Split (1)

train · 100k rows

id stringlengths 3 9	source stringclasses 1 value	version stringclasses 1 value	text stringlengths 1.54k 298k	added stringdate 1993-11-25 05:05:38 2024-09-20 15:30:25	created stringdate 1-01-01 00:00:00 2024-07-31 00:00:00	metadata dict
53122385	pes2o/s2orc	v3-fos-license	Indefinite theta functions and black hole partition functions We explore various aspects of supersymmetric black hole partition functions in four-dimensional toroidally compactified heterotic string theory. These functions suffer from divergences owing to the hyperbolic nature of the charge lattice in this theory, which prevents them from having well-defined modular transformation properties. In order to rectify this, we regularize these functions by converting the divergent series into indefinite theta functions, thereby obtaining fully regulated single-centered black hole partitions functions. Introduction and motivation In string theory compactifications, certain classes of microscopic states in the Hilbert space of bound systems of solitons and strings can be described by black hole solutions at strong t'Hooft coupling. In this context, exact counting functions have been developed that provide a statistical mechanical count of BPS states [1,2,3,4,5,6,7]. For a class of string theory compactifications such as type II on Calabi-Yau threefolds CY 3 , a topological twist creates a topological theory that captures the BPS aspects of the parent type II theory. In order to be able to write down the complete non-perturbative partition function of this theory, black holes must feed into the non-perturbative sectors of this theory and hence, writing down a well-defined black hole partition function becomes a significant step in this endeavour. In this paper we explore aspects of single-center black hole partition functions. We do this in four-dimensional N = 4 compactifications, since exact microstate counting formulae exist in these theories from which one can extract black hole degeneracies. In particular, we look at four-dimensional toroidally compactified heterotic string theory [1,3]. The first step in this program is to choose an ensemble to write down the single-center black hole partition function which can be used to extract the macroscopic black hole free energy. We consider the mixed statistical ensemble first introduced by OSV in [8]. It can be motivated by looking at partition functions in the near-horizon AdS 3 geometry of certain types of supersymmetric black holes. These black hole partition functions were explored by [9,10], where they were computed using formal Poisson resummation techniques. However, these partition functions are divergent due to the indefinite nature of the charge lattice in the theory. After an examination of the role played by the terms that contribute to the divergence in counting single-centered black holes, we propose a regularization of the divergent series by converting the sums into indefinite theta functions following a prescription by Zwegers [11], thereby obtaining fully regulated black hole partitions functions with well defined modular transformation properties. As a guiding principle we demand that the leading contribution to the free energy of this partition function equals the macroscopic black hole free energy. We now summarize some of the salient features of the dyonic degeneracy formula and set up relevant notation for the discussions that follow in this paper. Notation and background material Upon compactification of the type II string on K3 × T 2 physical charges are valued in the lattice Γ 6,22 ≃ H 2 (K3; Z) ⊕ 3 Γ 1,1 . Here Γ 1,1 is the hyperbolic lattice with bilinear form while the intersection form of the homology lattice of K3 decomposes into Γ 3,19 = ⊕2 Γ (−E 8 ) ⊕ 3 Γ 1,1 . The two vectors Q and P encoding the quantum numbers transform as a doublet under the S-duality group. The S-duality group is identified with the electric-magnetic duality in the heterotic frame and hence, these vectors can be labelled as electric and magnetic even if the individual charges are described in type II language. In this paper we will freely switch between the heterotic and type II dual frames, confident that no confusion should arise. The T-duality invariant charge bilinears of the theory are the norm squares of the electric and magnetic vectors, and their scalar product, explicitely −Q = Q · Q, −P = P · P and R = Q · P . The degeneracies of a class of micro-states 1 are expressed in terms of data associated with an auxiliary genus two Riemann surface. They are encoded in the expansion of the Siegel modular form 1 Φ 10 (σ, ρ, v) = Q,P ∈ 2Z, R ∈ Z d(Q, P, R) e −πi(Q σ+P ρ+R (2 v−1)) , (1.2) where the chemical potentials for the T-duality invariant bilinears parametrize the period matrix of said genus two Riemann surface The degeneracies d(Q, P, R) are non-vanishing for Q ≤ 2, P ≤ 2. For single-centered BPS black holes Q and P are negative while QP − R 2 ≫ 1, and hence convergence of the Q and P sums in (1.2) requires that Im ρ = M 1 ≫ 1 and Im σ = M 2 ≫ 1. Invariance of (1.2) under the large diffeomorphisms of the genus two Riemann surface given by ensures that we can always set the real parts of the chemical potentials to 0 ≤ Re σ < 1 , 0 ≤ Re ρ < 1 , 0 ≤ Re v < 1 . Therefore equation (1.2) can be rewritten by trading the sum over T-duality invariants with a sum over individual charges. The advantage of this rewriting is that it gives a direct path for comparing microscopic Hilbert space degeneracies to a macroscopic partition function over black hole backgrounds. The gravitational picture for these dyonic configurations includes extremal single-centered black holes which, at a specific point in their moduli space, have a near-horizon geometry described by a BTZ black hole in AdS 3 [13]. Approaching the horizon of the BTZ black hole yields an S 1 fibration over AdS 2 . The dual conformal field theory [14] has a central charge defined by the charge p 3 which sets the scale for the AdS 3 space; all states in this CFT are labeled as excitations above the vacuum by the quantum number q 0 and the angular momentum q 1 of the BTZ black hole. This provides a macroscopic partition function which counts single-centered black hole attractor geometries in a statistical ensemble where the p I are held fixed and the q I are summed over 2 . In a general charge configuration, the fixed p I define the AdS 3 spacetime, while the q I determine the BTZ excitations. Physically, this mixed ensemble captures all states in the near-horizon geometry of the black hole and should, in principle, capture the holographic entropy of the black hole, which is localized at the horizon. This mixed statistical ensemble was first introduced in [8] in the context of N = 2 Calabi-Yau compactifications of type II string theory. Hence, we are motivated to write down a black hole partition function in the mixed ensemble as Z OSV (p I , φ I ) = q I ∈Λe d(q I , p I ) e π q I φ I , (1.8) where Λ e denotes the lattice of electric charges in the large volume polarization, the variables φ I play the role of chemical potentials to be held fixed, and d(q I , p I ) denotes the absolute number (or a suitable index of) micro-states with electric/magnetic charges (q I , p I ). Observe that (1.8) is invariant under the shifts This formal invariance is a consequence of the fact that the charges are quantized and integer valued. In this paper we propose that the appropriate definition of the sum over the electric charges q a is in terms of indefinite theta functions. This will then also ensure the invariance under the shifts φ a → φ a + 2i. Definitions and properties of indefinite theta functions are briefly summarized in Appendix C. Models of interest In order to be able to use indefinite theta functions to define the sums over electric charges, these have to belong to sub-lattices defined in terms of quadratic forms of signature (r − 1, 1), respectively. This is the case in string models with N = 2 spacetime supersymmetry. However, models for which exact microstate degeneracies of dyonic black holes are known are models with N = 4 (or even N = 8) supersymmetry. In order to be able to apply the indefinite theta function regularization to N = 4 models we will focus on a subset of N = 4 charges, which we denote by (q I , p I ) (with I = 0, 1, . . . , n), and we will consider an effective N = 2 description of these models based on prepotentials of the form where n denotes the number of N = 2 vector multiplets coupled to N = 2 supergravity and the symmetric matrix C ab appearing in (1.10) has signature (1, n − 2), as required by the consistent coupling of vector multiplets to N = 2 supergravity [15,16]. In this N = 2 description, we take the associated charge bilinears and C ab to satisfy the same conditions as they do in N = 4. We can think of these models as appropriate sub-sectors of the N = 4 models where the charges associated with the extra indefinite directions have been set to zero such that the rank of the intersection form C ab and the gauge group is n − 1. In this paper we focus on toroidally compactified heterotic string theory 3 . Accordingly, we consider integer valued charges (q I , p I ) and integer valued matrices C ab and C ab , so that q a C ab q b ∈ 2Z, The T-duality subgroup of the N = 4 duality group that operates on the charges (q I , p I ) is SO(2, n − 1). The T-duality invariant charge bilinears are (1.11) The heterotic moduli fields are denoted by S = −iY 1 /Y 0 (the heterotic dilaton-axion field) and T a = −iY a /Y 0 . The effective N = 2 description also involves, in addition to F (0) , the S-duality invariant coupling function F (1) (S,S). As mentioned above we restrict ourselves to single-centered black hole states. These satisfy the following conditions, (1.12) The black hole attractor mechanism relates the near horizon values of the moduli fields S and T a to the charges as [17] 2 (1.14) This implies ̺ a C ab ̺ b > 0. Finally another ingredient we will need are indefinite theta functions. Aspects of the theory are reviewed in Appendix C. As in the case of ordinary theta functions, indefinite theta functions depend on a quadratic form Q : R r −→ R and its associated bilinear form B : R r × R r −→ R r . However in this case the quadratic form Q has signature (r − 1, 1). To retain convergence and modularity, one weights the sum with additional factors ρ as ϑ(z; τ ) = n∈Z r ρ(n + a; τ ) e 2πi τ Q(n)+2πi B(n,z) , (1.15) where τ ∈ H takes values in the complex upper half plane H, and z ∈ C r , with a, b ∈ R r defined by z = a τ + b. The factor ρ is the difference of two functions ρ c , ρ = ρ c 1 − ρ c 2 . The ρ c i depend on real vectors c i ∈ R r that satisfy Q(c i ) ≤ 0 and are given by [11] ρ c (n; τ ) = where E and sgn denote the error and sign function, respectively. In the main text we will take Q(c i ) < 0 for both c 1 and c 2 . The paper is organized as follows. We focus on the OSV ensemble and sum over the charges q 0 and q 1 without imposing any restrictions, following [10]. We compare the leading contribution to this sum with the macroscopic 1/4 BPS single-center black hole free energy. We then turn to the sum over the charges q a and specialize to the case p 0 = 0 in order to avoid a technical difficulty that arises when p 0 = 0. We regularize the sum over q a by converting it into an indefinite theta function. 2 Evaluation of Z OSV (p, φ) In the following we consider the evaluation of the mixed partition function (1.8) in toroidally compactified heterotic string theory, using an effective N = 2 description of this model based on (1.10). Summing over We first sum over the charges q 0 and q 1 following [9,10]. We convert the sum over (q 0 , q 1 ) into a sum over (Q, P ) using where, for the time being 4 , we assume that both p 0 and p 1 are non-vanishing, i.e. \|p 0 \| ≥ 1 , \|p 1 \| ≥ 1. In doing so, we need to ensure that when performing the sums over Q and P , we only keep those contributions that lead to integer-valued charges of q 0 and q 1 . These restrictions can be implemented by inserting the series L −1 L−1 l=0 exp[2πi l K/L], where K and L are integers (with L positive), which projects onto all integer values for K/L. The use of this formula leads to the following expression with R given by and [10] L(R,φ 0 ,φ 1 , φ a ) = Q,P d(Q, P, R) The range of the sums over l 0,1 enforces the condition that only those summands, for which (Q + q a C ab q b )/2p 1 and (P + p a C ab p b )/2p 0 are integers, give a non-vanishing contribution to (2.2). Now we introduce an additional sum over a dummy variable R ′ so as to be able to use the representation (1.2). To this end, we use a complex variable θ = θ 1 + iθ 2 , and write which holds for integer valued R, R ′ . Here, θ 2 is held fixed. Then we introduce (2.7) Next, using (1.2) and interchanging summations and integrations, we obtain [10] q 0 ,q 1 We note that identifying the d(q, p) on the lhs of (2.8) with the microcanonical dyonic degeneracy generated by the Siegel modular form 1 Φ 10 , automatically fixes the arguments (σ(θ), ρ(θ), θ) of the Siegel modular form to be the period matrix of a genus two Riemann surface, i.e. they have to take values in the Siegel upper half-plane Im ρ , Im σ > 0 , Im ρ Im σ > (Im θ) 2 . (2.9) Applying these restrictions to (2.7) imposes the constraints, Next, observe that (2.8) is invariant under the shifts [9,10] with n, m ∈ Z. Namely, under these shifts, σ(θ) and ρ(θ) transform as and since q a C ab q b ∈ 2 Z , p a C ab p b ∈ 2 Z, the exponent in the integrand of (2.8) is invariant under the shifts (2.11). Using Φ 10 (σ−n, ρ, v) = Φ 10 (σ, ρ, v) and Φ 10 (σ, ρ+m, v) = Φ 10 (σ, ρ, v), it follows that (2.8) is invariant under the shifts (2.11). This invariance, together with the sum over l 0 , l 1 -shifts, ensures that (2.8) is invariant under φ 0 → φ 0 + 2i , φ 1 → φ 1 + 2i. Φ 10 has various zeros [1]. The location of these zeros is parametrized in terms of five integers m 1 , n 1 , m 2 , n 2 ∈ Z, j ∈ 2Z + 1, which are subject to the condition The zeros are at 14) The zeros with n 2 encode the jumps in the degeneracies across walls of marginal stability corresponding to two centered small black holes which appear (or disappear) in the stable spectrum [18,19]. The zeros with n 2 ≥ 1 capture the entropy of single-center black holes [1]. The leading contribution to the entropy stems from the zeroes with n 2 = 1. Among them is the zero D with non-vanishing integers n 2 = j = 1, i.e. D = ρσ − v 2 + v = 0. In the following we focus on the zeros with n 2 = 1. These zeros can be generated from the zero D, which is described by (m 1 , n 1 , m 2 , n 2 , j) = (0, 0, 0, 1, 1), as follows. First, observe that Φ 10 (σ, ρ, v) is invariant under the discrete translations in v: Φ 10 (σ, ρ, v +p) = Φ 10 (σ, ρ, v) with p ∈ Z. Then, applying the shift transformation v → v + p as well as (2.12) to D results in zeros D (n,m,p) specified by the integers (n, m, −mn − p 2 + p, 1, 1 − 2p). In particular, D (0,0,0) = D. This provides a parametrization of the zeros with (m 1 , n 1 , m 2 , 1, j) satisfying (2.13). The same holds for the zeroes of Φ 10 (σ(θ), ρ(θ), θ), provided we make the compensating transformation φ 0 → φ 0 − 2ip p 0 , φ 1 → φ 1 + 2ip p 1 when performing the shift θ → θ + p. These compensating transformations constitute an invariance of (2.8), as discussed above. We may thus proceed as follows. The integral (2.8) will be evaluated in terms of the residues associated with the zeros of Φ 10 . Here we restrict ourselves to the zeros with n 2 = 1 which, as we just discussed, can be parametrized in terms of integers n, m, p. The contribution of these zeros can be accounted for by retaining the contribution of the zero D and extending the sum over l 0,1 to run over all positive and negative integers (i.e. l 0,1 ∈ Z) as well as extending the range of integration to −∞ < θ 1 < ∞. Hence, the poles of 1 Φ 10 corresponding to n 2 = 1 are characterized in terms of three numbers m, n and p, and by swapping the infinite ranges of m and n for the infinite ranges of l 0 and l 1 , respectively, and swapping p for the infinite range of θ 1 , we have fully expressed the entire subgroup of symmetries under which the poles corresponding to n 2 = 1 form a closed group, in terms of sums over two discrete variables and an integration over one continuous real variable. We now proceed with the evaluation of (2.8), focussing on the contribution of the zero D = 0. Free energy computation To compute the contribution from the zero D = 0, we follow the prescription given in [19], as follows. In the complex θ-plane, the contour of integration in (2.8) is taken to be −∞ < θ 1 < ∞ (as we just discussed) with fixed θ 2 , either θ 2 > 0 or θ 2 < 0. The θ-dependent part of the exponential in (2.8) can be written as with ̺ a given in (1.14). We consider single-center black holes so that ̺ a C ab ̺ b > 0. The choice of the sign of θ 2 then depends on the sign of p 0 p 1 . Namely, when p 0 p 1 < 0, we take θ 2 > 0. We can then deform the contour to θ 2 → −∞, where the integrand becomes vanishing. In doing so, we pick up the contribution from the zero D = 0, which will be specified below. Here, the zero is encircled in a clockwise direction. When p 0 p 1 > 0, we take θ 2 < 0. The contour can then be moved to θ 2 → ∞, where the integrand is again zero. In doing so, we pick up the contribution from the zero D = 0, but this time it is encircled in a counter clockwise direction. Thus, we obtain a non-vanishing contribution provided we choose the integration contour to satisfy p 0 p 1 θ 2 < 0. Then, the integral yields where Res denotes the residue which we now compute. Inserting (2.7) into the expression for D = v + ρσ − v 2 = 0, we find that the zero is located at the value (recall that now l 0,1 ∈ Z) which is complex, and hence away from the real θ 1 axis. In the vicinity of θ * , D takes the form For later convenience, we also introduce the notation Then, using (2.16), we obtain for (2.8) (we drop an overall numerical constant) we obtain Now, following [10], we generalize the definitions of S and Y 0 given in (A.8) and (A.9) and introduce the shifted fields Observe that in the presence of the l 0 , l 1 -shifts,S andȲ 0 are not any longer the complex conjugate of S and of Y 0 , respectively. Using (2.26), we obtain as well as (2.28) Inserting these expressions into (2.24), we get (dropping again a numerical constant) where in this expression the derivatives are with respect to S and toS, respectively. Next, let us relate (2.29) to the free energy of a macroscopic black hole. To this end, we first note that the mixed ensemble (1.8) involves summing (2.29) over q a . The macroscopic free energy, which corresponds to a critical point of the free energy functional, is obtained by extremizing the exponent in (2.29) with respect to q a . Performing this extremization we find . Then, inserting (2.31) into the exponent of (2.29) gives When l 0 = l 1 = 0, the value q B a can be thought of as a background charge that defines an attractor background geometry in view of the fact that (2.31) is simply the attractor equation for the real part of the scalar moduli fields Y a , cf. (A.11). Then, (2.32) equals the macroscopic free energy of this background charge black hole [10] (2.33) and the sum over the q a can be interpreted as a sum over fluctuations about this attractor background. In these expressions, Y 0 ,Ȳ 0 , S andS are defined with shifts φ 0 and φ 1 , as in (2.26). When l 0 = l 1 = 0, S and Y 0 become related to the attractor values for a singlecentered black hole. Indeed, using (2.32), we can rewrite (2.29) as where V a describes a fluctuation about the background charge (2.31). This follows by writing V a as where we used (2.31). One can see that the fluctuations can be space-like, time-like or null due to the hyperbolic structure of the charge-lattice metric C ab . In fact, if one thinks of the exponent as a free energy functional used to define the action for a partition function in a discrete hyperbolic lattice, then it is easy to see that there are no extrema of the action, but only critical points corresponding to single-centered black holes, since at any given point on this hyperbolic lattice there is always a space-like and a time-like direction. For the purpose of single-centered black hole entropy, we will only be interested in fluctuations in the l 0 = l 1 = 0 sector. The appearance of the other sectors in the mixed partition function function can be explained by analyzing the microcanonical degeneracy given in Here the contours are chosen such that the imaginary parts of the three arguments are fixed at certain values determined in terms of the charge invariants [19], and the real parts are chosen to run from 0 to 1. The integrand has second order poles corresponding to D (n,m,p) . In order to evaluate the residues at these poles, one can use the invariance of the integrand under imaginary translations in σ, ρ and v to map D (n,m,p) to D (0,0,0) while extending the range of the real parts of σ, ρ and v to the real line. In the case of Φ 10 (σ(θ), ρ(θ), θ), D (n,m,p) are mapped to D (n,m,0) . This involves an extension of the range of the real part of θ and a simultaneous translation in φ 0 and φ 1 in order to preserve the ranges of σ(θ) and ρ(θ), cf. (2.7). Here, the values of (l 0 , l 1 ) mod (p 1 , p 0 ) correspond to the increase in the ranges of the real parts of σ and ρ in Φ 10 (σ, ρ, v). The integral over v is done by expressing the leading divisor as a function of v and then evaluating the residue. The remaining two integrals are then performed by saddle point integration. The contour that passes through the saddle point is chosen in such a way that the two variables become conjugate to each other along the contour and that at the saddle point they correspond to the heterotic axion-dilaton pair and its conjugate [20]. Another way of expressing this is to say that the axion and dilaton scalars become real on this specific contour. The triple integral (2.37) also helps in defining background charges q 0 and q 1 , as follows. The imaginary parts of the integration variables, for the single-centered degeneracy, are expressed in terms of T-duality invariants as [19] Im σ = −2Λ P \|Q ∧ P \| , Using the definitions of the period matrix variables (2.7) in terms of φ 0 and φ 1 , and the definition of q 0 and q 1 in terms of Q and P (cf. (2.1)), respectively, we get an expression relating the background values of q 0 and q 1 to φ 0 , φ 1 and the other background charges, determined up to a positive constant Λ. Attractor geometry constraints on q a summation Summarizing, by summing over (q 0 , q 1 ), the number of integrations in (2.37) gets reduced from three to one, and the remaining integral over θ can be evaluated via residues. This is achieved by introducing an infinite sum over integers l 0,1 ∈ Z, which makes the shift symmetry φ 0,1 → φ 0,1 + 2i manifest. In the above we assumed that p 0 p 1 = 0. The result (2.29) remains valid when setting either p 0 = 0 or p 1 = 0. This can be checked (and we will do so in the next subsection) by redoing the above calculations using instead (Q, R) and (P, R) as summation variables, following [9]. Eq. (2.29) captures part of the OSV partition function for single-center black holes, namely the part associated with n 2 = 1. This yields the leading contribution to the partition function. Next, we would like to sum over charges q a . Here one faces the problem that one has to restrict to states with sgn(̺ a C ab ̺ b ) > 0,. Implementing this condition in a sum over charges q a is somewhat unwieldy. Note that this constraint becomes trivial in the rigid limit. Namely, when decoupling gravity, we recover a low-energy gauge theory based on a prepotential F (0) with a definite metric C ab , and the associated sum over the electric charges is unrestricted. To proceed, we note that a simplification occurs when setting p 0 = 0, since in this case ̺ a C ab ̺ b = −(p 1 ) 2 P and sgn(̺ a C ab ̺ b ) = sgn(−P ), which only depends on magnetic charges. Further, to make contact with a gravity partition function over single-centered black holes, one notes that black holes with p 0 = 0 have a near horizon geometry that, at an appropriate point of the moduli space, can be seen as a BTZ excitation of AdS 3 [13]. The OSV ensemble naturally sums over the q I charges and keeps the p I charges fixed. The fixed charges precisely define the AdS 3 background while the summed charges define excitations in this background. For these reasons, we will restrict ourselves to a summation over states with p 0 = 0 in the remainder of this paper. 2.4 Case p 0 = 0: summing over q a We will now compute the OSV mixed partition function (1.8) for the case when p 0 = 0. First, we redo the steps leading to (2.29) for the case p 0 = 0. Using we convert the sum over (q 0 , q 1 ) into a sum over (Q, R) and obtain [9] q 0 ,q 1 where now P is independent from Q and R. We set where we recall from (2.26), Once again, observe thatȲ 0 andS are not the complex conjugates of Y 0 and S when l 0,1 are non-vanishing. Next, we use the definition where ρ = ρ 1 + iρ 2 , and ρ 2 is fixed. We obtain [9] q 0 ,q 1 We consider single-center black hole solutions, so that P < 0. As before, identifying d(q, p) with the microcanonical dyonic degeneracy generated by 1 Φ 10 fixes the arguments (σ * , ρ, v * ) to satisfy the Siegel upper-half plane conditions (2.45) Next, we proceed as in the previous subsection. Using the characterization of the zeroes of Φ 10 corresponding to n 2 = 1 in terms of integers n, m, p we extend the sum over l 0,1 to run over all the integers, and we extend the range of integration to −∞ < ρ 1 < ∞. Using as well as Φ 10 ≈ D 2 ∆, we obtain the analogue of (2.30), where we used p 1 = Y 0 (S +S). Eventually, we obtain for the unregularized OSV partition function (up to an overall numerical constant), where where a = Im z/Im τ and b = Im (z τ )/Im τ . Here Q m (p) and Q e (q) are indefinite quadratic forms, and B e (z, q) is the bilinear form associated to Q e (q). Using τ e = σ * we obtain that τ e takes values in the complex upper half plane by virtue of the Siegel upper half plane conditions (2.45). Note that (2.47) agrees with (2.29) when setting p 0 = 0. For generic values of l 0 and l 1 both a a and b a are non-vanishing in the decomposition (2.48). On the other hand, when l 0 = l 1 = 0, (S +S) and i(S −S) are both real, and hence b a = 0. We will return to this issue in the next subsection when regularizing the sum (2.47). The matrix A ab has signature (n − 2, 1), and hence the quadratic form Q e (q) is indefinite, rendering the sum (2.47) over q a divergent, as discussed previously. We propose to regulate the divergence by turning the sum over the q a in (2.47) into an indefinite theta function ϑ(z; τ e ) following [11]. Regularizing Z OSV (p, φ) Recall that in our particular model, A ab is integer valued and indefinite, and τ e takes values on the upper half complex plane by (2.45). This is precisely the setting where indefinite theta functions can be defined. We will now modify the definition of the OSV sum in order to obtain an indefinite theta function, and discuss the consequences of this procedure. A physically motivated discussion of the regulatory procedure is given in Appendix B via a toy model. The main properties of indefinite theta functions are summarized in the Appendix C. An indefinite theta function (1.15) differs from an ordinary theta function by the presence of an extra factor ρ which deals with the indefinite directions, preserving modular properties. This factor ρ explicitly depends on two vectors c 1 and c 2 . Depending on the specific form of ρ these two vectors are used to project out the lattice points giving an exponentially growing contribution, or to weight them with a positive definite quadratic form. Thus, by introducing the weight ρ (e) in the OSV partition function (2.47), we obtain a convergent and modular sum Note that this should be intended as part of the definition of the electric sum, as we are not going to remove the weight ρ (e) in the following. Whether this factor can be derived from first principle, and not just by macroscopic arguments, is clearly an interesting question. Having obtained a modular object 5 , we now consider the modular transformation (τ e , z) → (−1/τ e , z/τ e ). Using that A ab is integer valued, ϑ(z; τ e ) transforms as [11] ϑ(z/τ e ; −1/τ e ) = 1 √ − det A (−iτ e ) (n−1)/2 e 2πi Qe(z)/τe ϑ(z; τ e ) = ν∈Z n−1 ρ (e) (ν +ã; −1/τ e ) e −2πi Qe(ν)/τe+2πi Be(z/τe,ν) , (2.50) Hence we obtain which makes it manifest that (2.49) has the shift symmetry φ a → φ a + 2i. Using (2.52), Z c 1 ,c 2 OSV (p, φ) gets expressed as where F E (p,φ) denotes the free energy (2.32) with φ a replaced byφ a . Apart from modular transformations, the indefinite theta function may also be subjected to elliptic transformations. One such transformation is induced by the S-duality transformation S → S + i λ with λ ∈ Z. This transformation induces the shift z a → z a + λ a τ e , where λ a = −C ab p b λ, as can be seen from (2.48). Under this transformation, the indefinite theta function picks up a factor [11] 2πiB(z, λ a )]. This particular elliptic transformation can also be viewed as inducing a shift of the background charge q B a given in (2.31). Namely, using (2.48), the above transformation can also be obtained by performing the shift φ a → φ a − 2λ p a /(S +S) which, using (2.31), translates into shifting the background charge by q B a → q B a + λC ab p b . This shows how the background charge dependence is encoded in the elliptic transformation. Introducing T a as T +T a = (S +S) p 1 p a , (2.55) and using p 1 = Y 0 (S +S) we get as well as This quantity equals the Kähler potential and Ω are given in (1.10) and in (2.27)), with T a = −iY a /Y 0 replaced by (2.55), and with Y 0 and S replaced by the shifted quantities (2.42). Observe that K is invariant under both S-and T-duality [10]. This extends toK and to the combination K + 4(S +S) 2 ∂ S ∂SΩ, provided S-and T-duality are defined in the same way when acting on the shifted fields (2.42) and (2.55). Note also that if we artificially take n = 27 (which corresponds to taking a model with 28 abelian gauge fields, just as in the original N = 4 model), the term (S +S) 12 in (2.57) precisely cancels against a similar term coming from Ω(S,S) in (2.27) [21], so that (2.57) becomes Thus, (2.60) takes a form reminiscent of \|e Ftop \| 2 (where F top denotes the holomorphic topological free energy), with an additional duality invariant measure factor [10] as well an extra weight factor ρ (e) . Our result for the regularized OSV partition function (2.57) contains a sum over indefinite theta functions over different (l 0 , l 1 ) sectors. In each (l 0 , l 1 )-sector, we can choose wedge vectors c 1 and c 2 to define the regulating error functions. Let us consider the sector l 0 = l 1 = 0 in more detail, and let us discuss a subtlety to which we already alluded to above. The sector l 0 = l 1 = 0 describes the semi-classical sector, and hence we must demand that our choice of wedge vectors and regularization yields sensible results in the semi-classical regime. The exact semi-classical point corresponds to ν = 0 in the l 0 = l 1 = 0 sector. However, as already mentioned, we have b a = 0 in this sector, and hence, both error functions in the regulator vanish. To resolve this conundrum, we propose a shift in the fluctuation (2.36) Here, we have used the fact that at the semi-classical point, Re(S −S) = Im(S +S) = 0. The modified free energy has an extra term −λ 2 Q(U ) S+S − iλU a C abφ b , as can be seen from (2.32). We will now pick appropriate value of c 1 and c 2 to preserve the classical free energy. In a well defined classical limit one should ensure that the exponential corrections coming from the error function do not affect the free energy. From (2.57) both the exponential and ρ go as S +S. If we denote y = −Im 1 τe and x 1 and x 2 the remaining factors in the error function, we can write Here, since we are interested in the semi-classical limit, we set l 0 = l 1 = 0 and obtain which are invariant under rescaling of the c i . In the semi-classical limit we have (S +S) → ∞, and hence we consider the expansion of the error function (C.3) around x = 0 as [22], In our case this yields (setting ν a = 0) (2.67) We choose U a to be a spacelike vector with norm U = U a C ab U b , so that in some Tduality frame we can bring it to the form U a = (1, 1 2 U 2 , 0), where vector 0 spans the timelike SO(r − 2) directions and the non-zero slots fill out a hyperbolic lattice. Then, in this frame, we choose c 1 a = ( 1 2 U 2 , 1, 0) and c 2 a = (1, 1 2 U 2 , 0), so as to ensure that the exponent in the regulator outside the brackets in (2.67) fully cancels the extra real term in the free energy πF E . In addition, for large U , the second term in the brackets is subleading compared to the first (when λ → 0). Then, normalizing the regulator in (2.67) by λ 1+α e −πiλU a C abφ b , where 0 < α < 1, and demanding that lim λ→0,S+S→∞ λ 2α (S +S) stays finite, we remove all leading order contributions to the semi-classical free energy. Let us now briefly comment on the symmetries of the regularized OSV partition function. The indefinite theta function has modular and elliptic transformation properties. The modular transformation (τ e , z) → (−1/τ e , z/τ e ) implements Poisson resummation, which we employed to extract the semi-classical free energy, see (2.57). The elliptic transformation z a → z a + λ a τ e , with λ a = −C ab p b λ and λ ∈ Z, induces a shift of the background charge, q B a → q B a + λC ab p b . This is a reflection of the underlying S-duality invariance that is present in the original N = 4 theory, see (A.4). The regularized OSV partition function also has the shift symmetry φ I → φ I + 2i, which expresses integrality of the charges. This shift symmetry was made manifest using the following steps. In going from the OSV partition function (1.8) to the dyonic degeneracy formula (1.2), the summation variables changed from charges to T-duality invariant charge bilinears. The chemical potentials appearing in the dyonic degeneracy formula (1.2) have translational symmetries associated with the integrality of the invariant charge bilinears. In order to ensure the stronger condition for the integrality of charges, we had to make the translation symmetry of the original OSV potentials explicit by introducing new dummy summation variables and complexifying the potentials. Hence, the shift symmetry φ I → φ I + 2i of (1.8) is manifest in the result (2.57), which is entirely expressed in terms of hatted potentialφ I that are complex. This was achieved by regularizing the sum over q a by turning it into an indefinite theta function ϑ(z; τ ), and subsequently applying a modular transformation to it. To achieve this, we had to introduce a vector U a with norm Q(U a ) < 0. We may identify U a with the real part of an asymptotic T a -modulus lying in the Kähler cone. This shows that, in general, only an SO(1, r) subgroup of the T-duality symmetry group is preserved by the regulator/the specific choice of the vectors c 1 and c 2 that enter in the definition of ρ, in the Zweger's prescription. 2.6 Z c 1 ,c 2 OSV (p, φ) from an expansion of 1 Φ 10 in powers of P In the OSV ensemble, the p I are kept fixed, which implies that when p 0 = 0, the charge bilinear P is constant. This allows us to obtain an exact expression for Z c 1 ,c 2 OSV (p, φ) by using a Fourier expansion of 1/Φ 10 in P -modes, as follows. Proceeding as above, we regularize the sum over the q a by turning it into an indefinite theta function, with z a given as in (2.48). Performing the modular transformation (2.52) we get Conclusions In this paper we have streamlined a new approach to deal with black hole partition functions in quantum gravity. The main feature is that the indefinite character of the charge lattice, a distinctive feature of gravity, and the need to preserve as many symmetries as possible, point towards the necessity of new mathematical structures to deal with the sums over microscopic states. We propose that the theory of indefinite theta functions may play a distinctive role in this program, as elucidated below. Partition functions and divergences In order to contextualize the divergences in the mixed ensemble, it is instructive to analyze the counting formula in two other ensembles. The first ensemble is defined by chemical potentials corresponding to the variation of the T-duality charge bilinears, where the partition function is given by In the Siegel upper-half plane, where this series is well-defined, Im σ ≫ 1 , Im ρ >≫ 1 and Im ρ Im σ ≫ (Im v) 2 , and hence the Q and P expansions are convergent. However, for a given value of Im v, the series is divergent in the R-sum, as the R sum goes over both positive and negative values. Hence, one cannot define the series for both positive and negative values of R, for a fixed value of Im v. This is related to the meromorphic structure of Φ 10 , arising from its double zero structure, which we dealt with, in detail, in section 2. In particular, the partition function has a double pole at y (1−y) 2 , where y = e −2πiv . One can expand this series about y = 0 or y = ∞, resulting in an expansion in positive powers of y , corresponding to positive R or an expansion in negative powers of y, corresponding to negative powers of R, respectively. But one cannot analytically continue from one expansion to another due to the pole at y = 1 corresponding to Im v = 0. As one moves through the pole, one picks up the residue around the pole, and this results in a jump in the degeneracy across a line of marginal stability corresponding to the appearance or disappearance of decadent dyons. One way to extract the microcanonical degeneracy and regulate the series is to compute the degeneracy for one sign of R, corresponding to the fixed sign of Im v, which makes the series well-defined. Then, we define the degeneracy of the charge configuration with the same value of Q and P , but with the opposite sign of R, as being equal to the degeneracy of the computed charge configuration, by applying parity-invariance [12]. A second ensemble in which we can write down a dyonic counting formula is obtained by fixing P and varying the other two charge bilinears. We can write down a partition function in this ensemble by going to a point in the Siegel upper-half plane where Im ρ ≫ Im σ. This allows us to expand the Siegel form as [23] 1 where ψ 10,m is a Jacobi form of weight 10 and index m. The Jacobi form inherits its meromorphicity from the Siegel modular form, and this leads to a divergence at the double poles. One can regulate this divergence by splitting the Jacobi form into two mock modular forms, one of which, the polar part, ψ P , encodes the double poles of the Jacobi form, and hence the jumps across the lines of marginal stability due to the decadent dyons, and the other is the finite analytic part of the Jacobi form which counts the immortal single-centered dyons, ψ F , as ψ 10,m = ψ P + ψ F . On the other hand, in the mixed statistical ensemble, the partition function is written as Z OSV (p I , φ I ) = q I d(q, p) e πq I φ I . For any sign of the chemical potentials, as we sum over both positive and negative values of q I , this series is divergent. As we saw in section 2, this divergence shows up in the evaluation of the fluctuations about the critical point contribution to the free energy of the partition function. In this paper we have physically motivated a reason to use a soft regulatory mechanism to handle the divergences in the partition function written in the OSV ensemble (see Appendix B). We used a soft regulator following Zwegers to convert this into an indefinite theta function. The regulator converts the divergent series into an indefinite theta function. However, this regulator could be one of many choices to define a convergent series. In order to provide a physical basis for it, we notice that the resulting well-defined partition function counts single-centered black holes. Hence, a physical justification for our regulator lies in establishing a connection between the regulated partition function in the mixed ensemble, and the finite mock modular form. Accordingly, one must turn to the connection between the theory of indefinite theta functions and mock modular forms analyzed in [11], and show that the mock modular form associated with this indefinite theta function is precisely the one that encodes the single-centered degeneracies. This connection can be worked backwards in theories like the STU or FHSV models, where there is a strong suggestion [24,25] that the wall crossing phenomena are encoded in terms of indefinite theta functions, to be able to extract mock modular forms and hence, partition functions for counting single-center black holes. These open questions are being currently pursued. Summary and context The simple idea described and implemented in this paper provides a starting point to define OSV-like sums over single-centered black hole microstates which have a semiclassical limit consistent with supergravity. Indeed, the need of having sums over states which are both mathematically meaningful (and not just formal) and compatible with an infrared gravitational description has been the guiding principle of our approach. One can extend the procedure implemented in this paper to the n 2 > 1 poles of the Siegel modular form that counts the microscopic states in the theory. Regularizing the OSV partition function using indefinite theta functions renders it convergent while maintaining the shift symmetry (1.9), and allows to make contact with semi-classical results through the usage of modular transformations that implement Poisson resummation. In order to prove the absolute convergence of the indefinite theta function series, one can show [11] that the series of the absolute values of the terms in the theta function converges faster than a canonical series obtained by effectively modifying the metric Q(c) of the charge lattice so as to remove the indefinite direction, and hence regulate the series. The norms of the vectors then change to Q(ν) → Q(ν) − B(c,ν) 2 2Q(c) (cf. (C.7)). Hence one could alternatively choose to simply modify the metric as above. It turns out that this is equivalent to introducing a canonical partition function regulator e −βH , where H is the Hamiltonian of the dyonic system seen as a bound state of D-branes in the theory. Indeed, if we consider the exponent Q c (ν) Imτ e in (C.7) and focus on the second term, given by B(c, ν) 2 /[−2(S +S)Q(c)] (here we set l 0 = l 1 = 0), and perform the replacements β = 1/(S +S) and u a = C ab c b / −2Q(c), we obtain β (ν a u a ) 2 with C ab u a u b = 1. This is precisely the H-regulator proposed in (6.15) of [26]. Thus, amusingly, the H-regulator used in [26,27] can be formally identified with the second term of Q c . We note, however, that the H-regulator has, so far, only been used at strong topological string coupling, c.f. (2.51) and (2.54) in [27]. We are not aware of its extension to weak topological string coupling, which corresponds to the semi-classical limit. One can also view the present work in the context of the picture of the quantum entropy function introduced by Sen in [28]. The quantum entropy function, which counts the microstates of a supersymmetric black hole, was defined in the AdS 2 background which formed the near horizon geometry of the black hole. In this background, the fluctuation over the various fields had to be performed keeping the charge fixed since, in two dimensions, the charge is associated with the non-normalizable part of the electric field. Hence, the quantum gravity partition function computed in this background is bound to be the microcanonical partition function of black hole microstates, and it can be expressed as the exponential of the Legendre transform of the full quantum action evaluated on this background with respect to the charges of the black hole, to give the full quantum entropy function. On the other hand, to compute a canonical partition function, we looked at black holes which, at some point in their moduli space, have a near horizon background factor of AdS 3 . The central charge of the holographically dual CFT and the radius of AdS 3 are fixed by the p I , while the chiral excitation that defines the black hole is determined by the q I 6 . Hence, the partition function is defined by summing over the q I while keeping p I fixed. This mixed ensemble counts fluctuations in AdS 3 , and the free energy computed in this ensemble will include not just the single-center black hole excitations, but also other excitations of the AdS 3 vacuum. The associated partition function can be asymptotically expressed as the exponent of a free energy which is the Legendre transform of the action with respect to the q I . Finally, the work presented here paves the way for rigorously defining black hole partition functions in a grand canonical ensemble. The grand canonical ensemble sums over every charge in the system [10] and can be thought of as summing over all fluctuations. The free energy computed in this case will have contributions from various AdS 3 backgrounds, each of which defines a mixed ensemble. This quantity will be the full Euclidean quantum action that encode the dynamics of black hole backgrounds in the theory. A well defined properly regulated partition function in the grand canonical ensemble is therefore quintessentially important in an understanding of the underlying stringy effective action of the theory. Finally, we note that our results indicate the need for defining indefinite theta functions on more general lattices. A S-duality, attractor equations, and the Hesse potential In this Appendix we will review some of the duality properties of the effective N = 2 description and their relation with the attractor equations. In particular we will also discuss the relation between the effective free energy and the Hesse potential. Consider the prepotential (1.10). Associated to each Y I is a pair (q I , p I ) of electric/magnetic charges. This is the charge vector in the so-called type IIA polarization. It is related to the one in the heterotic polarization byq = (q 0 , −p 1 , q a ) , p = (p 0 , q 1 , p a ) . (A.1) Under S-duality, where a, b, c, d are real parameters that satisfy ad − bc = 1. It acts as on S = −iY 1 /Y 0 , and as follows on the charges, In particular, under the transformation a = d = 0, b = −c = 1, we have S → 1/S and q a → −C ab p b , p a → C ab q b . We define φ I = Y I +Ȳ I , where F as well as the electric attractor equations for the q a , leads to a full determination of the Y I in terms of S, as follows [10], Using (A.9), the attractor values for (φ a , χ a ) are and Using (1.10) and (A.10) we compute Next, let us consider the macroscopic free energy based on (1.10). It is given by (2.32) with Ω = 0 (and l 0 = l 1 = 0), The modular parameters τ e and τ m that appear in (2.48) can be defined starting from the macroscopic free energy (A.14), as follows. We use the attractor equation (A.11) to express φ a in terms of S,S and the charges (p a , q a ). Then, viewing F E as a function of p a , q a (and S,S) we get If we now set 16) we obtain τ e = i/(S +S) and τ m = i\|S\| 2 /(S +S). Finally, let us consider the Hesse potential H (0) (φ, χ, S,S) that is obtained by Legendre transformation of the free energy (A.14) with respect to p a , (A.17) Using χ a = ∂F (0) E /∂p a we obtain B Toy model for a regulator We outline a simple model of regulating a divergent series so as to be able to provide a physical motive for the prescription of Zwegers. Consider the series V e V 2 , where V a is a vector in a two-dimensional Lorentzian space, and V 2 = V a C ab V b . The spacelike vectors render this series divergent. One simple way to regulate this series in a Lorentz-invariant fashion is simply to sum over only the time-like vectors. To implement this we go to the light cone gauge, where the norm-squared of the vector V a is given by ). This is precisely the regulatory proposal of Zwegers to define indefinite theta functions, which we review below. Note that in the case of a regulator based on error functions, all the vectors V a contribute to the sum, but those that would cause a divergence now appear with an exponential damping factor, rendering the sum convergent. Thus, no vectors are discarded in this case. C Properties of indefinite theta functions In this Appendix we review various properties of indefinite theta functions. Indefinite theta functions ϑ(z; τ ) were defined and studied by Zwegers in [11], and are modified versions of the sums considered in [29]. They have found recent string theory applications in [25,30,31,32,23]. Indefinite theta functions are based on quadratic forms Q : R r → R of signature (r−1, 1), defined in terms of symmetric non-degenerate r × r matrices A with integer coefficients, where τ ∈ H takes values in the complex upper half plane H, and z ∈ C r , with a, b ∈ R r defined by z = a τ +b. The factor ρ is the difference of two functions ρ c , ρ = ρ c 1 −ρ c 2 . The ρ c i depend on real vectors c i ∈ R r that satisfy Q(c i ) ≤ 0. In the following, we take Q(c i ) < 0 for both c 1 and c 2 . The other possibility, that is when Q(c i ) = 0, will be discussed in Appendix D. The set of vectors with Q(c) < 0 has two components, and we take c 1 and c 2 to be in the same component, so that B(c 1 , c 2 ) < 0. Then, the ρ c i are given in terms of error functions, Observe that ρ c (n + a; τ ) is non-holomorphic in τ . Also note that the definition of ρ(n + a; τ ) doesn't change if we replace c i by λ c i , with λ ∈ R + . This implies that two c i belonging to the same component of Q(c) < 0 should not be collinear, since otherwise ρ = 0, and that we may replace the condition Q(c i ) < 0 by Q(c i ) = −1 [11]. As shown in [11], (C.1) is convergent and has nice modular and elliptic transformation properties that are similar to those of theta functions based on positive definite quadratic forms. In the following, we briefly highlight various aspects that go into proving these remarkable facts. Following [11], we consider the indefinite theta function ϑ a,b (τ ) defined by ϑ a,b (τ ) = e 2πiQ(a)τ +2πiB(a,b) ϑ(z; τ ) = ν∈a+Z r ρ(ν; τ ) e 2πiQ(ν)τ +2πiB(ν,b) . (C.5) We begin by sketching the proof of convergence of (C.5). Since the proof is rather lengthy, we focus on the regulatorρ(x) = −sgn(x) β(x 2 ), which is related to ρ according to (C.3). To prove the convergence of the series usingρ, we will need the following two lemmata from [11]. The first lemma states that To show this, we consider the function f (x) = β(x 2 ) − e −πx 2 . It vanishes at x = 0, ±∞. Away from these values, it has extrema located at πx = sgn(x), which are local minima. Hence it follows that f (x) ≤ 0, which establishes the lemma. The second lemma that is needed states that the combination is positive definite. Here, c is a vector satisfying Q(c) < 0. Note that Q c (ν) > 0 regardless of the sign of Q(ν). This lemma can be proven as follows. First let us consider the case when ν ∈ R r is linearly independent of c. Then the quadratic form Q has signature (1, 1) on the two-span{c, ν}, and hence the matrix has determinant < 0, so noting that Q(c) < 0 we obtain On the other hand, when ν = λ c with λ = 0, we obtain Q c (ν) = −Q(c) λ 2 > 0, which shows that the combination (C.7) is always positive definite. Next, using Q(c) < 0, we compute ρ(ν; τ ) e 2πiQ(ν)τ +2πiB(ν,b) ≤ e π B 2 (c,ν) Q(c) Im τ e 2πiQ(ν)τ +2πiB(ν,b) = e −2πQc(ν) Im τ . (C.10) Since Q c (ν) > 0, the series ν∈a+Z r e −2πQc(ν) Im τ (C.11) converges, and thus ϑ a,b (τ ) is absolutely convergent for the above choice of regulator, and hence convergent. It is also uniformly convergent for Imτ ≥ ε > 0. The proof of convergence for the regulator ρ is much more involved, but proceeds along similar lines [11]. Now let us consider the behavior of ϑ a,b (τ ) under the modular transformation τ → −1/τ . The proof given in [11] establishing that ϑ a,b (τ ) has a good behavior under this transformation requires the regulator ρ(ν; τ ) to be an odd function of ν, and the derivatives ∂ρ/∂ν to exist. In [11] it is shown that under τ → −1/τ , ϑ a,b (τ ) transforms as We now sketch the proof leading to this result. It uses a lemma as well as Poisson resummation. The lemma states that for all α ∈ R r and τ ∈ H, R r ρ(a; τ ) e 2πiQ(a)τ +2πiB(a,α) da = 1 Its proof goes as follows. The integral on the left hand side is convergent. Using ∂ ∂α l e 2πiQ(aτ +α)/τ = 1 τ ∂ ∂a l e 2πiQ(aτ +α)/τ , (C.14) we obtain ∂ ∂α l e 2πiQ(α)/τ R r ρ(a; τ ) e 2πiQ(a)τ +2πiB(a,α) da = − 1 τ R r ∂ρ ∂a l (a; τ ) e 2πiQ(aτ +α)/τ da , (C.15) where we integrated by parts and used that the boundary terms do not contribute. Since ρ is the difference of two error functions, the derivatives ∂ρ ∂a l (a; τ ) yield derivatives of error functions. We therefore consider the following expression which appears on the right hand side of the above equation, where we used the property E ′ (x) = 2e −πx 2 . As shown in [11], the integral over R r can be split into an integral over R (corresponding to the direction associated with the negative eigenvalue of Q) and an integral over R r−1 associated with the directions that correspond to the positive eigenvalues of Q. The integration over R would yield a divergent result where it not for the presence of the additional factor E ′ which converts the factor τ appearing in the exponent on the right hand side into a factorτ , rendering the integration over R well behaved, R e 2πiQ(c) a 2 cτ +4πiQ(c)acαc da c , (C.17) where we refer to [11] for a detailed derivation of this remarkable result. Using this, the right hand side of (C.16) evaluates to Im τ e 2πiQ(aτ +α)/τ da Using this expression in (C.15) leads to (C. 19) It follows that the expressions in the two brackets have to agree, up to an α-independent expression. Since ρ(x) is an odd function, both brackets are odd as a function of α, and hence the α-independent expression has to vanish. This proves lemma (C.13). D An example of an indefinite theta function In this appendix we consider an explicit example [29] of an indefinite theta function. In doing so we explicitly show how indefinite theta functions differ from ordinary theta functions, and how the indefinite directions are dealt with. In this example the weight function ρ is taken to be the difference of two sign functions. In this case Zweger's "wedges" act as a projector onto a specific sublattice. We consider the indefinite theta function given by [29] x∈Γ 1,1 ρ(x + α) e 2πiQ(x)τ +2πiB(x,γ) , (D.1) defined over the indefinite lattice Γ 1,1 . We write the vector γ as γ = ατ + β. The other vectors are explicitly x = (m, n) , γ = (γ 1 , γ 2 ) , α = (α 1 , α 2 ) , β = (β 1 , β 2 ) . (D.2) For simplicity we will focus on the case where both c 1 and c 2 are chosen in such a way that Q(c i ) = 0 for i = 1, 2. In this case the weight function ρ = ρ c 1 − ρ c 2 simply plays the role of a projector and ρ c i (x + α) = sgn B (x + α; c i ) . We have now two possibilities. In the first case we pick the domain S 1 . Then c 1 = (0, +1) and c 2 = (−1, 0). Therefore we have to work with the condition {0 < α 1 < 1} ∩ {−1 < α 2 < 0} . (D. 6) In particular ρ(x + a; τ ) = sgn B(x + α, c 1 ) − sgn B(x + α, c 2 ) = sgn(m + α 1 ) + sgn(n + α 2 ) = The border values have to be treated separately: m = 0 =⇒ sgn(α 1 ) + sgn(n + α 2 ) = +1 + sgn(n + α 2 ) = 0 ⇐⇒ n > 0 n = 0 =⇒ sgn(m + α 1 ) + sgn(α 2 ) = sgn(m + α 1 ) − 1 = 0 ⇐⇒ m < 0 (D.8) Therefore the sum of the theta function is only in m ≥ 0, n > 0 and m < 0, n ≤ 0. Therefore x=(m,n)∈Γ 1,1 ρ(x + α) q mn e m(2πiγ 2 )+n(2πiγ 1 ) Note that in both cases the weight function ρ acts as a regulator projecting out certain lattice points, among which those that would have given an exponentially growing contribution. The situation where ρ is the difference between two error functions (C.2) can be treated similarly. Indeed as pointed out in (C.3) the error function can be written as E(x) = sgn(x) 1 − β(x 2 ) = sgn(x) − sgn(x)β(x 2 ) . (D.14) The first term in the sum behaves precisely as we have explained in the above text. The second term has a different role. Note that the second term is precisely the combination which we calledρ in the appendix C, cf. below (C.5). As discussed there, it follows from (C.10) that the combination ofρ with the exponential of an indefinite quadratic form is always bounded by a damped positive definite quadratic form. This was used to bound the whole series to prove its convergence. This means that in the function ρ = ρ c 1 −ρ c 2 , the competition between the sign terms in (D.14) acts as a projector, while the terms proportional to β sum over all the lattice points, but suppress their contribution exponentially via the error function.	2014-07-02T18:07:37.000Z	2013-09-17T00:00:00.000	{ "year": 2013, "sha1": "ad9c290b3d6fd40b0e033e6a8df869ca3b17a4cd", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/JHEP02(2014)019.pdf", "oa_status": "GOLD", "pdf_src": "Arxiv", "pdf_hash": "ad9c290b3d6fd40b0e033e6a8df869ca3b17a4cd", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
229306106	pes2o/s2orc	v3-fos-license	Liver–lung interactions in acute respiratory distress syndrome Patients with liver diseases are at high risk for the development of acute respiratory distress syndrome (ARDS). The liver is an important organ that regulates a complex network of mediators and modulates organ interactions during inflammatory disorders. Liver function is increasingly recognized as a critical determinant of the pathogenesis and resolution of ARDS, significantly influencing the prognosis of these patients. The liver plays a central role in the synthesis of proteins, metabolism of toxins and drugs, and in the modulation of immunity and host defense. However, the tools for assessing liver function are limited in the clinical setting, and patients with liver diseases are frequently excluded from clinical studies of ARDS. Therefore, the mechanisms by which the liver participates in the pathogenesis of acute lung injury are not totally understood. Several functions of the liver, including endotoxin and bacterial clearance, release and clearance of pro-inflammatory cytokines and eicosanoids, and synthesis of acute-phase proteins can modulate lung injury in the setting of sepsis and other severe inflammatory diseases. In this review, we summarized clinical and experimental support for the notion that the liver critically regulates systemic and pulmonary responses following inflammatory insults. Although promoting inflammation can be detrimental in the context of acute lung injury, the liver response to an inflammatory insult is also pro-defense and pro-survival. A better understanding of the liver–lung axis will provide valuable insights into new diagnostic targets and therapeutic strategies for clinical intervention in patients with or at risk for ARDS. Background Acute respiratory distress syndrome (ARDS) is a severe respiratory failure, due to noncardiogenic pulmonary edema [1,2], associated with a hospital mortality between 35% and 46% [1,3,4]. The pathology of ARDS involves diffuse alveolar damage (DAD), which comprises severe alveolar epithelial cell damage, neutrophil infiltration, activation of alveolar macrophages, production of cytokines and chemokines, plasma extravasation, procoagulant activity with fibrin deposition, hyaline membrane formation, myofibroblast proliferation, and fibrosis in the intra-alveolar spaces [2,4]. Formation of protein-rich edema in the airspaces due to the disruption of the alveolar-capillary membrane is one of the main factors that contributes to the severe impairment of blood and tissue oxygenation early in the evolution of DAD [2,4]. The DAD occurs not only in response to a direct injury to the lung (e.g., pneumonia), but it may also represent a pulmonary manifestation of diverse systemic immunoregulatory disorders, such as sepsis [4]. The pathogenesis of ARDS, therefore, is linked to changes in local and systemic host defense and immune responses [5], in which the liver plays an important role (Fig. 1). The liver has unique anatomic, cellular, and physiological characteristics that enable the clearance of circulating microbial products, tissue debris, altered platelets, products of intravascular coagulation, and different bioactive molecules ( Fig. 1) [6][7][8][9][10]. Also, the liver has a key role in the synthesis of proteins, metabolism of toxins and drugs, and in the modulation of systemic inflammatory responses and host defense (Fig. 1). It is becoming more evident that normal liver function exerts lung protection and is necessary for recovery from lung damage [11,12]. In this line, it has been observed that established ARDS during acute liver allograft rejection is resolved within hours of Fig. 1 Role of the gut-liver-lung axis in acute respiratory distress syndrome. There are several physiological mechanisms promoted by the liver that contribute to the development, progression, and resolution of the acute respiratory distress syndrome. ARDS acute respiratory distress syndrome, DAD diffuse alveolar damage, APPs acute-phase proteins, AA arachidonic acid, GF growth factor hepatic re-transplantation [13]. On the other hand, experimental studies suggest that the presence of the liver is also absolutely necessary for inducing lung injury in rats [14]. These apparently paradoxical observations highlight the relevant crosstalk between lung and liver in ARDS. Despite the well-recognized liver-lung interaction in the pathogenesis of ARDS, its underlying mechanisms and its effects on the outcome of these patients have been barely studied due to several reasons. First, patients with liver diseases are frequently excluded from studies of ARDS. In addition, liver function is not precisely reflected by the standard liver function tests in the clinical setting, and the liver is not as accessible as other organs such as the lung, making liver dysfunction not as evident as dysfunctions of other organs. Finally, its clinical consequences are also heterogeneous in critically ill patients [15]. The present work reviews the important role of the liver on the development and resolution of ARDS and aims to provide an integrated view of the underlying mechanisms that support the liver-lung interaction in critically ill patients. The reciprocal impacts of lung and liver dysfunctions Following hepatocellular damage, the liver may reduce its clearance function, increase the synthesis of deleterious substances, and dysregulate immune responses, leading to systemic complications such as coagulopathy, elevated risk of infection, hypoglycaemia, exacerbated inflammatory responses, encephalopathy, and damage of other extrahepatic organs, including lung injury [16][17][18][19][20][21][22]. In critically ill patients, hepatic dysfunction is recognized as a relevant clinical condition that influences the development, severity, and progression of ARDS [5,11,19,[23][24][25][26][27]. In ARDS patients, liver dysfunction is a major determinant of mortality [24][25][26]. It is well known that cirrhosis and other chronic liver diseases make the patients more susceptible for developing ARDS, which adversely affects patient outcomes [24][25][26]28]. A growing body of evidence suggests that liver damage activates and enhances inflammation in the pulmonary intravascular compartment and lower respiratory tract, leading to important changes in the structure and/or functions of the lung [29,30]. Although all these observations indicate that liver function is an important factor for the development and resolution of ARDS, there is also evidence that such interorgan communication is bidirectional. Thus, acute lung injury is known to impair hepatic function and to aggravate liver diseases by mechanisms involving hypoxemia, activation of systemic inflammatory responses, and cardiovascular changes [24,31,32]. Liver dysfunction is common in critical care patients The frequency of liver damage in critical illness has considerably increased over the last decades [23,[33][34][35], reaching up to 20% of ICU patients in some series and elevating their morbi-mortality [33,34,36]. In critically ill patients, liver dysfunction usually occurs after inflammatory insults such as sepsis and trauma [15,23,33,37], and the underlying interactive mechanisms are complex. The mechanisms of liver dysfunction in critically ill patients implicate microbial products, the paracrine action of cytokines and other inflammatory mediators, hypoxemia, oxidative stress, toxic compounds, hypoperfusion, passive congestion, and effect of nutrition support, among others [34,[38][39][40]. Liver dysfunction can be manifested by plasma elevation of liver enzymes (aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), γ-glutamyl transpeptidase) and bilirubin, decreased plasma levels of albumin and coagulation factors, and/or increased international normalized ratio (INR) [33,36]. The clearance rate of indocyanine green has been used as a dynamic test to assess the functional capacity of the liver. Although the indocyanine green test has shown to reflect better the excretory and/or microvascular dysfunction of the liver, its clinical use has certain limitations [41]. Increased plasma levels of bilirubin are associated with high mortality in critically ill patients [39,42]. Furthermore, hyperbilirubinemia has been proposed as a biomarker of ARDS and found to be an independent factor of mortality in patients with ARDS [26,35,39,43]. Unfortunately, neither bilirubinemia nor other hepatic parameters routinely measured in the clinical setting have the sensitivity and specificity required for an early identification of hepatic injury in critically ill patients [23,33,36,40]. Mechanisms of liver-lung interactions in ARDS The mechanisms by which the liver modulates lung injury involve interrelated elements of systemic and pulmonary host defense, inflammatory responses, and metabolism and include the following (see Fig. 1). Clearance by the hepatic mononuclear phagocyte system of systemic endotoxemia, bacteremia, vasoactive by-products, and procoagulant factors The mononuclear phagocyte system located in the liver, spleen, lung, and bone marrow constitutes the major mechanism to uptake and detoxify bacteria, fungi, viruses, and dying cells, limiting the magnitude and duration of infections [10,44,45]. Although these mononuclear phagocytic cells can exert this function in all these locations, their major mass is in the hepatic sinusoids [10,46]. The hepatic sinusoid is a unique vascular structure with highly specialized endothelial cells (liver sinusoidal endothelial cells) and liver macrophages (Kupffer cells) that reside within the lumen. The cells of the hepatic sinusoid are constantly exposed to gut-derived bacteria, microbial debris, and bacterial endotoxins. Kupffer cells, which line the extensive sinusoidal network, constitute nearly 80-90% of the tissue macrophages present in the body and exert an important role in host defense through phagocytosis and a multitude of secretory functions [46]. The hepatic mononuclear phagocyte system acts as a first line of defense in clearing bacteria and their products. Besides uptake of microbial pathogens and products, Kupffer cells also protect the lung and other extrahepatic organs by removing altered platelets and intravascular coagulation products (Fig. 1) [8][9][10]46]. Dysfunction of the reticuloendothelial system of the liver allows bacterial and microbial products, including the so-called pathogen-associated molecular patterns (PAMs) [47], to reach the lung and the systemic circulation, where they activate pulmonary and systemic inflammatory responses (Fig. 1) [29,[47][48][49]. Indeed, increased plasma levels of endotoxin, probably of intestinal origin, along with increased levels of some cytokines have been found in the blood of patients with acute and chronic liver diseases [50][51][52][53][54]. Pulmonary deposition of intravascular bacteria, and their products alter the structure and function of the lung by different mechanisms including (i) direct cytotoxic effect on alveolar epithelial and endothelial cells, (ii) modulation of local innate immune responses in the lung via activation of toll-like receptors (TLRs), resulting in activation of resident alveolar macrophages and neutrophil influx and in the production of reactive oxygen species, (iii) activation of the coagulation cascades and platelet aggregation, leading to pulmonary microvascular thrombosis [29,48,49,[55][56][57][58][59], and (iv) a sustainable increase in pulmonary vascular resistance [22,30] (Fig. 2). All these mechanisms alter the alveolar endothelial and epithelial cell functions and enhance barrier permeability leading to the formation pulmonary alveolar edema and respiratory failure [14,60], the two main characteristics of ARDS (Fig. 2). Metabolic inactivation and detoxification of endogenous inflammatory mediators The hepatobiliary system has an important capacity to inactivate and detoxicate proinflammatory cytokines, vasoactive mediators, and eicosanoids from the systemic circulation. Removal of all these mediators constitutes a critical element of systemic and pulmonary host defense, protecting the lung and other extrahepatic organs from injury ( Fig. 1) [8][9][10]46]. Like endotoxin, increased levels of cytokines (such as IL-8, IL-1β, ENA-78, TNF-α, MCP-1, MIP-1α,…) and arachidonic acid-derived eicosanoids (thromboxane, leukotrienes) not cleared by the liver have been shown to exert a direct cytotoxic effect on alveolar epithelial and endothelial cells, to activate local innate Fig. 2 Liver damage contributes to the development of acute respiratory distress syndrome. Liver injury leads to changes in the expression of acute-phase proteins (APPs) and to an increase in plasma levels of bacteria/bacterial products, pro-inflammatory cytokines, and pro-coagulant and vasoactive factors in the lung and systemic circulation. These mediators generate deleterious effects on the lung (passage of bacteria /bacterial products and inflammation) and on the gut (intestinal dysbiosis, impairment of gut barrier integrity, leakage of bacteria/bacterial products into the portal circulation and into the mesenteric lymph), resulting in relevant changes in the hepatic and pulmonary microbiota and promoting inflammation and oxidative stress in liver and lung tissues. In addition, lung-derived cytokines promote the synthesis of APPs and activation of inflammatory responses in the liver. All these responses mediated by the gut-liver-lung axis contribute to lung injury and multiple organ dysfunction in critical illness. IL interleukin, TNF tumor necrosis factor, INF interferon immune responses and to promote platelet aggregation in the lung, contributing to the development of diffuse alveolar damage (DAD) [55][56][57][58][59]. Hepatic synthesis of inflammatory mediators that can activate pulmonary alveolar macrophages and, consequently, increase inflammation in the lung Hepatic mononuclear cells include a heterogeneous population of lymphocytes, Kupffer cells (hepatic resident macrophages), monocytes, and granulocytes that perform vital functions for the innate and adaptive immune system. In response to injury, activation of these hepatic mononuclear cells enhances the production and release of inflammatory mediators, such as IL-1, IL-6, TNF-α, platelet-activating factor (PAF), and leukotrienes, into the systemic circulation [61], where they play an important role in the lung-liver interaction [18,31,51,[61][62][63][64]. These liverderived inflammatory mediators alter lung structure function early in acute inflammatory diseases (such as sepsis) and contribute to some extent to lung damage upon activation of pulmonary alveolar macrophages ( Fig. 1] [17,65]. In this line, elevated levels of TNF-α and IL-1β, two cytokines that are mainly synthetized by alveolar macrophages, have been found in the lungs of rats with carbon tetrachloride (CCl4)-induced cirrhosis, along with an increase in lipid peroxidation (TBARS) and antioxidant enzymes (superoxide dismutase and catalase) in the liver and lung tissues. These events are also associated with altered gas exchange and changes in the size of pulmonary vessels in these rats [66,67]. Besides high levels of endotoxin [50,68], patients with liver disorders also have high circulating levels of TNF-α, IL-1, and IL-6 [51][52][53][54] because of the altered capacity for inactivation and detoxification and the increased synthesis of pro-inflammatory mediators by the liver [9,44,46,61]. These specific cytokines have been shown to modulate systemic inflammatory responses and participate in the development of lung damage [69][70][71][72]. Therefore, it is possible that cytokines of hepatic origin may control and modulate the local host defense and immune system of the lung, contributing to lung injury ( Figure 2). The liver is the main organ responsible for the acute-phase response The organism responds to tissue injury or infection by local changes such as those associated to inflammation and by a coordinated sequence of systemic and metabolic process, known as the acute-phase response, aimed to restore homeostasis and recover from injury [63,[73][74][75][76]. One of the mayor characteristics of this acute-phase response is a change in plasma concentration (either increase or decrease) of the acute-phase proteins (APPs) expressed in the liver [74]. Cytokine-driving synthesis of acute-phase proteins in the liver modulates the systemic and pulmonary host inflammatory responses and intermediary metabolism (Fig. 1) [22,30,48]. The hepatic APPs have a variety of functions that include microbicidal and phagocytic activity (e.g., LPS binding protein, complement components, C-reactive protein), recruitment of immune cells to inflammatory sites (e.g., serum amyloid A), hemostasis (e.g., fibrinogen, α1-acid glycoprotein), antioxidant, and prevention of iron loss (e.g., haptoglobin) and antiproteolytic actions to counterbalance protease activity at sites of inflammation (e.g., α2macroglobulin, α1-antitrypsin, and α1-antichymotrypsin) (Fig. 1) [63,74,75]. While the local inflammation occurs in the alveolar airspaces of patients with ARDS, the acute-phase response is induced in the liver [73,76,77]. Interestingly, in pneumonia-induced ARDS, this liver-derived acute-phase response occurs independently of bacterial dissemination and depends instead on inflammatory signaling molecules derived from the pulmonary immune cells, such as the cytokines IL-1, IL-6, and TNF-α [10,62,63,73,74]. Then, these lung-derived cytokines can travel from the lung into the systemic circulation and ultimately modify acute-phase gene expression in the liver [63,73,78,79] upon activation of the transcription factors STAT3 (signal transducer and activator of transcription 3) by IL-6 and activation of RelA (v-rel avian reticuloendotheliosis viral oncogene homolog A, also known as NF-kB3) by the earlyresponse cytokines TNF-α and IL-1 [73,78,79]. In response to these cytokines, the liver changes the expression of many acute-phase proteins such as C-reactive protein, α-1 antitrypsin, serum amyloid A protein, and others [10,62,63,74,77], which in turn can directly travel back to the lung and pass into the airspaces where they cause inflammation, predominantly via activation of alveolar macrophages (Fig. 2) [5,14]. These phagocytic cells are targeted by multiple hepatic APPs such as SAA [80,81], SAP [82], LBP [83], and C-reactive protein [84,85]. Once activated by these hepatic APPs, alveolar macrophages release cytokines (IL-6 and CXCL1) that enhance local inflammation, in part by promoting neutrophil influx to the insterstitium and alveolar airspaces. Excessive inflammation in the alveoli may result in an increase in oxidative stress and lung injury [86,87]. Besides this potentially deleterious effect, hepatic APPs at the site of plasma extravasation has other potential functions, including opsonization of bacteria, leukocyte activation, antiprotease, antioxidant activities, and modulation of the coagulation pathway [63,75]. These mechanisms help to regulate host defense, limit excessive inflammation and immune responses, and promote bacterial clearance, preventing infection dissemination and reducing the risk of organ damage in the setting of pneumonia and sepsis. Also, hepatic APPs exert liver protection by countering TNF-dependent toxicity in the liver and attenuate systemic inflammation and mortality in sepsis and pneumonia-induced ARDS (Fig. 1) [79,86,[88][89][90]. Altogether, the bidirectional liver-lung axis mediated by APPs is critical for integrating systemic and pulmonary responses, balancing regulation of multiple host defenses and activation of inflammation to restore homeostasis and recover from organ injury [48,61,86]. Disbalance in this liver-lung communication can be an important factor in the initiation and progression of ARDS and of the damage to other organs [73]. Nutrients, bile, and hormone production The liver plays an important role in regulating metabolic homeostasis and in the synthesis and processing of lipids and carbohydrates that supply energy to other organs [91]. It is also the major site of synthesis of key proteins and bile acids that are critical for the normal uptake of vitamins and lipids [92]. Therefore, alterations in the flux of carbohydrates and lipids through the liver can indirectly impact distal organs due to alteration of their energy statuses [93]. In addition, hyperbilirubinemia in the context of liver diseases has been shown to cause some lung-specific deleterious effects, by entering the lung tissue, reaching the alveolar airspaces, and deteriorating the surface tension properties of the alveolar surfactant [94]. Although bilirubin has antioxidant properties, high bilirubin levels can also activate oxidative stress, apoptosis, and inflammatory responses in different cell types and organs [95][96][97][98]. Therefore, hyperbilirubinemia may actively participate in the development of ARDS, although the underlying mechanisms have not been fully elucidated. Finally, the liver produces several hormones that mediate diverse extrahepatic effects, such as insulin-like growth factor, angiotensinogen, and thrombopoietin, which have been shown to influence the development of ARDS (Fig. 1) [55,99,100]. The gut-liver-lung axis The pathogenic mechanisms of ARDS should be considered within a gut-liver-lung axis. Growing evidence indicates that intestinal microbiota and the mucosal immune system of the gut have an important impact on the function of the gastrointestinal tract itself and extra-intestinal organs, such as the lung and the liver [29,48,49]. Liver cirrhosis and other liver diseases favor the gut-derived bacterial translocation into the liver and lung by several mechanisms (Fig. 2). First, patients with liver disorders have intestinal dysbiosis characterized by a significant shift of the microbial composition toward pro-inflammatory bacteria. This gut dysbiosis is accompanied by activation of local intestine immune responses and impaired gut barrier function. A leaky gut barrier facilities bacterial translocation of live bacteria or their microbial products from the intestinal lumen to the liver, via portal circulation, and to systemic circulation and the lung via the mesenteric lymphatic system [29,101]. In the lung and the liver, gut microbiota can directly modulate their local immune cells (mainly alveolar macrophages and Kupffer cells, respectively) via activation of toll-like receptors (TLRs) and indirectly via different bacterial metabolites and signaling molecules, such as PAMPs [29,48,49,90]. Activated alveolar macrophages in the lung and Kupffer cells in the liver release proinflammatory cytokines, contributing to the initiation and/or progression of lung and liver damage and activation of systemic inflammation [29,101], which can also cause dysfunction in other organs. Second, liver dysfunction can imply less capacity of the liver to remove bacteria, bacterial products, and inflammatory mediators from circulation, leading to increased levels of these molecules in blood. Third, this pathological gut-derived bacterial translocation could cause important changes in the lung microbiome (Fig. 2) [90,102,103]. Indeed, pulmonary microbiome is frequently enriched with gut-related bacteria (Bacteroidetes and Enterobacteriaceae) in critically ill patients [102,103]. As a consequence of liver diseases, this gut-derived bacteria and accumulation of PAMPs, cytokines and other pro-inflammatory molecules in the systemic circulation can potentially cause or exacerbate lung injury upon TLR-4-mediated activation of intravascular and alveolar macrophages within the lung and recruitment of neutrophils and direct toxic effects of bacterial products on pulmonary microvasculature (Fig. 2) [68,[102][103][104][105][106][107][108]. Altogether, the gut-liver-lung axis seems to exert a relevant role in the initiation and modulation of hepatic, pulmonary, and systemic immune responses that contribute to the damage of the liver, the lung, and other organs. Extracellular vesicles Extracellular vesicles (EVs), a term that includes microvesicles (MVs), exosomes, and apoptotic bodies, represent an emerging mechanism of interorgan communication in many diseases, including liver diseases and ARDS [106,[109][110][111]. Extracellular vesicles are defined as membrane-bound vesicles, ranging 0.1-1.0 μm in diameter, which are released from cells by the budding of the cellular plasma membrane and carrying a diverse cargo, including lipids, proteins, RNAs, and miRNAs. The EVs are recognized as important mediators of interaction between different organs, and they are considered attractive therapeutic targets in different diseases [112,113]. Notably, the levels of circulating EVs have been reported to be increased both in patients with cirrhosis and ARDS. However, the potential role of circulating EVs in mediating liver-lung communication in the context of ARDS is not currently understood, representing an interesting topic for further investigation. Conclusions Liver injury and hepatotoxicity occur frequently in critically ill patients, and significantly influence their prognosis. Patients with severe hepatic dysfunction are at high risk for irreversible ARDS because of multiple defects in host defense and dysregulation of inflammatory responses. Interrelations between hepatic and pulmonary functions influence the development and progression of ARDS and play a central role in the resolution of lung damage by several mechanisms. First, the liver regulates host defense and modulates systemic inflammation. Also, the liver activates acute inflammatory responses in the lung early in the development of ARDS. Although promoting inflammation can be detrimental in the context of acute lung injury, the liver response to an inflammatory insult is also pro-defense and pro-survival. The understanding of the complex relation between the liver and the lung requires further research in order to improve the clinical management and to identify new diagnostic and therapeutic options for patients with or at risk for ARDS.	2020-12-18T14:29:51.775Z	2020-12-01T00:00:00.000	{ "year": 2020, "sha1": "dc59c85f3786636b04b668447b306f3f119037a9", "oa_license": "CCBY", "oa_url": "https://icm-experimental.springeropen.com/track/pdf/10.1186/s40635-020-00337-9", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "dc59c85f3786636b04b668447b306f3f119037a9", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
83586850	pes2o/s2orc	v3-fos-license	Descriptions of adults , immature stages and venom apparatus of two new species of Dacnusini : Chorebuspseudoasphodeli sp . n . , parasitic on Phytomyza chaerophili Kaltenbach and C . pseudoasramenes sp . n . , parasitic on Cerodontha phragmitophila Hering ( Hymenoptera : Braconidae : Alysiinae ; Di The imagines, cast skin of the final larval instar, and venom apparatus of Chorebus pseudoasphodeli sp. n., an endoparasitoid of Phytomyza chaerophylli Kaltenbach, 1856 on Daucus carota L., and Chorebus pseudoasramenes sp. n., an endoparasitoid of Cerodontha phragmitophila Hering, 1935 on Arundo donax L., two new species of Dacnusini from Spain, are described, illustrated, and compared with those of allied species. The immature larvae, mature larva, and pupa of C. pseudoasphodeli sp. n. are also described, illustrated, and compared. Morphological structures of phylogenetic value are discussed, and keys for the discrimination of the imagines are provided. INTRODUCTION The Alysiinae, a subfamily of braconid wasps, are largely characterized by their so-called "exodont" mandi bles, a condition in which the teeth of the mandibles face anteriorly and do not overlap when the mandibles are closed (Wharton et al., 1997).They are believed to use these to emerge from the puparia of the cyclorrhaphous Diptera that they parasitize.This subfamily has traditionally been divided into the tribes Alysiini (with three submarginal cells) and Dac nusini (with two submarginal cells).The latter lack the r-m vein.Chorebus Haliday, 1833 is the largest genus of the Dacnusini with approximately 215 Holarctic species, and can be defined as follows: mandibles with four teeth, or metapleuron with a rosette of setae around a central swelling; most species have both characters (Griffiths, 1968;Tobias, 1995).The additional tooth is located between the middle tooth and the original lower tooth (the four-toothed mandibles of other Dacnusini have devel oped the additional tooth in a different position: on the dorsal side of the elongate 2nd tooth).The rosette of setae around a swelling on the metapleuron and 4-toothed man dibles do not appear together in all species.Species of Chorebus are solitary endoparasitoids of Agromyzidae and Ephydridae (Diptera). Most of the taxonomy of braconid wasps is based on adult external morphology, and relatively little attention has been given to interspecific differences in larval or internal characters even though they could be useful. The reproductive apparatus of Hymenoptera, and its associated venom apparatus, vary in ways that may pro vide phylogenetic information and be useful for discrimi nating closely related species (Edson & Vinson, 1979;Quicke & Achterberg, 1990;Whitfield, 1992).For the Dacnusini, detailed studies of variation in the gland and morphology of the reservoir of the venom apparatus have so far only been done by Quicke et al. (1997). In this report the imagines, cast skin of the final larval instar and venom apparatus are described for two new species of Dacnusini that the authors discovered in Spain parasitizing agromyzid hosts: Chorebus pseudoasphodeli sp.n., and Chorebus pseudoasramenes sp.n.The imma ture larvae (first, second instars), mature larva (third instar) and pupa of C. pseudoasphodeli are also described.Additionally, the hosts of both species are cited, together with their host plants. MATERIAL AND METHODS Imagos and exuviae of the final larval instar of C. pseudoasphodeli sp.n. were obtained in August 1988 from pupae of Phytomyza chaerophylli Kaltenbach, 1856, whose larvae were found mining the leaves of Daucus carota L., at Foyos (Valen cia, Spain).Imagos and exuviae of the final larval instar of C. pseudoasramenes sp.n. were obtained in June 1989, from pupae of Cerodontha phragmitophila Hering, 1935, on Arundo donax L., at El Puig (Valencia, Spain).In both cases plant material infested with the host larvae was collected; this was placed in plastic bottles, of suitable dimensions, whose openings were covered with gauze held in place with a rubber band.These bot tles were kept under ambient conditions of temperature, RH and photoperiod.The methodology used for opening the puparia and preparing the cast skins is that proposed by Wahl (1984). For study of the egg and larval instars of C. pseudoasphodeli, leaves of D. carota L. mined by Phytomyza chaerophylli were collected in June 1990; these leaves were taken to the laboratory and placed in the same type of bottle as those described above, which were then kept in ambient conditions.The parasitoids that emerged from the host puparia were offered blotting paper impregnated with honey or sugar-water as a source food.Females, aged between 24-72 hours, were placed individually in Petri dishes (9 x 1.5 cm) and provided daily with leaves of D. carota infested with different larval instars of P. chaerophylli.To obtain recently deposited eggs of C. pseudoasphodeli the host larvae were dissected immediately after oviposition.To study the larval development of the parasitoid, parasitized mate rial was kept in a chamber at 21-23°C, 60-80 RH, and a photo period of 16L : 8D, and the hosts periodically dissected.When possible, these observations were complemented with dissection of hosts parasitized in the field.All dissections were carried out in 0.9% saline solution.For development of the egg and larvae, the methodology used by Guppy & Meloche (1987) was fol lowed.The microscopic preparations of the three larval instars were prepared by following the method of Evans (1987) and Sime & Wahl (1998). The venom apparatus of both species was prepared and illus trated according to the method described by Quicke et al. (1992Quicke et al. ( , 1997) ) (chlorazol black method) for dry museum specimens.The venom apparatus, of one specimen of each species, was treated with hydroxide solution, which removed the soft tissue so that the remaining characters of the chitinous gland intima can be seen (these characters are not apparent when an intact gland and reservoir is examined). The morphological terms for the body of the imago and wing venation follow Wharton et al. (1997); the biometric data fol low Wharton (1977).The terminology used to refer to the dif ferent structures of the immature stages is that of Finlayson & Hagen (1979), Hagen (1964), Short (1978), and Sime & Wahl (1998).The characters of the gland and reservoir parts of the venom apparatus studied were as follows: a) the presence/ absence of an unsculptured region anterior to the reservoir, b) the number of reservoir subdivisions, c) the reservoir neck region: 1) without narrowing, 2) parallel-sided neck, 3) with such a neck, d) reservoir: 1) less than 4 times longer than maximal width, 2) 4-6 times longer than maximal width, 3) more than 6 times longer than maximal width (all excluding the anterior unsculptured bulb, e) secondary venom duct with annular/spiral sculpture: 1) absent, 2) distinct but not extensive, 3) well developed and extensive, f) venom gland: 1) extensively branched or with multiple filaments or sacks not separated by two long unbranched ducts after division of secondary venom duct, 2) with extensive branching or numerous sacks at the end of a pair of long tertiary ducts, 3) with only one pair of unbranched filaments, 4) with a single filament, g) venom gland inserted on reservoir: 1) before posterior end, 2) at extreme pos terior end, h) secondary venom duct: 1) not narrow, more or less gradually developing into gland filaments, 2) narrow, termi nating abruptly at globular glandular sacks.The terminology used follows Quicke et al. (1997) (Fig. 6).All the material examined (imagines, immature stages and venom apparatus) is deposited at the Facultad de Biología, Universidad de Valencia (Valencia, Spain).The following abbreviations were used for the preimaginal stage descriptions: d = diameter; h = height; l = length and w = width. Female.Head transverse, in dorsal view between eyes 1.8-1.9times as wide as long; 2-2.1 times as wide as face; 1.3-1.4times as wide as mesoscutum; face 1.1-1.2times as wide as high; eyes, in lateral view, 1.2 times as long as temples.Face with short pubescence, directed both towards the centre and downwards.Mandibles (Fig. 1a) slightly expanded towards their apex, 1.2 times as long as wide, with all four teeth well developed.Antennae as long as body, with 18 flagellomeres (20 antennomeres). Mesosoma 1.6-1.7 times as long as wide.Sides ofpronotum bare.Mesoscutum 1.1 times as wide as long, roughened anteriorly, with white pubescence covering its anterior face and median lobe, but lateral lobes with only pubescence anteriorly, notauli only on declivity.Metapleuron smooth, with a swelling displaying a rosette of poorly developed setae that only show a weak tendency towards differentiation in the direction of the setae.Propodeum rugose-punctate with most of its surface covered with a dense and fine pubescence.Hind femur 0.75 times as long as hind tibia; hind tibia 1.05 times as long as hind tarsus. Colour.Body black, with exception of: palpi yellow; labrum yellow-brown; centre of mandibles reddishorange; antennae dark brown, excepting scape, pedicel, and basal half of first flagellomere, which are orangeyellow; legs orange-yellow; second metasomal segment orange-yellow. Body length 1.77 mm. Male. Like female but with: a) longer antennae, with 19 flagellomeres (22 antennomeres, b) petiole more styl ized, 1.8 times as long as wide apically, with almost par- allel sides, and with more pronounced stigmatiferous tubercles; c) shorter: 1.55 mm. Differential diagnosis.This new species, in which the swelling of the metapleuron is smooth with a poorly developed rosette of setae, is included in the "plesiomorph species of Chorebus" (Griffiths, 1968), and is closest to C. asphodeli Griffiths, 1968, from which it is distinguished by the following characters: 1) antennae as long as body; 2) sides of pronotum bare; 3) wings with 2-1A totally decoloured; 4) metapleuron smooth; 5) col our: a) labrum (yellow-brown), b) palpi (yellow), c) centre of mandibles (orange-red), d) antennae (dark brown, with the exception of scape, pedicel, and basal half of first article of the flagellum, which are orangeyellow), e) legs (orange-yellow) and f) second metasomal segment (orange-red). This species can be inserted in the keys of Griffiths Etymology.The specific name of this species refers to C. asphodeli Griffiths, 1968, which it strongly resembles.Female.Head transverse, in dorsal view, between eyes 2-2.1 times as wide as long; 2-2.1 times as wide as face; 1.3-1.4times as wide as mesoscutum; face 1.2 times as wide as high; eyes, in lateral view, as long as temples.Face strongly punctate.Mandibles not expanded, 0.8-0.9times as wide as long, with tooth 2 long and pointed, but teeth 1, 3 and 4 well developed (Fig. 2a).Antennae with 30 flagellomeres (32 antennomeres). Chorebus pseudoasramenes Mesosoma elongate, about 1.8 times as long as wide.Sides of pronotum covered with fine pubescence.Mesoscutum 1.2-1.3times as wide as long, roughened anteriorly, with white pubescence covering its median lobe, and the lateral lobes almost bare, notauli only devel oped at their bases.Metapleuron with a rugose-punctate swelling; metapleural and propodeal pubescence whitish and dense.Hind femur 0.6 times as long as hind tibia. Colour.Body black, with the exception of: palpi yel low; labrum orange-red; clypeus red-brown; centre of mandibles red-brown; antennae yellow; legs yellow, with the hind coxae infuscated; second metasomal segment brownish. Body length 2.1 mm. Differential diagnosis.This new species belongs to the "Chorebus ovalis/lateralis complex" (Griffiths, 1968), and is closest to C. asramenes (Nixon, 1937) from which it can be distinguished by the following characters: 1) Mandibles with all four teeth well developed.2) Face strongly punctate.3) Lateral lobes of mesoscutum almost bare.4) Colour of labrum (orange-red) and palpi (yellow). This species can be inserted into the keys of Griffiths (1968) (Part VI: 131) and Tobias (1995) Etymology.The specific name of this species refers to C. asramenes (Nixon, 1943), to which it is very similar. Chorebuspseudoasphodeli sp. n. Eggs and first larval instar were found in larvae of the host that were at different stages of development.Second and third larval instars and pupae were only found in hosts that had already pupated. Larva.1st instar.Some larval measurements are given in Table 1.General aspect (Fig. 3c).Body (length and width, Table 1) with head well defined and 13 body seg ments, caudate, vermiform, transparent, slightly curved ventrally.The last two abdominal segments almost fused together; the last one slightly modified into a short organ, blunt, rounded, in the shape of a tail (length, Table 1), with 30-45 microtrichia (l = 2 pm, n = 8) in groups around the place where one would expect to find the anus.Segments 2-12, with the exception of the protho racic and last abdominal ones, each with a row of short spines (l = 6 p.m) on back (n = 8): mesothorax (row with 6 spines), metathorax (10 spines) and abdominal seg ments (between 15-28 spines).Cranium (Fig. 3d) (length and width, Table 1) slightly flattened dorsoventrally, with the following differentiated but only very slightly sclero-tized sclerites: epistoma (e), (p), anterior (pa) and posterior (pp) pleurostomal processes and hypostoma (h).Mouthparts: Mandibles well defined, with an oblong molar lobe and with a sharp blade, slightly curved, and well sclerotized. 3rd instar.General aspect (Fig. 3g).Typical hymenopteriform (l = 1.88-2.03mm, w = 0.71-0.85mm, n = 2), with head, the three thoracic segments and the abdominal segments well defined; yellowish-white.Integument with conical papillae (5x5 pm) covering thoracic and abdominal segments, except in intersegmental zones and around the spiracles and anus; posterior part of each seg ment surrounded by a ring of setae (40-70 per segment, mean length of setae = 6 pm).Nine pairs of spiracles (d = 10 mm), with the atrium and closure apparatus well dif ferentiated, one pair in intersegmental zone between first and second thoracic segments and one pair on anterior edge of each of the first eight abdominal segments. Pupa.Exarate.In the single pupa examined a cocoon was not observed. Exuviae. Cast skin of Chorebus pseudoasphodeli.Of the five exuviae available for study only one was measured.Tegu ment weakly sclerotized with the exception of the spira cles, with campaniform sensilla (d = 2 pm).Spiracles with small atrium (d=10 pm). Chorebuspseudoasramenes sp. n. Cast skin of Chorebus pseudoasramenes.Of the two exuviae available for study, only one was measured refer to only one of them.In this description, the unmentioned structures or non-specified measurements are assumed to be similar or of the same order as those of the previous species.Only differences are detailed. The first instar larva of this species, according to the classifications of Clausen (1962) and Hagen (1964) is of the caudate-mandibulate type.This larva is more similar to the first instar of D. areolaris [Haviland, 1922] body segments.It is possible that the mandibles, in this first instar, serve to break the chorion, as reported by Guppy & Meloche (1987) for D. dryas.Alternatively, they could be used to kill other larvae and prevent super or multiparasitism as suggested by Hagen (1964) for the caudate-mandibulate larvae of Braconidae.The functions of the tail and spines is highly controversial (Ullyett, 1944;Hagen, 1964). The second instar larva is similar to the mature larva, as in many Hymenoptera Parasitica.The second instar has lost its tail and spines, and the mandibles are very reduced.This latter character "presence of mandibles" separates the second instar of this species from those described for Dacnusa (Haviland, 1922;Guppy & Meloche, 1987), but brings it closer to the third instar of C. posticus.According to Wright et al. (1946) the third instar of this species, although lacking mandibles, has a pair of weak oral papillae.Perhaps these structures are very reduced and depigmented mandibles, as in C. pseudoasphodeli. The mature larva is very similar to that already described for C.posticus, from which it differs, mainly, in the arrangement of the setae on the tegument and the absence of cephalic papillae. The cast skin of the final larval instar of C. pseudoasphodeli and C. pseudoasramenes, like that of C. denticurvatus, which is the only species of the genus whose final larval instar has been described in detail, shows: a) sim ple untoothed mandibles b) a reduced labial sclerite, c) a pleurostoma with well defined mandibular processes and d) a long stipital sclerite.The only appreciable differences between them and the above species is in the type, number and arrangement of the tegumental differentia tions: sensilla, setae and papillae. VENOM APPARATUS The two species of Chorebus described here have a venom apparatus with the characters specified by Quicke et al. (1997) for Chorebus s.str.Both C. pseudoasphodeli (Fig. 7) and C. pseudoasramenes (Fig. 6) have: large swollen region anterior to the reservoir; an undivided res ervoir; a reservoir neck region without narrowing, parallel-sided; a reservoir not more than six times longer than its maximal width; a secondary venom duct without annular/spiral sculpture; an extensively branched venom gland; a venom gland inserted at the extreme posterior end of the reservoir, and no narrow secondary venom duct.Both species can be differentiated as follows: From the data available on the preimaginal states and morphology of the venom apparatus of the Dacnusini, it is clear that further descriptions are necessary, both of the larvae and the venom apparatus, before inferring phyloge netic relationships from their morphological features.Thus, although Quicke et al. (1997) observed morpho logical differences between the venom apparatus of C. posticus and those of other Chorebus s.str., which allow the separation of this species from the rest of the genus, this is not confirmed by significant differences in the prei maginal instars.The most marked differences between the mature larvae of C. posticus and those of Chorebus spe cies, which are described in greatest detail: C. denticurvatus, C. pseudoasphodeli and C. pseudoasramenes, lie in the arrangement of the setae on the tegument and the absence of cephalic papillae.In the immature larvae, the only potentially interesting aspect is the absence of spines on the tegument of the first instar larva of C. posticus and the presence of such structures in C. pseudoasphodeli.Finally, although the number of larval instars in the last species differs from that in C. posticus, this is not impor-tant since despite the tendency of endoparasitoid Hymenoptera to show fewer than 5 larval instars (Quicke, 1997), it may vary even within the same genus (Hagen, 1964). than to that of the only species of Chorebus described: C. posti cus.Like the larva of the species of Dacnusa Haliday, 1833 it displays spines on the back of a large number of sp.n.; 7 -venom apparatus of Chorebus pseudoasramenes sp.n.The terminology used for the venom gland and reservoir parts is shown in figure 6: a = anterior swollen region (bulbous expansion); b = reservoir with spiral sculpture; c = reservoir neck region; d = venom gland; e = gland filament (sack); f = secondary venom duct.	2018-12-03T21:05:55.788Z	2003-09-15T00:00:00.000	{ "year": 2003, "sha1": "94db2a128a058152158cae76577aade52a202a80", "oa_license": "CCBY", "oa_url": "http://www.eje.cz/doi/10.14411/eje.2003.061.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "94db2a128a058152158cae76577aade52a202a80", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology" ] }
53368981	pes2o/s2orc	v3-fos-license	LINKING ORGANIZATIONAL STRUCTURE, TECHNOLOGICAL SUPPORT AND PROCESS INNOVATION: THE MEDIATING ROLE OF KNOWLEDGE SHARING IN THE IRAQI TEXTILE INDUSTRY This paper investigates the relationships among organizational structure, technological support, knowledge sharing and process innovation and whether knowledge sharing has a mediating effect on these relationships. Based on the survey among employees in the Iraqi textile industry, the results revealed that organizational structure and technological support positively and significantly influence knowledge sharing. Knowledge sharing was also found to be an important mediator between organizational structure, support technology and process innovation. The findings bear implications to the Iraqi government’s call for innovation in the Iraqi textile industry. INTRODUCTION Process innovation is unique, complex and difficult to control (Gerybadze, Hommel, Reiners, and Thomaschewski, 2010). It describes the activities that are performed at each stage of the development of innovation (Ortt and Duin, 2008). Process innovation is defined as developing a new or substantially improved production process through new equipment or reengineering of operational process (Wong and He, 2003). Capon, Farley, Lehmann and Hulbert (1992) suggest process innovation as tools, devices, and knowledge in throughput technology that mediate between inputs and outputs and are new to an industry, organization, or subunit. Knowledge management can play a pivotal role in supporting and nurturing innovation (Chen and Huang, 2009;Gold, Malhotra and Segars, 2001). Knowledge sharing plays a potentially mediating role in connecting organizational structure and technological support with organizational innovation process. Successful knowledge sharing is believed to have the potential of enhancing an organization's competitive advantage, customer focus, employee relations and development, innovation, and lower costs. Finally, the organizational promotion of knowledge sharing is changing traditional ideas concerning managing intellectual resources and employee work styles by providing new processes, disciplines and cultures, thus constituting an organizational innovation (Darroch and McNaughton, 2002). Knowledge sharing and process innovation in the Iraqi textile industry are relatively new concepts whereby the industry is dominated by 6 government-owned manufacturers who run the textile factories. The researchers conducted preliminary interviews with the manufacturers and were informed that whereas the factories share common organizational structure, the operations were lacking in technological support and knowledge sharing as they were still using traditional technology and focused on production, not process innovation. Therefore, the purpose of this paper is to examine the possible mediating effect of knowledge sharing on the relationship between organizational structure and technological support withprocess innovation in the Iraqi Textile industry. Process Innovation Today's competitive business environment, innovation is absolutely essential. Superior innovation provides companies with opportunities to grow faster, better, and smarter than their competitors. Ultimately this can enable a company or organization to influence the direction of its industrial sector (Davila, Epstein and Shelton, 2006). Cetindamar and Ulusoy (2008) defined "innovation" as the process leading to the adoption and diffusion of new technologies with the express aim of creating new processes, products and services. Damanpour (1991) defines innovation in a broader way by viewing it as a new product or service, a new production process technology, a new structure or administrative system, or a novel plan or program pertaining to organizational members. Process innovation is viewed as a pervasive and embracing process which includes research, development, and implementation of new ideas and behaviors (Damanpour, 1996). Extensive literature reviews suggest that process innovation depends on several dimensions: level, driver, direction, source and locus (Crossan and Apaydin, 2010). Level explains the split between individual, group and firm's processes. Driver deals with both, internal and external driver. Internal drivers of the innovation process can be available knowledge and resources. Prajogo, Power and Sohal (2004) identified two types of innovation performance, product and process. These classifications enable the grouping of organizational innovation into product innovation and process innovation. Prajogo et al. (2004) stressed that knowledge management has substantial positive relationship to both product innovation and process innovation. Gerybadze, Hommel, Reiners and Thomaschewski (2010) mentioned that it is important to monitor innovation from different perspectives such as structure related, market related, product/project related, process/performance related and culture related.In considering innovation as a process, unitary sequence model is used to explain process innovation (Gopalakrishnan and Damanpour, 1997). Organizational Structure Organizational structure refers to the formal allocation of work roles and administrative mechanisms to control and integrate work activities (Ghani, Jayabalan, and Sugumar 2002;Robbins 1990). According to Claver-Cortés, Zaragoza-Saez and Pertusa-Ortega (2007), organizational structure can be defined as the result of the combination of all the ways in which work can be divided into different tasks, the coordination of which must afterwards be ensured. Structures that promote individualistic behavior in which individuals, divisions, or functions are implicitly or explicitly required to keep knowledge private have been shown to inhibit effective management of knowledge in organizations (Gold et al. 2001, O'Dell andGrayson 1998). The knowledge-based view emphasizes the importance in understanding the processes through which organizations access and utilizes knowledge possessed by its individual members (Grant, 1996). Structure can influence knowledge management processes through shaping patterns and frequencies of communication among organizational members, stipulating locations of decisionmaking and affecting efficiency and effectiveness in implementing new ideas (Zheng, Yang and McLean, 2010). Despite inconclusive findings regarding the relationship between organizational structure and knowledge management (Tsai, 2002), a decentralized structure has often been seen as facilitative to knowledge management success (Damanpour, 1991;Deal and Kennedy, 1982;Gold et al., 2001). High centralization inhibits interactions among organizational members (Gold et al., 2001), reduces the opportunity for individual growth and advancement (Kennedy, 1983), and prevents imaginative solutions to problems (Deal and Kennedy, 1982). Therefore, it is posited that: H1. Organizational structure relates positively to knowledge sharing. Two important aspects of organizational structure are centralization and formalization (Eppler and Sukowski 2000;Jarvenpaa and Staples 2001). Accordingly, organizational structure is construed as a second-order construct encompassing these dimensions. Centralization refers to the locus of decision authority and control within an organizational entity Caruana, Morris and Vella 1998;Ein-Dor and Segev 1982). Centralization of decision making can result in complex and time-consuming communication channels (Bennett and Gabriel 1999), which may introduce distortion and discontinuity of ideas and knowledge as they are passed along multiple levels of authority (Stonehouse and Pemberton 1999). Formalization refers to the degree to which decisions and working relationships are governed by formal rules, standard policies, and prescribed procedures. Smith (2006) argues that organizational structure facilitates the discovery, creation, sharing, exchange, and transference of new knowledge within the organization. It is the main mechanism through which organizations supervise the behavior of employees and coordinate individuals and functions (Holsapple and Joshi 2001;Rapert and Wren 1998). Therefore, it is posited that: H2.Organizational structure relates positively to process innovation. Technological Support Sher and Lee (2004) indicated technology is certainly conceptually complicated and multidimensional. It exists in several types which include artifact, knowledge, and process. Artifact technology can relate like tools, techniques, and actions used to change organizational inputs into outputs. Similarity study by Tippins and Sohi (2003) show the technological support enhance individual's connectivity and facilitate knowledge structuring and interpretation. According to (Kim and Lee, 2005) show support technology is multi-dimensional and organization must invest in an extensive infrastructure that supports the different kinds of knowledge. Technology tools for knowledge sharing include electronic bulletin boards, discussion forums, knowledge directories, groupware, databases, intranets, intelligent search engines, personal web pages, electronic mail, virtual conference rooms, libraries, corporate yellow pages, among many others (Alavi and Leidner, 2001;Bender and Fish, 2000). It is evident that technology enables and aids core knowledge activities such as knowledge creation, knowledge sharing, knowledge distribution, and knowledge application (Gold, Malhotra and Segars, 2001). Similarly, it is clear that technology is an integral part of the KM application. Specific technologies play a fundamental role in promoting the KM movement (Barney, 1991). Kharabsheh (2007) draws a very clear relationship between technology and knowledge sharing. Therefore, it is posited that: H3. Technological support relates positively to knowledge sharing. H4. Technological support relates positively to process innovation. (2009) defined knowledge sharing as a process where individuals exchange knowledge (tacit or explicit) and together create new knowledge. Knowledge sharing can play a pivotal role in supporting and nurturing innovation (Chen and Huang, 2009). As noted earlier, various scholars have proposed the importance of managing knowledge in the organizations, hence implying that the implementation of KM would be conducive to process innovation. Knowledge creation involves a continuous process through which one overcomes the individual boundaries and constraints imposed by information and past learning, by acquiring a new context, a new view of the world, and new knowledge. By interacting and sharing tacit and explicit knowledge with others, the individual enhances the capacity to define a situation or problem, and apply his or her knowledge to solve the problem effectively (Nonaka, Von Krogh and Voelpel, 2006). Nonaka (1991), and Nonaka and Takeuchi (1995) contend that the concept of innovation and "knowledge creation" are closely related. Along this line, it is generally assumed that the process of innovation consists of an ongoing pursuit of harnessing new and unique knowledge (Subramaniam and Youndt, 2005). Hence, it is posited that: Mohd and Zawiyah H5. Knowledge sharing relates positively to process innovation. Mediating role of Knowledge Sharing Effective knowledge management contributes to competitive advantage (Connor and Prahalad, 1996), improved performance (Teece, 1998), and innovation (Nonaka and Takeuchi, 1995).Knowledge resources are an outcome of organizational culture, structure, because knowledge is created, made sense of, and utilized in accordance with a set of cultural values and norms, embedded in structural relationships, For example, knowledge sharing practices are affected by cultural expectations such as what knowledge should be shared with the organization and what should be hoarded by individuals, by structural relationships such as how quickly the knowledge flows through formal reporting relationships. In turn, organizational knowledge reflective of cultural, structural, characteristics of the organization is utilized to help produce new products and services, improve efficiency, and enhance effectiveness (Nonaka, Toyama and Konno, 2000).Finally, the organizational promotion of knowledge sharing is changing traditional ideas concerning managing intellectual resources and employee work styles by providing new processes, disciplines and cultures, thus constituting an organizational innovation (Darroch and McNaughton, 2002). In this light, knowledge sharing can be seen as mediating the relationship between organizational factors and process innovation, hence it is posited that: H6. Knowledge sharing mediates the relationship between organizational structure and process innovation. Research Model The intersection of the resource-based view lays the theoretical grounding for this study. The resource-based view posits that firm competitiveness comes from unique bundles of tangible and intangible assets that are valuable, rare, imperfectly imitable, and sustainable (Barney,1991). Researchers such as Barney(1991), Peteraf (1993), andWernerffelt (1984) argue RBV views a firm as a bundle of resources and capabilities. Resource consist of tangible components like financial and physical assets like property; plant and equipment; and intangible components like human capital, patent, technology know-how (Grant,1996). Capabilities are "invisible assets", tangible or intangible organizational processes developed by a firm over a period of time that "cannot be easily bought; they must be built" (Teece, Pisano and Shuen, 1997, p.514).It has been widely accepted among scholars and practitioners that organizational context and processes have considerable influences on the SHS Web of Conferences 01007-p.4 organizational effectiveness of practices (Zheng, Yang and McLean;Gold, Malhotra and Segars, 2001).A vast number of studies and surveys indicate that there is a positive relationship between an efficient organizational context and effective knowledge sharing and the organizational competitiveness of implement KM. In this conceptual model, the organizational contexts (organizational structure and technological support) as independent variables; knowledge sharing as a mediating variable and process innovation as the dependent variable. Hung, Lien,Yang, Wu and Kuo(2011).Ten items to evaluate organizational structure, technological support and knowledge sharing were adapted from Gold et al.(2001).The results of the principal component factor analysis yielded eigen values greater than 1.0, which accounted for 58% and 65% of the variance for the independent and dependent variables respectively. Measurement models Results from the confirmatory factor analysis demonstrated that all of the scales used in the study formed adequate measurement models and thus provided evidences for the construct validity of the measures. Table 1 shows the fit indices of the measurement models whereas Table 2 shows the descriptive statistics of the constructs. Assessing Structural Model The goodness of fit indices show that the hypothesized model fit the data well .The path coefficients in Figure 2 and Table 2 was used in testing hypotheses 1 to 5, the indirect effects between the variables were tested in hypotheses 6 and 7. Holbert and Stephenson (2003) indicated specific indirect effects are not calculated by the major SEM software packages, so Sobel test for significant indirect effects are shown in Table 4. Hypotheses testing All of the hypotheses were examined through the investigation of the path coefficients and the statistical significance .Based on the results in Tables3 and4, Hypothesis 1 is supported. There is a significant path coefficient of.24 (p<.05) from organizational structure to knowledge sharing. Hypothesis 2 is also supported. There is a significant path coefficient of .35 (p<.01) from organization structure to process innovation. Hypothesis 3 is supported. There is a significant path coefficient of .37 (p<.05) from technological support to knowledge sharing. Further, Hypothesis 4 is supported. There is a significant path coefficient of .21 (p<.05) from technological support to process innovation. Hypothesis 5 is supported. There is a significant path coefficient of .22(p<.05) from knowledge sharing to process innovation. From the results in Table 4, knowledge sharing was found to have partial mediating effect between organizational structure and process innovation, and between technological support and process innovation. So, hypotheses 6 and 7are partially supported. DISCUSSION AND CONCLUSION The results of the study are in line with the knowledge-based view of the firm which emphasizes the importance of understanding the processes through which organizations access and utilize knowledge possessed by its individual members (Grant, 1996). Organizational structure can influence knowledge management processes through shaping patterns and frequencies of communication among organizational members, stipulating locations of decision-making and affecting efficiency and effectiveness in implementing new ideas (Zheng, Yang and McLean, 2010). It also corresponds with Penrose's (1959) opinion that the usefulness of organizational resources varies with changes in organizational knowledge and knowledge management serves as a key leverage point in organizations. One practical implication of this study is that the Iraqi textile factories should take cognizant and try to improve their knowledge sharing practices. This study bear theoretical implications as it provide some insights in integrating the resourcebased view when applied in the Iraqi textile industry. It reveals that whereas the resources (knowledge) in the textile factories may be hierarchical (organizational structure), knowledge sharing may bring the factories one step closer towards organizational effectiveness (process innovation) by mediating the resources to process innovation. Further exploration is needed to examine this proposition. The social implication of this study for the Iraqi government which owns the factories is to enhance knowledge sharing activities in the factories in order to promote process innovation in the Iraqi textile industry. One of the limitations of this study is that the study was done in one industry and in one middleeastern country, specifically in the Iraqi textile industries which limits generalizablity of the findings. To further test the organizational contexts of the model,future research are encouraged in the context of an Asian or developing countries or emerging economiesfrom which the measurement model was tested (Gold, Malhotra and Segars, 2001;Wong, 2005). The study can also be replicated in other industries as well as in public or private factories. In conclusion, based on the survey among employees in the Iraqi textile industry, the results revealed that organizational structure and technological support positively and significantly influence knowledge sharing. Knowledge sharing was also found to be an important mediator between organizational structure, technological support and process innovation. The findings bear implications to the Iraqi government's call for innovation and knowledge management practices in the Iraqi textile industry.	2018-10-13T08:10:12.871Z	2015-01-01T00:00:00.000	{ "year": 2015, "sha1": "6188ac8648bc1252c80ec7275fd1aea22776c3f8", "oa_license": "CCBY", "oa_url": "https://www.shs-conferences.org/articles/shsconf/pdf/2015/05/shsconf_icolass2014_01007.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "6188ac8648bc1252c80ec7275fd1aea22776c3f8", "s2fieldsofstudy": [ "Business" ], "extfieldsofstudy": [ "Business" ] }
119303699	pes2o/s2orc	v3-fos-license	Geodesic motion in R- charged black hole space-times We study the geodesic motion of massive and massless test particles in the background of a particular class of multiple charge black holes in gauged supergravity theories in D = 4. We have analysed the horizon structure along with the nature of the effective potentials for the case of four equal charges. In view of the corresponding effective potentials, we have discussed all the possible orbits in detail for different values of energy and angular momentum of the incoming test particles. The periods for one complete revolution of circular orbits and the advance of perihelion of the planetary orbit have also been investigated in greater detail for massive test particles. We have also discussed the time period of unstable circular motion and cone of avoidance of massless test particles in detail. All the corresponding results obtained for massive and massless test particles are then compared accordingly. I. INTRODUCTION: Over the last century, the Einstein's theory of gravitation i.e. General Relativity (GR)[1 -4] has been extremely successful to understand various observational facts like gravitational redshift, the precession of Mercury's orbit, the bending of light etc. On the other hand, various black hole spacetimes with or without rotation were obtained as exact solutions of Einstein field equations in GR namely Schwarzschild metric [5]. Other black hole spacetimes including charges and rotations have also been discovered. Though GR has enjoyed the great success, but it is still not a complete theory to understand the physics at sufficiently smaller length scale e.g. near the spacetime singularity which arises in case of gravitational collapse [4]. In the vicinity of the spacetime singularity, the quantum effects should be taken care of seriously and recently string theory has become the promising candidate for the same purpose which comprises gravity in the frame work of quantum theory [6][7][8]. In continuation pursuit of the search for a quantum theory of gravity, the gauged supergravity theories [9][10][11] have also captured considerable attention in recent times. In such models of gravity, the maximally supersymmetric gauged supergravity is realized as truncation of string theory or D = 11 dimensional M theory compactified on a sphere where the gauge group is the isometry group of the sphere. In particular, in AdS/CF T correspondance [12], this gauge group becomes the R-symmetry group of the boundary CFT which is related to the string theory on anti-de Sitter space. Black hole solutions in the AdS sector have drawn much attention to understand strongly coupled gauge theory at finite temperature on the CFT side [13], [14]. For the present work, we consider four dimensional charged black holes in N = 8 1 , gauged supergravity upto four charges. The construction of these black holes have been explicitly performed in Ref. [15] and according to [16], such charged black hole congurations are termed as R-charged black holes. In gauged N = 8 supergravity models, the bosonic part of the complete Lagrangian has a negative cosmological constant Λ proportional to the square of the gauge coupling constant 2 g [17] and the black hole solutions are asymptotically AdS. ory [36][37][38][39][40][41][42][43], we try to make an attempt to understand the geodesics of massive neutral as well as massless particles in the exterior region of such R-charged black holes. Such investigations may indeed useful to capture the effect of cosmological constant in addition to the charges on the motion of the test particles. The main objective of this paper is to study all the possible orbits of test particles in these class of spacetimes. We analyze the equivalent one dimensional effective potentials in order to study the geodesics equations of motion of the test particles in R-charged black holes background. On the other hand, we also focus our study on the various parameters (mass parameter, multiple charge parameters and gauge coupling constant) of the solutions and present various possible orbits by tuning those parameters. Due to the presence of four non zero charges, the form of the one dimensional effective potentials are little complicated in our case. We mainly concentrate on the numerical approach without performing the analytic solutions [22][23][24]. The form of potentials can be retraced back to the known form in the limit of zero charges which otherwise takes quite nontrivial form. This paper is organised as follows. In the next section, we discuss about the spacetime of our interest and the horizon structure of these particular class of R-charged black holes [15] with multiple charges, whereas the ungauged situation (i.e. g = 0) is same as the Schwarzschild black hole with flat asymptotes in GR. In section III, we derive the geodesics equation and obtain the form of the effective potential. In section IV, we study geodesics equation of motion for massive neutral particles (i.e. the timelike geodesics) under one dimensional effective potential by considering different values of the parameters involved in. The radial and non-radial geodesics are studied analytically and otherwise numerically. Using the effective potential techniques, the motion of test particles and the structure of corresponding orbits are discussed in greater detail. In order to have a complete analysis of the geodesic motion in the background of the abovementioned black hole spacetime, we also study the motion of massless test particles (i.e. null geodesics) in section V. Finally, we conclude and summarize our results. II. THE SPACETIME WITH MULTIPLE CHARGES We consider the following spacetime metric of four electric charge black hole solution in D = 4, N = 8 gauged supergravity, Here g is the gauge coupling and µ is the non-extremal parameter like mass term in pure The scalar curvature R for the above metric can be given as below, One can easily verify that for H α = 1 (no charge solution), the scalar curvature R = −8g 2 , which is exactly same as that of an Anti-de Sitter Schwarzschild black hole. Let us first discuss about the event horizons of the metric eq.(2.1) in presence of n equal charges. So for n equal charges with k = 1, equation (2.2) reduces to, certain values of β, one will get two zeros of the function f (r) whereas smaller than that two zeros merge into a single zero and larger than that real zeros will disappear completely by leaving with a naked singularity as explained in [44]. The same arguments are also plausible when the constant g varies keeping charge parameters β fixed, as shown in fig(1h). III. THE GEODESICS EQUATIONS AND EFFECTIVE POTENTIAL One can simplify metric given in eq.(2.1) by choosing θ = π 2 , to restrict the motion of the particles in the equatorial plane i.e. we restrict ourselves to study of the equatorial geodesics. Using the following geodesic equation we obtainẗ φ where dot denotes the derivative with respect to proper time τ . For a timelike geodesics (u µ u µ = −1), From the constant of motion, the first integrals of eqs. (3.2) and (3.4) result as, where E and L are the integration constants which represent the conserved energy and angular momentum of the test particles respectively. Using eqs.(3.6) and (3.7), one can now rewrite the timelike constraint eq.(3.5) in the following The effective potential for timelike geodesics may be identified from eq.(3.8) as, One can reproduce the effective potential for the schwarzschild black hole with AdS asymptote as a limiting case when H α = 1. It seems that under scaling of the metric components like f → f H − 1 2 and r 2 → r 2 H 1 2 , one can exactly write down all the corresponding equations above for the case of a schwarzschild black hole in flat space [3]. However, the effective potential for null geodesics is given as, The difference in the nature of effective potentials for both the cases as given in eq.(3.9) and eq.(3.10) manifests itself in the structure of orbits as presented in the later sections. BITS FOR TIMELIKE GEODESICS From effective potential given in eq.(3.9) for particles moving along the timelike geodesics, the equation of motion (3.8) is investigated for two cases: radial and non-radial geodesics in the following subsections, respectively. A. Effective potential for radial geodesics (L = 0) For radial geodesic (with zero angular momentum L), the effective potential given in eq. (3.9) reduces to the following form, The behaviour of effective potential is shown in fig.(2) for a particular set of parameters, considering all values of n ranging from 0 to 4. If the particle is released from rest at a distance r = r i , the initial energy of the particle is given as, So from the radial equation of motion, one obtainṡ which can be integrated as, . (4.4) In order to know the proper time experienced by a particle falling from r i to coordinate radius r, one needs to evaluate the integral given in eq. Here we study the timelike geodesics for incoming test particles with nonzero angular momentum. Let us try to understand orbits of the particles in the background of the R-charged black holes by using the effective potential V ef f given by eq.(3.9). For r → r H (the horizon ra- . These are the asymptotic behavior of the potential. In fig.(4) (ii) For the other possible orbit at this energy value, the test particle starts from the point B and falls into the singularity. Hence it is a terminating bound orbit. II) E = E 1 : (i) A bounded planetary orbit between points A and P (as shown in fig (6)). (ii) The other possible orbit corresponds to point D which falls into the singularity (i.e. a terminating bound orbit). The possible orbits are unstable circular orbit at point F . The particle starts from point F and then it can go either to point G or to the singularity after crossing the horizon. There exists terminating bound orbit for particle crossing point J. Hence there are no fly-by orbits possible. C. Analysis of the geodesics: Orbit equation can be obtained by using eq.(3.7) and eq.(3.8) as, uniquely determine the type of particles orbit in the background of these charged black holes. For n = 4, the polynomial P n (r) reduces to, where the mass parameter is fixed as µ = 1. Note that for any value of the parameters β, g, L, E, the function P 4 (r) → −∞ for r → ±∞. Also, for r = 0, P 4 (r) = 2g 2 p 4 L 2 ( E 2 2g 2 − L 2 − p 2 ) indicates that P 4 (r) becomes positive in the domain of the parametric space where E 2 2g 2 > L 2 + p 2 . We restrict the values of the parameters in such a way that the previous bound satisfied throughout hereafter. (I) For E = E C : As discussed above that for energy value E = E C , two type of orbits are possible, one is a stable circular orbit and another one is the terminating bound orbit. In fact the number of zeros of the function P 4 (r) uniquely characterizes the orbit structure as mentioned in [24]. Solid line represents the orbit (terminating bound orbit) for test particle starting from a distance at r = r B in the allowed region, dotted line represents the event horizon at r = r H ; with L = 10, β=1, g = 0.02, E 2 = 1.3632. Circular Orbits: In case of circular orbit, r = r C = constant, where r C is the distance of the circular orbit from the singularity and henceṙ = 0. In order to calculate the time periods for circular orbits by using eqs.(3.6) and (3.7), we have The circular orbit condition V ′ ef f = 0 can be used to express L in terms of r C as, where X = µnp r C H(r C ) 1 − H(r C ) + p r C , p = µ sinh 2 (β), H = 1 + p r C and H = H n . As radial velocityṙ vanishes for circular orbit, it provides another condition for this orbit as E 2 = V ef f (r C ). Substituting this in eq.(4.8) with ∆t ≡ T t and ∆φ = 2π for one period then we obtain, where T t,sch = 2π Similarly time period in proper time can also be obtained using eqs.(3.7) and (4.8) as, Advance of perihelion for planetary orbits: We use the elementary derivation of the advance of perihelion of a planetary orbit for the Schwarzschild solution as in [45]. The advance of the perihelion for our case is obtained by comparing a Keplerian ellipse in a Lorentzian coordinates with the spacetime of interest given in eq.(2.1). The relevant relation communicating the two ellipse is the areal constant of Keplers second law. The following transformation of the coordinates, r and t, in the binomial approximation can be obtained for the spacetime used in present study, given by eq.(2.1), dr. Similarly in given spacetime one has, where r ′ is given in eq.(4.13) and the radial function R = r(1 + p r ) n 4 . The binomial approximation of the radial function R is The polar form of an ellipse is given by (4.20) where e is the eccentricity and l is the semi-latus rectum. Therefore applying the binomial approximation alongwith the integration (4.19) and using the eq. (4.20), we obtain the perihelion shift as At this energy value of the incoming particles, the two zeros of the polynomial P 4 (r) in the previous plots ( fig.(8)) merge where P 4 (r) = 0 and also d dr P 4 (r) = 0 at r = r F . The representative plots have been shown in figs.(9a) and (9c). The radius r F is the radius of the circular orbit which is unstable. In fig.(9b), a polar plot of the timelike geodesic is given for particle arriving from a finite radius r F and plunges in to the black hole singularity. fig.(9d) another possible orbit has shown where the particles coming from r G and having an unstable circular orbit at r = r F . (IV) For E = E 3 : At this energy value E = E 3 , the degenerate root at r = r F will disappear leaving only a single root at r = r J as shown in fig.(10b). Again we have numerically simulated the path of the particles carrying energy E = E 3 is shown in fig.(10c) represents a terminating bound orbit. In order to have a detailed view of the geodesic motion in the background of R-charged Black Holes, the Null geodesics are discussed in the next section. A. Radial Geodesics Since the radial geodesics are the trajectories followed by zero angular momentum test particles(i.e. L=0) in given spacetime geometry, the effective potential given by eq. where τ 0 is the integration constant corresponding to the initial position of the test particle. Eq.(5.2) shows that in terms of proper time, the radial coordinate depends only on the constant energy value E. Using eq.(3.6) and eq.(5.1), the radial equation of motion can also be obtained in terms of coordinate time t (for n = 4) as, where f is given in eq. In this section, we study the null geodesics for incoming test particles with non zero angular momentum. Variation in the effective potential with charge parameter β is shown in fig.(12) while all other parameters are fixed. One can observe from fig.(12), that height of the effective potential decreases with increasing the value of charges for a particular value of n. The following orbits are allowed depending on the values of the constant E (i.e. energy of FIG. 13: Effective potential of a unit mass black hole (i. e. µ = 1) for a massless test particle in the presence of four equal charges with L = 10, β=1, g = 0.02. the incoming test particle) as shown in fig.(13). I) E = E 0 : A particle starting form point M drops into the singularity, forming a terminating bound orbit. II) E = E 1 : (i) A fly-by orbit is present for the particle coming from infinity, which turns at point P as shown in fig.(13). (ii) For the other possible orbit at this energy value, the test particle starts from the point Q and falls into the singularity. Hence it is a terminating bound orbit. III) E = E 2 : A particle starting from infinity follows an unstable circular orbit at point R which finally drops into the singularity as depicted in fig (13). Hence starting from point R, it follows a terminating bound orbit. IV) E = E 3 : A particle starting from the infinity finally drops into the singularity after crossing the horizon. Hence it is a terminating escape orbit. C. Orbit Analysis for null geodesics: Using the constraint for null geodesics (i.e. u µ u µ =0), the corresponding orbit equation reads as, dr dφ There exists an interesting type of geodesics for massless test particles with definite angular momentum, specifically known as Cardioid type Geodesics [46]. From eq.(5.4), it follows that with specific energy value of E 2 = 2g 2 L 2 , there also exist cardioid type geodesics for R-charged black holes. For this special value of energy, the eq.(5.4) reduces to, An elemental integration of eq.(5.7) leads to, (1 + cos φ). (5.8) which is the equation of the cardioid. This solution is independent of the angular momentum of incoming test particle and depends only on the mass of black hole. (II) For E = E 1 : The numerical solution of orbit equation (5.4) is shown in fig.(14). Here one may notice FIG. 14: (a) P 4 null (r) represents the shaded allowed region for motion of a massless test particle with two real positive zeros at point P and Q, (b) Solid line represents the orbit of a massless test particle in region II, (c) Solid line represents the orbit of a massless test particle in region I while dotted line represents the event horizon; with L = 10, β=1, g = 0.02, E 2 = 1, n = 4. the presence of a fly-by orbit with turning point at P and a terminating bound orbit starting from point Q as depicted in fig.(13). (III) For E = E 2 : The trajectory of a massless test particle with energy E = E 2 is presented numerically in fig.(15). The particle starting from infinity forms an unstable circular orbit at point R, which finally terminates into the singularity. The circular orbit condition (i.e. V ′ ef f (r) = 0) for null geodesics is reduces to, where V ef f (r) is given in eq.(3.10) with n =4, p = µ sinh 2 β. Solution of eq.(5.9) describes the radius of unstable circular orbit and is given as, where p = µ sinh 2 β and eq.(5.9) has a real solution for (2p + 3µ) 2 ≥ 8µp. The larger root of eq.(5.9) locates the position of unstable circular orbit, which again has the minimum value for the condition (2p + 3µ) 2 = 8µp as, (r c ) min = 1 4 (2p + 3µ) . integrating afterwards as follows, Cone of Avoidance: The cone of avoidance can be defined at any point [41,46,47] and the light rays included in the cone must necessarily cross the horizon and get trapped. If ψ denotes the half-angle of the cone directed towards the black hole at large distances, then here dr = (r + p) rf 1/2 dr, (5.15) where f is given by eq.(2.4). The radial function R described in eq.(4.16) and eq. f or r >> 1, (5.18) where, r c denotes the radius of circular orbit and r H denotes the radius of event horizon. (IV) For E = E 3 : The path of a massless test particle with energy E = E 3 is simulated numerically in fig.(16) FIG. 16: (a) P 4 null (r) represents the shaded allowed region for motion of massless test particle no real positive zero, (b) Solid line represents the orbit of massless test particle in the allowed region while dotted line represents the event horizon; with L = 10, β=1, g = 0.02, E 2 = 2, n = 4. which shows the presence of a terminating escape orbit. VI. SUMMARY AND CONCLUSIONS In this article, we have investigated geodesic motion of massive as well as massless test particles around a particular class of R-charged black holes in N = 8, D = 4 gauged supergravity theory. Some of the important results are summarised below: • It is observed that in R-charged black hole spacetimes, the horizon is always present for single and two charges, while it occurs only for a particular condition in case of charges more than two. • Different types of orbits (such as terminating bound, planetary, stable and unstable circular orbits) are present for incoming massive test particles, corresponding to their initial energy. No fly-by orbits are observed in any case for massive test particles. Fly-by and terminating escape orbits are however present for the case of massless test particles unlike the case of massive test particles. There exist no stable circular orbits for a massless test particle. An interesting type of orbit for a massless test particle namely, Cardioid type geodesic is found to depend only on the mass of the black hole. • Radius of stable circular orbits for massive test particles and radius of unstable circular orbits for massless test particles are found to increase with the enhancement in the number of charges. We have also computed the time periods both in coordinate and proper time to visualise the changes with the number of charges. The radius of unstable circular orbit for a massless test particle is found to be independent of the gauge coupling constant. • Advances of perihelion in planetary orbits is calculated for massive test particles by using the Keplerian orbit method and the corrections arising from non zero charge parameters apart from the gauge coupling constant are obtained. • Cone of avoidance for a massless test particle is calculated and it is observed that at large distances, it depends on both the charge and gauge coupling constant. All the results obtained in the present study reproduces the results manifestly corresponding to well known spacetimes such as Schwarzschild and Schwarzschild AdS black hole spacetimes in the prescribed limits respectively.	2015-09-17T03:13:24.000Z	2015-09-16T00:00:00.000	{ "year": 2015, "sha1": "51b85e8307c6fd8f561084d85771a997444c6b34", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1509.05110", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "51b85e8307c6fd8f561084d85771a997444c6b34", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
42045417	pes2o/s2orc	v3-fos-license	Implicitization of Hypersurfaces We present new, practical algorithms for the hypersurface implicitization problem: namely, given a parametric description (in terms of polynomials or rational functions) of the hypersurface, find its implicit equation. Two of them are for polynomial parametrizations: one algorithm,"ElimTH", has as main step the computation of an elimination ideal via a \textit{truncated, homogeneous} Gr\"obner basis. The other algorithm,"Direct", computes the implicitization directly using an approach inspired by the generalized Buchberger-M\"oller algorithm. Either may be used inside the third algorithm,"RatPar", to deal with parametrizations by rational functions. Finally we show how these algorithms can be used in a modular approach, algorithm"ModImplicit", for avoiding the high costs of arithmetic with rational numbers. We exhibit experimental timings to show the practical efficiency of our new algorithms. Introduction Let K be a field and let P = K[x 1 , . . . , x n ] be a polynomial ring in n indeterminates. Then let f 1 , . .., f n be elements in the field L = K(t 1 , . . . , t s ), where {t 1 , . . . , t s } is another set of indeterminates which are viewed as parameters. We consider the K-algebra homomorphism ϕ : K[x 1 , . .., x n ] −→ K(t 1 , . . . , t s ) given by x i → f i for i = 1, . . . , n Its kernel, which will be denoted by Implicit(f 1 , . . . , f n ), is a prime ideal, and the general problem of implicitization is to find a set of generators for this ideal. The task of computing Implicit(f 1 , . . . , f n ) can be solved by computing a suitable Gröbner basis (see Proposition 3.1). However, in practice this method does not work well since in most non-trivial cases it is far too slow. The poor computational speed is aggravated when computing with rational coefficients (rather than coefficients from a finite field). There is definitely a big need for new, efficient techniques, and many authors have investigated alternative ways. The literature about implicitization is so vast that it is almost impossible to mention the entire body of research on this topic. This interest derives from the fact that the parametric representation of a rational variety is important for generating points on it, while the implicit representation is used to check whether a point lies on it. Besides its theoretical importance, the double representation of a rational variety is used intensively for instance in Computer Aided Geometric Design. A good source of bibliography up to ten years ago is [10]. More recently, new ideas have emerged. As we said, it is almost impossible to cite all of them, and we content ourselves to mention a few. In particular, new methods for computing implicitizations have been described in [12], [13], [14], and [19]. Some of these new ideas respond to the fact that in many cases the computation of Implicit(f 1 , . . . , f n ) is too hard, hence one seeks a way to check whether a point lies on the rational variety without actually computing its equations. So, what is the content of this paper? And what are the novelties and the new algorithms presented here? First of all, we concentrate on the "hypersurface case" where Implicit(f 1 , . . . , f n ) is a principal ideal, and hence generated by an irreducible polynomial which is therefore unique up to an invertible constant factor. Remark 1. 1. Let I be the ideal Implicit(f 1 , . . . , f n ) and let m be the minimum number of generators for I. From the facts that dim(K[f 1 , . . . , f n ]) ≤ dim(K[t 1 , . . . , t s ]) = s and dim(K[f 1 , . . . , f n ]) = dim(K[x 1 , . . . , x n ]/I) ≥ n − m it follows that m ≥ n − s. So, whenever s ≤ n − 1 then I has at least one generator, and in particular I is non-zero. The hypersurface case typically arises when s = n − 1, in accordance with the remark above. However, this is not always the case, as the following examples show. Example 1. 2. We consider the "atypical" case where n = s = 2 and f 1 = f 2 = t 1 +t 2 . Clearly we have Implicit(f 1 , f 2 ) = x − y , which is obviously principal. This does, however, become a typical case if we use a "better parametrization" in terms of u = t 1 +t 2 , where we then have f 1 = f 2 = u, and consequently also have s = n − 1 with this better parametrization. Another "atypical" example is the following. Let n = 3, s = 2 and f 1 = . Here we do have s = n − 1 but the implicitization is not principal, in fact it turns out that Implicit(f 1 , f 2 , f 3 The reason here is that there is a "better parametrization" in terms of u = t2 t1 , where we have f 1 = u 2 , f 2 = u 2 + 1, f 3 = u 2 + u + 1; and with this better parametrization we have s = n − 1. In this paper we do not examine the interesting question of finding a good parametrization, which is a problem of a quite different nature. The ideas explored in this paper can be summarized in the following way: • Exploit homogenization to improve elimination (RatPar, ElimTH): using elimination is known to be an elegant but impractical way to achieve implicitization. We show that any problem (polynomial or rational parametrization) can be homogenized (see Proposition 3.3 and Theorem 5.3). Thereafter, the result is given by the first polynomial not involving any of the parameters, so the computation can be stopped as soon as it is found, avoiding the remaining "useless reductions". • Use a direct algorithm which does not need elimination (Direct) : Wang in [21] described an algorithm based on searching for a linear relationship among the images of the power-products. We refine this idea and make it incremental, thus leading to several important insights and opimizations (see Subsection 4.2). • When the coefficient field is Q use modular methods (ModImplicit): computing the solution polynomial modulo several primes, and then obtaining the solution over Q by Chinese Remaindering is a powerful tool, but needs to be finetuned to any specific context. In Section 6 we use an incremental approach combined with fault-tolerant rational reconstruction to resolve the problem of how many primes are needed and to "tolerate" computations with bad primes (some of which cannot be detected a priori). Remark 1. 3. Regarding the first item, one of the referees pointed out that we made no reference to the "projective view of the implicitization problem, which is relatively classical". The main reason was to avoid complications for the typical computer-algebra people who, generally, are much more familiar with algebra than geometry. Nevertheless, let us give some hints in this direction to the interested reader. If f 1 , . . . , f n ∈ K[t 1 , . . . , t s ] then the map ϕ can be seen as the algebraic counterpart of the map of the affine schemes Φ : A s −→ A n . We let d i = deg(f i ) for i = 1, . . . , n, then homogenize the f i with a new indeterminate h such that deg(h) = 1, and set deg(x i ) = deg(f i ). Now we consider the projective space P s with coordinates t 1 , . . . , t s , h and the weighted projective space P(d 1 , . . . , d n , 1) with coordinates x 1 , . . . , x n , h (see for instance [7] for an introduction to the theory of weighted projective spaces). The map Φ can be viewed as the restriction to A s of the rational map Ψ : P s P(d 1 , . . . , d n , 1) given by [t 1 : Observe that Ψ is a rational map, but not necessarily a map, since it may have a non-trivial base locus. The algebraic explanation of this fact is exactly the proof of Proposition 3. 3. The situation is more complicated when f 1 , . . . , f n ∈ K(t 1 , . . . , t s ). Using a common denominator, we may assume that f i = pi q with f i , q ∈ K[t 1 , . . . , t s ] for i = 1, . . . , n. If we let D q denote the open subscheme A s \ {q = 0}, then ϕ can be seen as the algebraic counterpart of the map of affine schemes Φ : D q −→ A n , and the standard way to proceed is to take care of this limitation, as explained in Remark 5. 1. But there is a different way to interpret Implicit(f 1 , . . . , f s ). We let d i = deg(p i ) for i = 1, . . . , n and d 0 = deg(q). Then we let d = max{deg(q), deg(p 1 ), . . . . , n (see Definition 2.5), so that all the polynomials Q, P 1 , . . . , P s are homogeneous of the same degree d. Next we consider the projective space P s with coordinates t 1 , . . . , t s , h and the projective space P n with coordinates x 0 , x 1 , . . . , x n . If we let A n be the affine open chart of P n defined by x 0 = 0, the map Φ can be interpreted as the restriction to D q of the corresponding rational map Ψ : P s P n defined by [t 1 : t 2 : · · · : t s : h] → [Q : P 1 : · · · : P s ]. The algebraic explanation of this fact is exactly the proof of Theorem 5. 3. Why not try to use other embeddings into suitable projective or weighted projective spaces, as we do in the case of polynomial parametrizations? The reason is explained in all the remarks and examples following Theorem 5. 3. There is a further idea: computing implicitizations with constraints, in particular using a method of "slicing" the variety with suitable parallel hyperplanes. This technique was introduced and used in [20]. However, it is rarely better than our new methods when Implicit(f 1 , . . . , f n ) is a hypersurface. We shall investigate the "implicitizations with constraints" for the general case in a later paper. The algorithms described in this paper are implemented in CoCoALib [2], and are also available in CoCoA 5 [3]. With our new methods most of the examples mentioned in the literature become "easy", that is we can compute the implicitization in less than a second -see Table 1 in Section 7. Consequently, we introduce new, challenging examples, and the last table shows the performance of our implementation. We thank the referees for their useful comments and suggestions which helped us to improve this paper. Notation and Terminology Here we introduce the notation and terminology we shall use. Definition 2.1. Let K be a field, and let P = K[x 1 , . . . , x n ] be a polynomial ring in n indeterminates. Let t 1 , . . . , t s be further indeterminates which are viewed as "parameters". Given elements f 1 , . .., f n in K[t 1 , . . . , t s ], we define the ideal J = x 1 − f 1 , . . . , x n − f n of the ring P [t 1 , . . . , t s ] to be the eliminating ideal of the n-tuple (f 1 , . . . , f n ). We define Implicit(f 1 , . . . , f n ) to be the kernel of the K-algebra homomorphism ϕ : Definition 2.2. We extend naturally Definition 2.1 to parametrizations by rational functions. Let p1 q , . . . , pn q be rational functions in the field L = K(t 1 , . . . , t s ) with common denominator q; so that we have q, p 1 , . . . , p n ∈ K[t 1 , . . . , t s ]. We define Implicit( p1 q , . . . , pn q ) to be the kernel of the K-algebra homomorphism ϕ : P −→ L which sends x i → pi q for i = 1, . . . , n. Definition 2. 3. An enumerative ordering is a total ordering such that for every element there are only finitely many elements smaller than it. In particular, an enumerative term-ordering is a term-ordering which is also enumerative; consequently, an enumerative term-ordering is defined by a matrix with strictly positive entries in the first row. Example 2. 4. Any degree-compatible term-ordering is enumerative because for any power-productT all smaller power-products, T <T , must have deg(T ) ≤ deg(T ), and so they are finite in number. In contrast, the lex-ordering (for 2 or more indeterminates) is not enumerative because if indeterminate x 2 is less than x 1 then all powers x d 2 are smaller than x 1 . Definition 2. 5. In section 5 we shall use two different kinds of homogenization: • traditional homogenization and dehomogenization: with respect to h we denote them by the superscripts hom(h) and deh(h) respectively; and with respect to x 0 , by the superscripts hom(x0) and deh(x0) . • d-shifted-homogenization: for a non-zero polynomial f and degree , which is a homogeneous polynomial of degree d. As a special case, since 0 hom(h) = 0, we have 0 hom d (h) = 0 for all d. The following easy properties of the shifted-homogenization will help the reader understand the proof of Theorem 5.3 Lemma 2. 6. Let P be a polynomial ring over the field K, and let f, g ∈ P . Proof. The proofs are elementary exercises in algebra. Observe that the special definition 0 hom d (h) = 0 is indeed compatible with this lemma, since from the equality f −f = 0 we deduce the equalities Polynomial Parametrizations In this section we consider a parametrization given by polynomials f 1 , . . . , f n in the ring K[t 1 , . . . , t s ], where {t 1 , . . . , t s } is a set of indeterminates which are viewed as parameters. We will look at parametrizations by rational functions in section 5. Proof. Claims (a) and (b) are standard results (see for instance book [17], Section 3.4). Claim (c) follows from the isomorphism K[t 1 , . . . , t s , x 1 , . . . , x n ]/J ∼ = K[t 1 , . . . t s ], whence J is prime, and so J ∩ P is prime too. ✷ Remark 3.2. We shall later find it convenient to assume that in the parametrization no x i maps to a constant. This is not a restriction because if, say, f n ∈ K then we obtain the simple decomposition: Implicit(f 1 , f 2 , . . . , f n ) = x n − f n + Implicit(f 1 , . . . , f n−1 ). Thus any indeterminates x i which map to constants can simply be taken out of consideration, letting us concentrate on the interesting part. Henceforth we shall assume that none of the f i is constant. The very construction of the eliminating ideal (in Definition 2.1) looks intrinsically non-homogeneous. And it is well-known that the behaviour of Buchberger's algorithm can be quite erratic when the input is not homogeneous: usually the computation for a nonhomogeneous input is a lot slower than a "similar" homogeneous computation (though there are sporadic exceptions); for instance, see Example 4. 3. We now look quickly at how to use homogenization during implicitization. A first idea is to give weights to the x i indeterminates by setting deg(x i ) = deg(f i ) for each i. If we do so, and if the original f i are homogeneous polynomials, then the eliminating ideal J turns out to be a homogeneous ideal. Even when the f i are not all homogeneous, in the process of ordering and choosing the power-products of a given degree, we may reasonably expect that Buchberger's algorithm will "behave" more similarly to a homogeneous ideal than with the standard grading, where all indeterminates have degree 1. Although this trick improves the computation in most cases, it is not a miraculous panacea. Much better ideas come from the following Proposition 3.3 and Theorem 5.3 which reduce the computation of the implicitization ideal to the case of prime ideals whose generators are homogeneous polynomials. In the proofs we use the fundamental properties of homogenization and dehomogenization as described in [18], Section 4. 3. A general discussion about the topic treated in the following proposition can be found in [18], Tutorial 51. In the proposition below we use a single homogenizing indeterminate h; so, to simplify notation, homogenization and dehomogenization are tacitly taken with respect to h. , and letJ be the eliminating ideal of (F 1 , . . . , F n ). Then: Proof. The proof of claim (a) follows immediately from the fact thatJ is an eliminating ideal, hence prime. Let J be the (non-homogeneous) eliminating ideal of the tuple (f 1 , . . . , f n ). Now, sinceJ is prime, it is saturated with respect to h; furthermore we haveJ deh = J, so we Let σ be an elimination ordering for {t 1 , . . . , t s } on R. . . , x n − F n , the eliminating ideal of (F 1 , . . . , F n ). ElimTH-2 Main Loop: Start Buchberger's algorithm for the computation of a σ-Gröbner basis GB of J. Perform its main loop degree by degree (i.e. always choose the lowest degree pair). When you add to GB the first polynomial G such that LT σ (G) is not divisible by any t i exit from loop. Hypersurfaces Parametrized by Polynomials In this section we start to treat the "hypersurface case", namely the case where it is known that the implicitization ideal is principal. In this situation we typically have s = n − 1, although this equality is not equivalent to the implicitization being a principal ideal, as shown in Example 1. 2. There is no easy way to determine whether the implicitization is going to be a principal ideal, but this information might already be independently known for the particular example under consideration. So this is usually taken as hypothesis by the papers on this topic. A Truncated Homogeneous Computation As already observed, the idealJ in Proposition 3.3 is homogeneous, hence the computation of the (elimination) Gröbner basis ofJ can be performed degree by degree. Moreover, using the methods described in Proposition 3.3 we get the following extra bonus in the hypersurface case: as soon as we obtain a Gröbner basis element, G, which does not involve the parameters, we may stop the computation of the Gröbner basis because the solution polynomial is just the dehomogenization of G. Proof. We recall the equality Implicit(f 1 , . . 3.a, hence it is saturated with respect to h, and so it is generated by g hom . ✷ Proof. Termination: The Main Loop in the algorithm is simply Buchberger's algorithm, and that terminates in a finite number of steps. Moreover, Corollary 4.1 guarantees that J contains a polynomial not involving the t i indeterminates, and since σ is an elimination ordering for the t i there is such a polynomial in the Gröbner basis, so the Main Loop will set G and exit. Correctness: In the Main Loop we execute Buchberger's Algorithm with respect to an elimination ordering for all the t i ; thus the elements of the Gröbner basis whose leading terms are not divisible by any t i form a Gröbner basis for the elimination idealJ ∩ P [h]. By Corollary 4.1 this ideal is principal, so Buchberger's Algorithm (computing degree by degree) will produce exactly one polynomial whose leading term is not divisible by any t i . The Main Loop stops as soon as this polynomial is found. In step ElimTH-3 the polynomial G will be the generator of Implicit(f hom 1 , . . . , f hom n ), and by Corollary 4.1 we We consider briefly what happens if the input to Algorithm 4.1 (ElimTH) does not correspond to a principal implicitization ideal. If Implicit(f 1 , . . . , f n ) is the zero ideal then Buchberger's Algorithm in step ElimTH-2 will terminate without finding any candidate for G; we could in that case simply set G = 0. By Remark 1.1 this cannot happen if s ≤ n − 1. If, instead, Implicit(f 1 , . . . , f n ) is non-zero and non-principal then the polynomial G found in step ElimTH-2 will be a lowest weighted-degree element of a Gröbner basis for that ideal (and consequently a lowest weighted-degree non-zero element of the ideal). This next example illustrates the good behaviour of the algorithm above. The usual elimination of t takes more than one hour, even if we give the weights 15 and 23 to the indeterminates x 1 , x 2 respectively; whereas the truncated homogeneous elimination takes less than a second (this is one of our test cases: see Example 8.14). The solution polynomial has 176 power-products in its support. Remark 4. 4. If the eliminating ideal is not homogeneous, the idea of truncating the computation as soon as a polynomial in P is found does not work well, since it may happen that the first such polynomial computed by the algorithm is a proper multiple of the solution polynomial. The phenomenon is similar to the case where the reduced Gröbner basis of an ideal is {1}, yet before discovering that 1 is in the basis it often happens that many other (non-reduced) Gröbner basis elements are computed. One could take the polynomial found and factorize it, then substitute into the various irreducible factors to see which factor is the good one. But this is unlikely to be efficient. A Direct Approach We briefly recall the setting of this paper: we have been given a K-algebra homomorphism ϕ: K[x 1 , . . . , x n ]−→K[t 1 , . . . , t s ] sending x i → f i and we assume that its kernel is a principal ideal: the problem is to find the generator of ker(ϕ) = Implicit(f 1 , . . . , f n ) = g . Direct-1 Initialization: Following Remark 3.2, we shall find it convenient to assume that each f i ∈ K[t 1 , . . . , t s ] is non-constant. In this section we compute the polynomial g via a direct approach. We use the notation LPP(f ) to indicate the leading power-product of the polynomial f (also denoted in the literature by The idea behind our direct approach is to directly determine g by searching for a linear dependency among all the ϕ(T ): we generate, one by one, the polynomials ϕ(T ) as T runs through the power-products in K[x 1 , . . . , x n ] until a dependency exists. We shall now see how to reduce this apparently infinite problem to a finite, tractable one. Proof. Termination: The main loop of the algorithm considers the power-products in the ring K[x 1 , . . . , x n ] in increasing σ-order until the condition in step Direct-2.3 breaks out; since σ is enumerative, every power-product will be considered at some (finite) time. The initial values for QB and PPL, and the updates to these two variables in step Direct-2.3 (no) guarantee that whenever we enter step Direct-2.1 the set PPL satisfies PPL = {x i T : 1 ≤ i ≤ n and T ∈ QB} \ QB; in other words it comprises those powerproducts outside QB and which border on QB. As σ is a term-ordering, PPL therefore always contains the σ-smallest power-product outside QB (as well as many others). In step Direct-2.3 the algorithm looks for a K-linear dependency amongst the polynomials {ϕ(T ) \|T ≤ σ T }. Every such linear dependency corresponds to a monic element of ker(ϕ). By hypothesis ker(ϕ) contains the polynomial g (which we may assume to be monic wrt. σ), so if we reach step Direct-2.3 with T = LPP(g) then a linear dependency will surely be found (e.g. corresponding to the coefficients of g). Since σ is an enumerative ordering, there are only finitely many power-products less than LPP(g); so we will break out of the main loop when T = LPP(g), if not earlier. Correctness: We shall show that we do not break out of the main loop until T = LPP(g), and that when we do break out, the polynomial we construct in step Direct-3 is g. The test in step Direct-2. 3 gives true if and only if there is a polynomialg, of the form T − i a i T i with each T i < σ T , satisfying ϕ(g) = 0 or equivalentlyg ∈ ker(ϕ). Note thatg is monic, thus non-zero by construction. By hypothesis ker(ϕ) is a principal ideal (generated by g). So every non-zero element of ker(ϕ) has leading term σ-greater-than-or-equal to LPP(g), thus step Direct-2.3 will not find any linear dependency if T < σ LPP(g). If Implicit(f 1 , . . . , f n ) is the zero ideal then the Main Loop never exits (as no nontrivial linear dependency exists). However, if s ≤ n − 1 then the ideal Implicit(f 1 , . . . , f n ) cannot be the zero ideal as proved in Remark 1. 1. If, instead, Implicit(f 1 , . . . , f n ) is non-zero and non-principal then the main loop will exit, and the polynomial g calc found in step Direct-3 will be the monic polynomial with σ-smallest leading term in the reduced σ-Gröbner basis for that ideal. Remark 4. 7. This approach is inspired by the Generalized Buchberger-Möller algorithm [6], and is somewhat simpler (e.g. the list G for storing the Gröbner basis is not needed, and the update to the list PPL is simpler). But there is an important difference: here we cannot specify a priori a finite dimensional vector space as the codomain of the normal form vector map. For the generalized Buchberger-Möller algorithm the finiteness of the codomain led to an easy proof of termination; instead here we had to introduce the concept of enumerative ordering. suggest exploiting the fact that ϕ is a homomorphism to compute the value cheaply. Apart from the very first iteration when T = 1, we always have T = x j T ′ for some indeterminate x j and some power-product T ′ for which we have already computed ϕ(T ′ ); so we can calculate with just a single multiplication ϕ(T ) = ϕ(x j ) ϕ(T ′ ). Usually there are several choices for the indeterminate x j , so we can choose the one which leads to the cheapest multiplication. Note that in step Direct-2.3(no) we manipulate just power-products when updating PPL. (c) In step Direct-1.1 we pick some enumerative ordering on the power-products of the ring K[x 1 , . . . , x n ]. Here we describe a specific good choice; the idea is that as we pick (in step Direct-2.1) the power-products T in increasing order then the corresponding LPP(ϕ(T )) are in non-decreasing order. We start with a (standard) degree-compatible term-ordering τ on the powerproducts of K[t 1 , . . . , t s ]. Let M τ be an s × s integer matrix representing it (so all entries in the first row are 1). We define the order vector of a power-product t e1 1 t e2 2 · · · t es s to be M τ e; the ordering τ is thus equivalent to lex comparison of the order vectors. Let E be the s × n integer matrix whose columns are the exponents of LPP τ (f i ); put M = M τ · E, an s × n matrix whose i-th column is the order vector of LPP τ (ϕ(x i )). The first row of M is strictly positive: the i-th entry is deg(f i ). We complete M to a term-ordering matrix M ′ for the power-products of K[x 1 , . . . , x n ]: i.e. we remove rows linearly dependent on those above it, and adjoin new rows at the bottom to make M ′ square and invertible. The term-ordering defined by M ′ is enumerative since M and M ′ have the same first row, and it has our desired property. Remark 4. 10. We contrast Algorithm 4.2 (Direct) with the method presented by Wang in [21]. The underlying idea is the same: find the generator by searching for a linear relationship among the images of power-products. Wang's method adjoins new powerproducts in blocks. Each block comprises all power-products of a given standard degree (where each indeterminate has degree 1). Wang observed that the linear systems produced "tend to be almost triangular". Our approach adjoins new power-products one at a time, and this lets us use several important optimizations (described in Remark 4.8). On each iteration we do a single multiplication to obtain ϕ(T ), and then a single "row reduction". Using the term-ordering described in Remark 4.8(c) guarantees that our linear system is as triangular as possible, and by adjoining power-products one by one we keep the system small (and avoid computing extraneous images of power-products under ϕ). The importance of these optimizations is illustrated by the computation time for Example 8.7: our implementation of Algorithm 4.2 (Direct) took less than 8 seconds, while Wang reported about 47000 seconds -no doubt some (but not all) of the speed gain is due to improvements in hardware. Remark 4. 11. The requirement that σ be a term-ordering is stronger than necessary. For instance, it is sufficient that σ orders by degree (how it orders within a fixed degree is unimportant). However, if we use such a general ordering then the list PPL will then have be to updated differently in step Direct-2.3(no) (e.g. fill it with all power-products of the next degree when it becomes empty, much like Wang's method [21]). Hypersurfaces Parametrized by Rational Functions In this section we consider a parametrization given by rational functions f 1 , . . . , f n in the field L = K(t 1 , . . . , t s ), where {t 1 , . . . , t s } is a set of indeterminates which are viewed as parameters. We can write this parametrization with a common denominator q, so that we have f i = pi q with q, p 1 , . . . , p n ∈ K[t 1 , . . . , t s ]. We construct the ideal I = tx − s, ty − s, z − s , and observe that I = x − y ∩ s, t, z . Hence we get I ∩ K[x, y, z] = z (x − y) which is not prime. The correct result, which may be obtained by including the generator ut − 1, is Implicit( In the following we present a "homogeneous" method for the computation of the ideal Implicit(f 1 , . . . , f n ) when f 1 , . . . , f n are rational functions; compared to the classical method the big advantage we have is that the Gröbner basis computation (i.e. the elimination) is performed on a homogeneous ideal. Proof. The kernel Implicit( p1 q , . . . , pn q ) is an ideal in the ring P = K[x 1 , . . . , x n ], which we view as a subring of R. We write P [x 0 ] to denote the subring K[x 0 , x 1 , . . . , x n ]: observe that a polynomial in P [x 0 ] is homogeneous in the induced grading if and only if it is homogeneous in the standard grading (i.e. in the usual sense of the word). As in Proposition 3.3, the proof of claim (a) follows immediately from the fact thatJ is an eliminating ideal, hence prime. To prove claim (b) we introduce the following sets: . . , pn q ) = 0} Clearly, the conclusion is reached if we prove the following claims. Claim (1) is just the definition of Implicit. To prove claim (2) we recall that the idealJ is homogeneous, henceJ ∩ P [x 0 ] is too. ThusJ ∩ P [x 0 ] can be generated by homogeneous elements, and S 2 contains all homogeneous elements inJ ∩ P [x 0 ], and so it surely generates the ideal. Finally, we prove claim (4). We have . . , pn q ), hence the claimed equality follows. Since claim (5) 5. We may relax the restriction in the theorem that each p i be non-zero; it is there just to allow an easy definition of d, the upper bound for the degrees. Remark 5. 6. We could be tempted to use the general method highlighted in Remark 5.1: namely, we homogenize the generators of J given there to get the eliminating idealJ † , and then imitate Proposition 3. 3. However, even if the ideal (J † ∩ P [h]) deh is principal, the idealJ † ∩ P [h] need not be principal, as the following example shows. The main drawback is thatJ † need not be saturated with respect to h. Example 5.7 (Ex 5.2 continued). We return to Example 5.2, but this time homogenize the generators of J = I + ut − 1 to produce the following idealJ † = tx − hs, ty − hs, z − s, ut− h 2 . However, elimination yieldsJ † ∩K[x, y, z, h] = xzh− yzh, xh 2 − yh 2 which is not principal. Even if we homogenize the generators and bring them all to the same degree, we get the idealJ ‡ = tx − hs, ty − hs, h(z − s), ut − h 2 , and again elimination producesJ ‡ ∩ K[x, y, z, h] = xzh − yzh, xh 2 − yh 2 . Remark 5. 8. In the case of rational functions, we could also be tempted to homogenize the input in a similar way to Theorem 5.3 but applying just hom(h) instead of equalizing the degrees with hom d (h) . This does not work because claim (4) of the proof fails, as the following easy example shows. Example 5. 9. In the ring K(t 1 , t 2 )[x 1 , x 2 , x 3 ] we consider the eliminating ideal The correct answer is Implicit(· · · ) = x 1 x 3 − x 2 2 . However, if we consider the ring 1 whose degree is 4 while the actual solution is the polynomial x 1 x 3 − x 2 2 whose degree is 6. We now turn Theorem 5.3 into an explicit algorithm: Remark 5. 10. In step RatPar-4 we may use any algorithm to compute the implicitization from the (homogeneous) polynomial parametrization, e.g. Algorithms 4.1 or 4.2 Modular Approach for Rational Coefficients It is well known that computations with coefficients in Q can be very costly in terms of both time and space. When possible, it is generally a good idea to perform the computation modulo one or more primes, and then "lift" the coefficients of these modular results to coefficients in Q. There are two general classes of method: Hensel Lifting and Chinese Remaindering. We shall use Chinese Remaindering. The modular approach has been successfully used in numerous contexts: polynomial factorization [22], determinant of integer matrices [5], ideals of points [4], and so on. In any specific application there are two important aspects which must be addressed before a modular approach can be adopted: • knowing how many different primes to consider to guarantee the result (i.e. find a realistic bound for the size of coefficients in the answer); • handling bad primes: i.e. those whose related computation follows a different route, yielding an answer with the wrong "shape" (i.e. which is not simply the modular reduction of the correct non-modular result). There is no universal technique for addressing these issues. For our particular application there is no useful coefficient bound, and only a partial criterion for detecting bad primes (see Remark 6.8). We shall use fault-tolerant rational recovery to overcome our limited knowledge about these two aspects (see section 6.3). Definition 6.1. (reduction modulo p) Given a prime number p we denote the usual "reduction mod p" ring homomorphism by ψ p : Z −→ Z/ p . We can extend ψ p naturally to Z[t 1 , . . . , t s ] by mapping the coefficients but preserving the power-products, and extend it further to rational functions (over Q) by localizing away from its kernel in Z[t 1 , . . . , t s ]. Bad Primes Let f 1 , . . . , f n ∈ Q(t 1 , . . . , t s ) be non-constant and such that Implicit(f 1 , . . . , f n ) = g is a principal ideal, for some g ∈ Q[x 1 , . . . , x n ]. Clearly the generator g is defined only up to a non-zero scalar multiple; we resolve this ambiguity by requiring g to be monic (with respect to some fixed term-ordering on Q[x 1 , . . . , x n ]). We can now define den(g) ∈ Z to be the least common denominator of the coefficients of g. Definition 6. 2. We say that the prime p is unsuitable if any of the following happens: . In other words p is unsuitable if it divides any denominator, or if the degrees of numerator and denominator of some f i change modulo p. It is easy to check whether p is unsuitable. We exclude all unsuitable primes from subsequent discussions. Definition 6. 3. We say that the prime p is bad if it is suitable but either of the following happens: (A) g is not in the domain of ψ p , that is p divides a denominator in g. We say that a prime is good if it is neither unsuitable nor bad. We say that p is very-good if it is good and Supp(g) = Supp(ψ p (g)); in other words, it does not divide the numerator of any coefficient in g. Example 6.4 (Bad primes). Given So the prime 3 is bad because, even though the modular implicitization is principal, it is not equivalent modulo 3 to the correct result. Indeed, even when Implicit(f 1 , . . . , f n ) is principal in Q[x 1 , . . . , x n ] we cannot be sure that Implicit(ψ p (f 1 ), . . . , ψ p (f n )) is principal too. For instance, given the parametrization whereas modulo 2 we find that From Remarks 4.2 and 4.6, we see that in cases such as this, where the modular inputs do not satisfy the assumption that Implicit(· · · ) be principal, our Algorithms 4.1 (ElimTH) and 4.2 (Direct) for computing Implicit(ψ p (f 1 ), . . . , ψ p (f n )) will simply return the first polynomial in the ideal that they find. Remark 6. 7. Only finitely many primes are good but not very-good. By definition a prime is good but not very-good if it divides the numerator of some coefficient of g, or equivalently if it divides the least common multiple of the numerators of the coefficients of g. Clearly only finitely many primes do so. In conclusion, only finitely many primes are not very-good. Remark 6.8 (Detecting bad primes). We do not have an absolute means of detecting bad primes, but given the implicitizations modulo two different primes we can sometimes detect that one of them is surely bad (without being certain that the other is good). What we can say depends on which algorithm we used to compute the implicitizations -we must use the same algorithm for both modular computations! If we run Algorithm 4.1 (ElimTH) with a bad prime p to produce the output g p then we know that deg(g p ) ≤ deg(g). Thus if we run Algorithm 4.1 with two different primes p 1 and p 2 , and if deg(g p1 ) < deg(g p2 ) then surely p 1 is a bad prime. Note that even if deg(g p ) = deg(g), we need not have g p = ψ p (g) as shown by the non-principal ideal in Example 6.4 above. If we run Algorithm 4.2 (Direct) with a bad prime p to produce the output g p then we know that LPP(g p ) < σ LPP(g) provided we use the same, fixed enumerative termordering σ. Thus if we run Algorithm 4.2 with two different primes p 1 and p 2 , and if LPP(g p1 ) < σ LPP(g p2 ) then surely p 1 is a bad prime. Single Prime Method Given input f 1 , . . . , f n we can pick a suitable prime p, and run one of our algorithms to get an output g p . If p is very-good then Supp(g p ) = Supp(g). We can then determine the coefficients of monic(g) by solving a linear system over Q. Let N = \| Supp(g p )\| and pick N distinct s-tuples of random integers; evaluating all the f i at each such s-tuple produces a "random point" on the hypersurface, i.e. a zero of g. If the N × N matrix whose (i, j)-entry is the value of the i-th power-product (in Supp(g p )) at the j-th tuple is of full rank then knowing that every point on the hypersurface is a zero of g, and knowing that the leading coefficient of monic(g) is 1 we can solve the linear system to get all coefficients of g calc , our "informed guess" for the value of monic(g). We must now verify that g calc is correct; we do this by simply substituting f 1 , . . . , f n into it. If our choice of prime was very-good then the substitution will verify that g calc is correct. Conversely, if our choice of prime p was not very-good then the candidate "informed guess" for the support of monic(g) was wrong, and g calc will lie outside Implicit(f 1 , . . . , f n ), so the substitution will give a non-zero result; in this case we must start again with a different prime, hoping that this time it will be very-good. This technique is advantageous when the implicitization is especially sparse (since then the linear system will be small). Remark 6. 9. We can make a cheaper initial verification by picking another random point on the hypersurface, and verifying that that point is a zero of g calc . Naturally, if this "randomized check" passes then a full verification must still be done. Multiple Prime Method A disadvantage of the single prime method is that if the prime chosen is not verygood then we discover this only at the end of a potentially expensive verification. We can greatly reduce the risk of a failed verification by using several different primes, and combining the corresponding modular answers using Chinese Remaindering. Our strategy must handle bad primes. Using the checks in Remark 6.8 we can detect and discard some bad primes, however it is possible that a few bad primes pass undetected. We use faulttolerant rational reconstruction to cope with any undetected bad primes; we will find the right answer so long as the good primes sufficiently outnumber the undetected bad ones. Moreover, when using several primes we do not require that any of the primes be very-good; it is enough for most of the primes to be good and "complementary" (i.e. the union of the supports of the answers from all the good primes tried must include the support of the true answer). The key ingredient in this approach is a fault-tolerant rational reconstruction procedure (e.g. see [1] and [8]): this enables rational coefficients to be reconstructed from their modular images even if some of those images are bad. The reconstruction procedure normally returns either the correct rational or an indication of failure, though there is a low probability of it producing an incorrect rational. So for certainty, the reconstructed implicit polynomial must be verified. We chose the HRR algorithm from [1] because it is better suited to our application: compared to ETL from [8] it requires fewer primes (and therefore fewer costly modular implicitizations) when reconstructing "unbalanced" rationals, i.e. whose numerator and denominator have differing sizes. From the test in step we have that g calc (f 1 , . . . , f n ) = 0, so the value returned is surely an element of Implicit(f 1 , . . . , f n ); consequently, g calc is a nonzero multiple of g. We show by contradiction that g calc is a scalar multiple of g. Suppose not, then g calc = f g for some non-constant polynomial f . Let σ denote the enumerative term-ordering used inside Algorithm 4.2 (Direct); and let deg * denote the weighted degree used inside Algorithm 4.1 (ElimTH) -note that condition (b) in our definition of "unsuitable" makes sure that the same weighted degree is used every time. Let T 1 = LPP σ (g calc ), then clearly T 1 > σ LPP σ (g). Let T 2 be a term of g calc of maximal weighted degree; then deg * (T 2 ) > deg * (g). Note that T 1 and T 2 could be the Remark 6. 11. Since each g p is defined only up to a scalar multiple, we normalize the polynomial by making it monic; this guarantees that for every good prime p, the corresponding polynomial g p is equal to ψ p (g). Timings In this section we show the practical merits of our algorithms. We conducted two series of experiments, which we report in the two tables below. The experiments were performed on a MacBook Pro 2.9GHz Intel Core i7, using our implementation in CoCoA 5. The columns headed "ElimTH" and "Direct" report the computation times for the respective algorithms: in each case there are separate columns for computations over a finite field (char 32003), and over the rationals (char 0). The column headed "Len" says how many terms there are in the resulting polynomial. The symbol ∞ in the tables means that the computation was interrupted after 20 minutes, and 0 means that the computation takes less than 0.001 seconds. A horizontal line in the middle of the tables separates examples with polynomial parametrizations from examples with rational parametrization. Table 1 contains statistics related to examples taken from the literature, which we list in Appendix 8. 1. It shows that, with the sole exception of Example 8.7, they are computed in almost no time. Examples from the Literature We found only two examples which defeated us: listed in our Appendix as Examples 8.9 and 8.10 -originally they were Examples 5.2 and 5.3 in [11]. We suspect they are essentially incalculable because the implicitizations are almost certainly polynomials of high degree (over 100) having very many terms (over 100000). Table 2 contains statistics related to our own examples, which we list in Appendix 8. 2. The small numbers in brackets in the columns for characteristic 0 are the number of moduli used in Algorithm 6.3 (ModImplicit) for reconstructing the rational coefficients. The time to compute the answer is essentially the product of the number of moduli and the time for a single finite field; the rest of the time is for verification, which can represent more than half the total time as in Example 8.27.	2016-10-13T15:11:56.000Z	2016-02-12T00:00:00.000	{ "year": 2017, "sha1": "75c3224584c09017306c6c906ac77ff3ee358486", "oa_license": "elsevier-specific: oa user license", "oa_url": "https://doi.org/10.1016/j.jsc.2016.11.002", "oa_status": "BRONZE", "pdf_src": "Arxiv", "pdf_hash": "75c3224584c09017306c6c906ac77ff3ee358486", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Mathematics", "Computer Science" ] }
228965259	pes2o/s2orc	v3-fos-license	A Review of the Combination of Lean and Green in manufacturing for Sustainable Development Lean and green are concepts that can not be separated from the business processes of a company, including in manufacturing. A company is easy to achieve but sometimes difficult to maintain it. Maintaining the achievements that have been achieved is a challenge for the company that must be faced. Sustainability is a concept that can help companies maintain their achievements. In this paper we focus on the combination of lean and green in manufacturing to achieve sustainability. Sustainability that is the focus of this paper is based on the triple bottom line, namely economic, environmental and social. Lean implementation is believed to increase the productivity of a company’s production system that can improve performance from an economic standpoint. While by applying green the company can achieve productivity in terms of environmental and social aspects. So that by synergizing these two concepts, the triple bottom line can be achieved in accordance with predetermined targets. Introduction Since last year, several specialized companies in developed countries have faced considerable competition. Big competition is caused by increasing customer demand so every company must deal with it. Lean Manufacturing has been widely adopted. by reducing waste and increasing the value of the production process, making products with more shortcomings, focusing on customers (high quality, low cost, short time), robust production operating systems, and shipping costs are the goals of most companies to face competition [1]. One of the biggest challenges facing manufacturing companies is environmental problems. A growing industry has a major influence on environmental damage. Increasing global warming, effects on human health, ozone depletion, water pollution, biodiversity loss, and economic impacts. Manufacturing systems then evolved by increasing awareness of environmental risks, the need to compete not only efficiently, but migrated from less environmentally friendly to environmentally friendly. An industry can be said to be environmentally friendly if it can minimize or eliminate the presence of environmental waste. Aspects to consider in environmental waste include energy, water, materials, waste, transportation, emissions and biodiversity [2]. In the late eighties, a concept called the concept of sustainable development. The purpose of the mini concept is Development that meets the needs of the present generation without reducing the ability of future generations to meet their own needs. Sustainable companies that contribute to sustainable development by providing economic, social and environmental benefits or what is called a "triple bottom line" [1]. Lean and green are concepts that support the achievement of the bottom line. lean is able to reduce waste in a production process to make the company achieve maximum productivity and increase company profits. Meanwhile, by applying the green principle to the production system, the company releases the waste generated from the production process. by combining these two concepts into the production system, a company can also achieve young development with sustainable development. Lean focuses on seven forms of waste consisting of overproduction, defects, unnecessary inventory, improper processing, excessive transportation, waiting, and unnecessary movements. while basically environmental waste can also be considered as additional waste. Therefore, the concept of lean and green must be applied simultaneously to reduce waste in the company. Based on this, this paper discusses the integration between lean and green in reducing waste based on lean and waste based on green to reach the triple bottom line. Lean in manufacturing In terminology Lean is a series of activities or solutions to eliminate waste, reduce non-value added (NVA) operations and increase value added (VA) operations [3]. A lean strategy is a paradigm / concept to produce a repair system that is supported and eliminates all forms of waste from production and supply chains to improve quality, reduce costs and increase value for customers [4]. Lean can be identified as a systemic and systematic approach to identifying and eliminating waste, or non-value-adding activities through continuous improvement by flowing products (material, work-in process, output) and information using internal and external pull systems to pursue excellence and perfection [5]. Lean which is applied to the whole company is called Lean Enterprise. Lean which is applied in manufacturing is called Lean Manufacturing. There are five basic principles of Lean [2]. 1. Identifying product value based on customer perspective 2. Identify Value Stream Mapping for each product. 3. Remove waste that does not add value from all activities along the value stream. 4. Organizing so that material, information and products flow smoothly and efficiently throughout the value stream process using a pull system 5. Continue to look for techniques and tools for improvement (improvement tools and techniques) to achieve excellence and continuous improvement. Figure 1. Principles of Lean Lean manufacturing is a systematic approach to identifying and eliminating waste through continuous improvement, in order to create a smooth production process flow with fast lead times and little waste [6]. lean manufacturing is a production system that uses very little energy and waste to fulfill what consumers want precisely. The purpose of lean manufacturing is to eliminate waste (non value adding activity) from a process so that activities along the value stream are able to produce value adding [7]. in the Toyota Production System (TPS) there are seven waste in the production process, namely as follows [8]: 1. Overproduction, which is a waste caused by excessive production, the intention is to produce products that exceed what is needed or produce earlier than the schedule that has been created. 2. Waiting, which is a waste of time waiting for the next process. Waiting is an interval when the operator does not use the time to do value adding activity because it waits for the product flow from the previous process (upstream). 3. Transportation, transportation is an important activity but does not add value to a product. Transportation is the process of moving material or work in process (WIP) from one work station to another, using either a forklift or conveyor 4. Excess processing, occurs when work methods or work sequences (processes) used are deemed unfavorable and flexible. 5. Inventories, are inventories that are less necessary. The point is that there is too much material inventory, too much work in process between one process and another that requires a lot of space to store it, the possibility of this waste is a very high buffer. 6. Motion, is an unnecessary activity / movement carried out by the operator that does not add value and slow down the process so that the lead time is long. 7. Defects, are products that are damaged or not according to specifications. This will cause an ineffective rework process, high customer complaints, and very high level inspection. Green in Manufacturing Green is a concept that deals with supply chains and production in the environment and handling in the environment consisting of air, energy, air, solid and hazardous waste [4]. Green Manufacturing aims to continue the integration of industrial processes and products, to reduce or avoid pollution in air, water and soil, to reduce waste at its source and to minimize risks to humans and other species [9] [2]. Green Manufacturing includes several practices namely pollution prevention and reduction of toxic use and design for the environment. Pollution prevention focuses on avoiding or minimizing waste or emissions through source reduction or on-site recycling. Reducing resources can be achieved in various ways that relate well to the process and the product [10] [2]. 1. Product Modification, when the composition of the shape and material of the product changes. 2. Input substitution, which means raw materials and non-polluting additives are used as process aids (for example lubricants and lubricants) with longer service time. 3. Modification of technology includes improving the automation process, process optimization, redesign of equipment and substitution processes. 4. Good Housekeeping, means changes in operational and management procedures to reduce or eliminate waste and emissions. For example spill prevention, increase worker instruction and training. Comparison of Lean and Green Manufacturing There are five steps in Lean Manufacturing, namely defining the flow of customers, defining the flow of value, making it "flow", "pulling" from customers, and striving for excellence [11]. In the first step, the customer's value must be determined, because you have to know what the customer wants and what the customer does not want from the production. However, customers can be interpreted in different ways, so there must be different types of waste, as shown in Figure 2. Some people argue that being environmentally friendly will incur costs, but being environmentally friendly means also saving costs from disposition of waste, emissions and exemption from collecting fines due to pollution, as well as the use of recycled materials. There will be a break even point between costs and benefits of being environmentally friendly. For those who have limited financial capacity, they can try to find a break-even point and plan investments that are appropriate for the company's sustainability [2]. Lean and Green for Sustainable Development in Manufacturing Sustainable Manufacturing itself is defined as "the creation of products of economic value through processes that minimize negative impacts on the environment, save energy and natural resources, and conserve natural resources and energy to guarantee their availability in the future. The process must also be safe for employees, the public, and consumers. " Sustainable Manufacturing "is an evolution of manufacturing systems ranging from traditional manufacturing systems, then lean manufacturing that focuses on reducing waste (waste reduction based), green manufacturing with 3R, until finally on the concept of sustainable manufacturing with the 6R approach to the product life cycle [12]. The implementation of Sustainable Manufacturing leads to the achievement of (sustainable development) as stated by the World Commission on Environment and Development is defined as "development that meets current needs without compromising the ability of future generations to meet their own needs" [13]. Tripple bottom line is the company's sustainability and growth does not depend on operating profit, but also concrete actions taken by the company on the environment, and justice. and everything is done in order to create sustainable development [14]. The concept of sustainability can be broken down into three aspects of understanding, namely [15]: 1. Economic sustainability, which is defined as development able to produce goods and services continuously to maintain the sustainability of government and avoid it happening sectoral imbalances that can damage agricultural production and industry. 2. Environmental sustainability: An environmentally sustainable system must be able to maintain stable resources, avoid the exploitation of natural resources and the function of environmental absorption. This concept also concerns the maintenance of biodiversity, stability of airspace, and other ecosystem functions that are not categorized as economic resources. 3. Social sustainability: Social sustainability is defined as a system capable of achieving equality, providing social services including health, education, gender, and political accountability	2020-11-05T09:08:41.749Z	2020-10-29T00:00:00.000	{ "year": 2020, "sha1": "0385b53f73d93a052e1e850bfb215151770f92c3", "oa_license": null, "oa_url": "https://doi.org/10.1088/1755-1315/575/1/012066", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "8cc7374daf29038537017705a98034ba3becb5a8", "s2fieldsofstudy": [ "Environmental Science", "Business", "Engineering" ], "extfieldsofstudy": [ "Physics", "Business" ] }
31651404	pes2o/s2orc	v3-fos-license	Incorporating Intermediate-Severity Disturbances in Oak Stand Development We propose a conceptual framework for the development of closed canopy oak (Quercus) stands that incorporates the influence of intermediate-severity canopy disturbance events, provides for the possibility of multiple developmental pathways, and does not narrowly define an endpoint given the stochastic nature of natural disturbances. The proposed model differs from the current oak stand development model in three primary ways. First, our proposed model acknowledges more than one mixed stage of development after an intermediate-severity disturbance based on the pre-disturbance condition and disturbance agent. Second, we suggest that these discrete mixed stages may progress in their development along different pathways and stands may be structurally dissimilar when they reach the complex stage. Third, we contend that the complex stage of development in oak stands is not usually achieved in the absence of the mixed stage because the return interval of these events is shorter than the period required for oak stands to reach the complex developmental stage via gap-phase processes alone. Our proposed framework for oak stand development should aid decision making in oak-dominated systems. Introduction The ability to forecast stand growth and development is critical in forest management as some degree of predictability is required for decision making [1].Thus, stand development models provide a basis for long-term forest planning.Oliver and Larson [2] provided a process-based model of growth and development for even-aged stands following catastrophic disturbance.In this model, stand development is divided into four discrete phases based primarily on competition for growing space.Listed sequentially, these are stand initiation, stem exclusion, understory reinitiation, and complex stages of development.Each phase exhibits unique stand structures and marks a discrete period in the development of the stand.The Oliver and Larson [2] model established the relationships between canopy disturbance events and the progression of stands along the developmental pathway from one stage to the next.Based on the Oliver and Larson [2] model of stand development, Johnson et al. [3,4] established a conceptual model for the development of oak (Quercus) stands (Figure 1).Johnson et al. [4] categorized canopy disturbances based on spatial extent, severity, and frequency, and disturbance events were broadly grouped as stand initiating (i.e., catastrophic), incomplete stand-scale (i.e., intermediate-severity events too localized to be classed as stand initiating and too large to be classed as gap scale), or gap scale (i.e., localized).Unlike the Oliver and Larson [2] model, the oak development model includes a mixed stage of development following intermediate-severity disturbances.Johnson et al. [4] noted that oak stands in the mixed developmental stage are ubiquitous because the events that create them are so common.Our purpose is to elaborate upon the importance of intermediate-severity disturbances and the resultant mixed stage of oak stand development, which are poorly understood and we contend underappreciated. Background on Disturbance and Development in Oak Stands In the oak stand development model, a catastrophic disturbance regenerates a new stand and so begins the stand initiation stage of development.In this developmental phase, stem density is high and thus, competition-induced mortality is common [5].The trees that are able to overtop competitors (i.e., the winners) will eventually stratify to form a true canopy and crown closure marks the beginning of the stem exclusion stage of development.During the stem exclusion phase, competitioninduced mortality remains high.The death of canopy trees during this stage results in relatively small gaps and freed growing space is rapidly captured by adjacent trees.By age 40 years the loss of these Background on Disturbance and Development in Oak Stands In the oak stand development model, a catastrophic disturbance regenerates a new stand and so begins the stand initiation stage of development.In this developmental phase, stem density is high and thus, competition-induced mortality is common [5].The trees that are able to overtop competitors (i.e., the winners) will eventually stratify to form a true canopy and crown closure marks the beginning of the stem exclusion stage of development.During the stem exclusion phase, competition-induced mortality remains high.The death of canopy trees during this stage results in relatively small gaps and freed growing space is rapidly captured by adjacent trees.By age 40 years the loss of these canopy trees is sufficient to result in increased rates of crown extension, height, and radial growth [6,7].The recruitment of seedlings to sapling and small tree size classes marks the onset of the understory reinitiation stage of development.In oak stands, these understory stems are typically comprised of more shade-tolerant species [8][9][10].With stand age, the size of canopy gaps created by a fallen tree increases and the ability of neighboring trees to close gaps decreases [5], but during the understory reinitiation phase, the gaps are still of the size that they most often close by lateral branch extension rather than subcanopy height growth [11][12][13][14].The period in stand development when most gaps are sufficiently large so that they fill by subcanopy recruitment rather than lateral crown expansion marks the beginning of the complex developmental stage.At this stage, oak stands should exhibit structural characteristics that are typically used to define old-growth conditions.Some authors have proposed multiple old-growth stages for some forest cover types [15], but the current oak model recognizes a single complex stage. In the current oak development model, the large gaps that characterize structure of the complex stage result from the gap-scale disturbance processes that occur continuously throughout prior stages of development.Although gaps in older stands occur less frequently than in earlier stages of development, because older stands contain fewer individuals and have reduced competition and mortality rates [5], the gaps in complex stage stands are relatively large.These gaps take long periods to close, which increases the probability of a new individual recruiting to the canopy [12,[16][17][18].In the current oak stand development model, the structural changes with maturity are driven by these gap-scale disturbance events in the absence of larger exogenous disturbances.Johnson et al. [4] posit that stand initiation commences following a catastrophic disturbance and development progresses from one stage to the next via gap-scale disturbances, but a disturbance of the intermediate scale alters development to the mixed stage.The model does not discern different mixed stages based on the stage of development at the time of the intermediate-severity disturbance or specific intermediate-severity disturbance agents.Once the mixed stage is created by an intermediate-severity disturbance, the stands may then continue progression along a pathway driven by more localized disturbances until the complex stage is reached, a catastrophic disturbance regenerates the stand, or development is altered by a subsequent intermediate-severity event. We propose a revised conceptual model of oak stand development driven by canopy disturbance events (Figure 2).Our goal is to build on the concepts provided by prior authors using the Johnson et al. [4] framework as an archetype.Our revised model of oak stand development differs from the existing model in three fundamental ways.First, our proposed model acknowledges more than one mixed stage of development (i.e., a broad range of structural conditions) after an intermediate-severity disturbance based on the pre-disturbance condition (i.e., the stage of development) and disturbance agent.Second, we suggest that these discrete mixed stages may progress in their development along different pathways and may be structurally dissimilar when they reach the complex stage (i.e., they do not necessarily converge).Third, we contend that the complex stage of development is not usually achieved in the absence of the mixed stage.Below we provide our basis for these modifications and summarize our revised model of oak stand development.Based on the variability inherent in oak-dominated systems across the temperate zone, we present a conceptual framework rather than a mathematically derived classification scheme for understanding developmental pathways of oak stands.We note that our conceptual framework is appropriate for closed canopy oak stands (i.e., forests and woodlands) and may not be an appropriate framework for the development of oak stands with perpetual open canopies (i.e., oak savannas). Multiple Mixed Stages The current model of oak stand development only recognizes a single mixed stage following an intermediate-severity disturbance, regardless of the stage of development at the time of the event or disturbance agent.The response of a stand to a disturbance event is controlled in part by the predisturbance stand condition.Elements of stand structure, such as stem density, canopy tree height, quadratic mean diameter, crown volume, and the vertical distribution of foliage, differ between the stem exclusion and understory reinitiation stages of development.Therefore, it follows that the same intermediate-severity disturbance event would impact oak stands of these two stages of development differently (i.e., the same wind event would result in different structures based on the initial structural condition).For example, larger trees are more likely to be killed by strong wind events relative to smaller trees [19][20][21][22][23].The mortality of these larger trees would result in larger canopy gaps and larger inputs of coarse woody debris.Characteristics of the disturbance may be driven by predisturbance structure.More work is needed to quantify the effects of intermediate-severity disturbances on oak stands at different stages of development, but it serves to reason that stands in the stem exclusion and understory reinitiation stages would not structurally resemble each other after an intermediate-severity disturbance event.We contend that intermediate-severity disturbances do not homogenize residual structures of stands that exhibit different pre-disturbance conditions. Multiple Mixed Stages The current model of oak stand development only recognizes a single mixed stage following an intermediate-severity disturbance, regardless of the stage of development at the time of the event or disturbance agent.The response of a stand to a disturbance event is controlled in part by the pre-disturbance stand condition.Elements of stand structure, such as stem density, canopy tree height, quadratic mean diameter, crown volume, and the vertical distribution of foliage, differ between the stem exclusion and understory reinitiation stages of development.Therefore, it follows that the same intermediate-severity disturbance event would impact oak stands of these two stages of development differently (i.e., the same wind event would result in different structures based on the initial structural condition).For example, larger trees are more likely to be killed by strong wind events relative to smaller trees [19][20][21][22][23].The mortality of these larger trees would result in larger canopy gaps and larger inputs of coarse woody debris.Characteristics of the disturbance may be driven by pre-disturbance structure.More work is needed to quantify the effects of intermediate-severity disturbances on oak stands at different stages of development, but it serves to reason that stands in the stem exclusion and understory reinitiation stages would not structurally resemble each other after an intermediate-severity disturbance event.We contend that intermediate-severity disturbances do not homogenize residual structures of stands that exhibit different pre-disturbance conditions.Therefore, our revised model of oak stand development recognizes a broad range of mixed stages based on the stage of development at the time of the disturbance event. Examples of natural intermediate-severity disturbance agents include low intensity tornadoes, ice storms, floods, mixed severity fires, insect outbreaks, and pathogens (Table 1).By definition, intermediate-scale disturbances are those greater in extent than gap-scale events, but not stand replacing.Thus, the range of disturbance events that may be classed as intermediate severity is considerably greater than the range of gap-scale and stand replacing events.This inherent variability in biological legacies, which represent the effects of prior disturbances [24,25], from intermediate-severity events makes forecasting the developmental pathway difficult and complicates model development.For example, two different intermediate-severity disturbance events may leave the same residual basal area in a stand, but the size, number, and spatial distribution of residual trees may be considerably different (note the range of mixed stage conditions depicted in Figure 2 with uniform, aggregated, and linear patterns of disturbance).A species-specific insect outbreak may remove trees singly or in small groups beyond the background mortality rate from throughout a stand (canopy gaps formed beyond the background rate would be an intermediate-severity event), whereas a low intensity tornado may remove most all trees in a linear swath, but leave the majority of the stand undisturbed [23,26,27].Based on pre-disturbance condition and disturbance agent, resultant structures from intermediate-severity disturbances may vary widely, not only horizontally, but also vertically.Impacts of an intermediate-severity disturbance in stem exclusion stage stands, with a single stratum, may substantially differ from the impacts of understory reinitiation or complex stage stands, which have two or more vertical strata.Tree crowns have not differentiated in stem exclusion stands and thus, these stands may not contain individuals that are more or less likely to be removed by intermediate-severity disturbances because of tree traits.However, in multi-strata stands larger trees in dominant positions may be more susceptible to windthrow or damage from glaze accumulation and small trees may be more susceptible to some fire events. We emphasize that the mixed stage should not be confused with the complex stage of development.Mixed stage stands may contain a single age class of canopy trees and may not contain large trees at wide spacings that characterize older, complex stage stands.Although an intermediate-severity disturbance may enhance structural complexity, these events do not by themselves create the structures epitomized by stands in the complex stage.In this respect, our proposed model does not differ from Johnson et al. [4]. Development of Mixed Stage Stands Based on the current model of oak development, in the absence of additional exogenous disturbance, stands in the mixed stage would transition to the complex stage via gap-scale processes.As the impacts of intermediate-severity disturbances would be different based on stage of development, so should the responses of the stands.The stem exclusion phase, for example, lacks advanced reproduction to take advantage of new growing space, the smaller and younger canopy trees during this phase may sprout more prolifically, and the residual canopy trees may respond more vigorously relative to later stages of development [50][51][52][53].Additionally, if the intermediate-severity event resulted in uniform mortality (i.e., trees killed individually rather than in groups), the openings formed by the loss of canopy trees in the stem exclusion phase may be smaller and the vigorous perimeter trees may close the voids relatively quickly.Although these openings may provide a mechanism for seeds to germinate and perhaps for seedlings to establish, the probability of canopy accession by one of these new germinates would be relatively low.Depending on the severity of the intermediate-severity event, the developmental pathway of an impacted stem exclusion stage stand may result in a mixed stage stand that subsequently progresses to either the understory reinitiation stage or the complex stage of development.For example, if the intermediate-severity event consisted of gap-scale disturbances created beyond the background rate, the stands would enter the mixed stage.This phase may be short-lived however as the disturbance may provide for the establishment of advanced reproduction.As an understory layer develops and canopy openings close, the stand would shift to the understory reinitiation stage, but material and information legacies would remain in the system [24,25].In this example, the duration of the mixed stage may be influenced by a variety of factors including the sprouting potential of killed trees, seed production of residual trees, and the suitability of the seedbed for germination and stem establishment.In contrast to the above scenario, canopy openings created by an intermediate-severity disturbance with a uniform mortality pattern during the understory reinitiation phase would be larger, as gap size is influenced by canopy tree size.Advanced reproduction would exist during the understory reinitiation stage of development and thus, the probability of gap capture by a sub-canopy tree would be much higher relative to the same event during the stem exclusion phase [12,17].Understory reinitiation stage stands impacted by an intermediate-severity event would enter the mixed stage and in the absence of another incomplete stand-scale disturbance would progress to the complex stage.Thus, oak stand development may not be linear and deterministic.Multiple developmental pathways may exist, and these pathways may be more web-like than linear. The spatial pattern of the intermediate-severity disturbance, resulting from different disturbance agents, may be a strong influence on subsequent development.Intermediate-severity events that remove trees uniformly throughout a stand cause a transition to the mixed stage, but this stage may be ephemeral.Such events may ultimately accelerate development to later stages.Intermediate-severity disturbances that result in clustered mortality patterns (patch or linear mortality) may allow for the establishment of a new age class.These openings would be larger and thus exist for longer periods, which increases the probability of canopy accession by new germinates or advanced regeneration.The spatial pattern of canopy disturbance may influence the scale of structural heterogeneity of the affected stand.A uniform disturbance may create fine-scale heterogeneity, whereas patch or linear mortality would exhibit structural complexity at a broader scale.The spatial configuration of dead canopy trees and the sizes of the canopy openings are strong influences on subsequent development. The structural characteristics of stands in the complex stage may reflect their unique disturbance histories (i.e., ecological memory) and may be dissimilar to stands that were impacted by an intermediate-severity disturbance at another point during their development or a different intermediate-severity disturbance agent (note the range of complex stage conditions depicted in Figure 2).The different effects on stand structure and responses to an intermediate-severity disturbance based on pre-disturbance condition and disturbance agent, may in part explain why even simple structural measures, such as stem density, basal area, quadratic mean diameter, and q-factor, vary so widely even in stands with similar species composition and in the same biophysical region [54,55].Additionally, complex stage stands may be impacted by intermediate-severity disturbance and enter the mixed stage of development.Intermediate-severity disturbances in complex stage stands would result in an increased land fraction in high light environments, and may result in a reduced number of large canopy trees that epitomize the complex phase as these individuals may be more susceptible to some exogenous canopy disturbance agents relative to smaller sized stems. Intermediate-Severity Disturbance Frequency We contend that the complex stage of development is not usually reached in the absence of the mixed stage.This hypothesis underscores why it is important to acknowledge that a range of mixed stages can occur based on pre-disturbance condition and disturbance agent, and that mixed stage stands may progress along different developmental pathways.We do not suggest that an intermediate-severity disturbance is essential to achieve a structurally complex oak stand or initiate some essential function, but rather the return interval of intermediate-severity disturbances in oak stands is shorter than the period required for stands to reach the complex developmental phase via gap-scale disturbance alone.Theoretically, an oak stand could reach the complex stage via gap-scale disturbances only but in reality, this does not often occur based on the return interval of intermediate-severity events.Disturbance history reconstructions using tree-ring records and forest inventory data have been conducted in some oak stands of the eastern USA.The studies have shown that the return interval of disturbance events that removed trees from at least 25% of the stand was typically 20-50 years [6,26,[56][57][58][59][60][61].Disturbances of this spatial extent and severity would be classified as intermediate-scale events.Collectively, such disturbances are relatively common in the temperate zone.For example, over 1250 tornadoes occur in the USA annually [62] and over 13,500 instances of damaging winds (wind speeds ≥ 80 kph) were reported in the USA in 2016 alone [63].For some oak-dominated sites the return interval of ice storms may be as short as 20 years [58,64].Thus, we posit most all oak stands pass through the mixed stage of development at least once prior to reaching the complex stage. Implementation and Management Implications Quantitative stand structural measures for pre-and post-intermediate-severity disturbance are needed to test this conceptual model of development.Specifically, data for stand basal area, tree density, diameter structure, horizontal patterns of live and dead stems, vertical distribution of tree crowns, and canopy openness are required to test the efficacy of the proposed conceptual framework [15].With additional quantitative data, it may be possible to subdivide the broad mixed stage we propose into discrete categories and to provide developmental pathways for each.Information on long-term disturbance history from retrospective studies or permanent monitoring plots is also needed to provide more information on the return interval of disturbances across different severities.Pre-and post-disturbance stand structural data and disturbance history reconstructions are available for some oak-dominated stands, but the sample size is insufficient because of the variability of stand conditions in oak-dominated systems of the temperate zone.Disturbance response in oak stands may differ depending on co-occurring taxa.For example, stands with abundant shade-tolerant species, such as red maple (Acer rubrum), may respond differently than those where shade-intolerant species, such as tulip-poplar (Liriodendron tulipifera) often colonize disturbed environments.With a larger dataset, it would be possible to develop a dichotomous key based on simple stand structural measures (e.g.basal area by tree size classes) to classify stands to the stage of development [15].Such a model may have to be adjusted to accommodate oak stands on extremely high productivity and low productivity sites. Our conceptual model highlights the wide range of potential stand structures that may occur naturally because of canopy disturbance in oak-dominated systems.These structures may be used to guide management targets and actions.It is also important for managers to understand how stands might respond to natural or anthropogenic intermediate-severity disturbances and how development might proceed following such events.Our conceptual framework provides managers with a more comprehensive understanding of stand developmental pathways and disturbance events, which should prove especially useful for those that wish to develop silvicutural systems based on natural disturbance processes and developmental patterns.Most oak stands in the eastern USA are managed in an even-aged system that treats entire stands uniformly.Managers that desire to follow a natural disturbance-based silviculture approach, in attempt to enhance native forest diversity, maintain ecosystem functions, and promote resiliency to future perturbations, might consider applying even-aged treatments in groups rather than evenly through a stand.The group opening sizes and shapes could be based upon those that result from natural intermediate-severity disturbance events.The return interval of stand entries may be based upon the return interval of natural intermediate-severity disturbances.Reserve trees in the group openings could be retained at variable densities.Such a system would promote intra-stand heterogeneity and result in two or more age classes at the stand-scale.An example would be a system that uses group shelterwoods with reserves with successive harvest entries timed at 20-50 year intervals.Such a system would entail relatively frequent entries, but would also allow for a wide range of potential stand structures to achieve a wide range of management objectives. Conclusions We hypothesize that oak stand development is not necessarily deterministic and linear, but may follow multiple pathways.Our conceptual framework of oak stand development incorporates the influence of intermediate-severity disturbance events, outlines the possibility of multiple pathways, and does not narrowly define an endpoint given the stochastic nature of natural disturbances.Our model differs from the current oak stand development model in three primary ways.First, our proposed model acknowledges more than one mixed stage of development after an intermediate-severity disturbance based on the pre-disturbance condition and disturbance agent.Second, we suggest that these discrete mixed stages may progress in their development along different pathways and may be structurally dissimilar when they reach the complex stage.Thus, the mixed and complex stages are relatively broad classes as they may represent a wide range of structural conditions.Third, we contend that the complex stage of development is not typically achieved in the absence of the mixed stage, not because intermediate-severity disturbances are essential, but rather because the return interval of such events is shorter than the period required for oak stands to reach the complex developmental phase via gap-scale disturbance alone.Our proposed model illustrates that oak stands progress from one stage to the next via gap-scale disturbance processes, but superimposed over this background disturbance regime are intermediate-severity disturbances that alter, temporarily or permanently, the developmental pathway.Catastrophic disturbances at any point during development reset the system to the stand initiation stage of development.Intermediate-severity disturbances play important roles in the development of temperate oak stands.These stochastic events make stand developmental pathways complex, and acknowledging that oak stand development may have a web-like nature with multiple pathways should improve our ability to manage the oak resource. Forests 2017, 8 , 284 2 of 11 Figure 1 . Figure 1.Model of oak stand development adapted from Johnson et al. [4] illustrating forest disturbances and their relation to stages of stand development and disturbance-recovery rates. Figure 1 . Figure 1.Model of oak stand development adapted from Johnson et al. [4] illustrating forest disturbances and their relation to stages of stand development and disturbance-recovery rates. Figure 2 . Figure 2. Conceptual model of oak stand development shaped by canopy disturbance events.Light gray lines indicate the developmental pathway driven by gap-scale disturbances, medium gray lines by intermediate-severity disturbances, and black lines by catastrophic disturbances.Note the variability of residual stand structures after intermediate-severity disturbance events is depicted to vary left to right from uniform to aggregated to linear patterns of tree mortality based on disturbance agent.The mixed stage is also depicted to vary top to bottom based on pre-disturbance condition (MSE: mixed stage from stem exclusion, MUR: mixed stage from understory reinitiation, and MCS: mixed stage from complex stage).Likewise, a wide range of structural conditions is depicted in the complex stage based on disturbance history. Figure 2 . Figure 2. Conceptual model of oak stand development shaped by canopy disturbance events.Light gray lines indicate the developmental pathway driven by gap-scale disturbances, medium gray lines by intermediate-severity disturbances, and black lines by catastrophic disturbances.Note the variability of residual stand structures after intermediate-severity disturbance events is depicted to vary left to right from uniform to aggregated to linear patterns of tree mortality based on disturbance agent.The mixed stage is also depicted to vary top to bottom based on pre-disturbance condition (MSE: mixed stage from stem exclusion, MUR: mixed stage from understory reinitiation, and MCS: mixed stage from complex stage).Likewise, a wide range of structural conditions is depicted in the complex stage based on disturbance history. Table 1 . Examples of studies that have examined intermediate-severity disturbances from a variety of disturbance agents in oak-dominated stands.	2017-08-25T12:44:15.450Z	2017-08-05T00:00:00.000	{ "year": 2017, "sha1": "86e47f83481c0240f5bfa9e45b1a2c2bfd1398ee", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1999-4907/8/8/284/pdf?version=1501921675", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "86e47f83481c0240f5bfa9e45b1a2c2bfd1398ee", "s2fieldsofstudy": [ "Environmental Science" ], "extfieldsofstudy": [ "Environmental Science" ] }
15710951	pes2o/s2orc	v3-fos-license	Therapeutic Effects of Korean Red Ginseng Extract in a Murine Model of Atopic Dermatitis: Anti-pruritic and Anti-inflammatory Mechanism Korean red ginseng (KRG) and ginsenosides exhibit diverse biological effects, including anti-inflammatory and anti-allergic. We aimed to investigate the therapeutic effect of KRG in a murine model of atopic dermatitis (AD) is mediated whether by diminishing the pruritus or by suppressing the inflammation. Thirty NC/Nga mice were randomly divided to 5 groups. AD-like skin lesions were induced by percutaneous challenge with 2,4,6-trinitro-1-chrolobenzene (TNCB) on the ears and backs of NC/Nga mice. KRG extract, evening primrose oil, cyclosporine, and phosphate-buffered saline were administered orally by a gastric tube. Each study group was also divided into scratching-permitted and scratching-restricted subgroups to evaluate the impact of scratching behavior on AD. The effects of KRG and the other agents were assessed by measuring the clinical severity score, ear thickness, extent of transepidermal water loss (TEWL), number of scratching movements, total systemic immunoglobulin E (IgE) and interleukin (IL)-31 levels, histologic changes of cutaneous lesions, and mRNA expression levels of tumor necrosis factor (TNF)-α, interferon (IFN)-γ, thymic stromal lymphopoietin (TSLP), and IL-31. KRG exerts therapeutic effects against AD by inhibiting the T helper 2 (Th2) mediated inflammation as well as by diminishing the itching sensation. Moreover, restricting scratching behavior suppresses the vicious cycle of itching and scratching, thus reducing clinical and systemic inflammation in our murine model of AD. INTRODUCTION Itching is the most serious clinical symptom and a major diagnostic criterion of atopic dermatitis (AD) (1). Itching provokes the scratch reflex; this reflex is harmful because the itching-scratching cycle aggravates the existing skin lesions (2). Scratching behavior not only intensifies pruritus, but also further compromises the skin barrier and induces inflammation (3). Controlling inflammation and the sensation of abnormal itching are important for successfully managing AD (4). Several studies have shown that red ginseng has anti-inflammatory and antiallergic properties (5)(6)(7)(8)(9). However, few studies have determined whether red ginseng has an anti-pruritic effect. We previously examined the effects of systemic administration of Korean red ginseng (KRG) on a 2,4,6-trinitro-1-chrolobenzene (TNCB)-induced model of AD in NC/Nga mice and found that KRG prevented the development of acute AD-like lesions and the late onset of chronic AD-like lesions in NC/Nga mice model (7,8). KRG extract significantly reduced the total clinical severity score and the level of systemic immunoglobulin E (IgE) in a mouse model of AD. In the chronic stage, oral administration of KRG not only reduced the T helper 1 (Th1)-mediated response, demonstrated by the reductions in tumor necrosis factor (TNF)-α and interferon (IFN)-γ levels, but also inhibited regulatory T (Treg) cells and Langerhans cells (LCs) (7). Oral administration of KRG was found to partially suppress thymic stromal lymphopoietin (TSLP), dendritic cells (DCs), and the T helper 2 (Th2) response of early AD-like skin lesions in NC/Nga mice (8). The authors attempted to identify whether the therapeutic effects of KRG on AD is the anti-inflammatory effect on the primary inflammatory process or is mediated by reducing itching to decrease the secondary inflammation by itch-scratch cycle in this experiment. In the present study, we developed a novel method of protecting the skin from scratching. Animals Thirty SPF NC/Nga female mice (5 weeks old) were purchased from Charles River Laboratories (Kanagawa, Japan) and housed at 22°C with a 12 hours light-dark cycle. Food and water were freely available. Induction of AD-like skin lesions TNCB was purchased from Sigma (St. Louis, MO, USA) and used to induce AD-like skin lesions as previously described (7,8). Briefly, the abdomen of each mouse was sensitized epicutaneously with 150 µL of 5% TNCB dissolved in a 4:1 mixture of ethanol and acetone. On day 7 postsensitization, the dorsal skin and ears of the mice were challenged with 190 µL of 1% TNCB dissolved in a 4:1 mixture of acetone and olive oil. Three days after the first challenge, a 1% TNCB solution was repeatedly applied to the dorsal skin and ears. The study design is summarized in Fig. 1. Animal groups and drugs Thirty mice were divided randomly into 5 groups: AD control, KRG-treated (KRG), cyclosporine-treated (cyclosporine), evening primrose oil-treated (EPO), and sham. Each experimental group consisted of 6 mice. The AD control and sham group received 0.2 mL phosphate-buffered saline (PBS). The KRG, cyclosporine, and EPO groups were orally administered 200 mg/ kg KRG, 2.5 mg/kg cyclosporine, and 50 mg/kg EPO for 10 days, respectively and they did not received PBS. The AD control, KRG, cyclosporine, and EPO groups was applied TNCB on the back and ears. The sham group received only PBS (0.2 mL) by a gastric tube without TNCB challenge. Each study group was divided into 2 subgroups ( Fig. 2A). Mice in the first subgroup (scratching-permitted) were allowed to freely scratch their TNCB-induced AD-like dorsal lesions. In contrast, mice in the second subgroup (scratching-restricted) were prevented from scratching their skin via a custom-designed plastic chamber affixed to their backs before starting the experiment (Fig. 2B). Each subgroup consisted of 3 mice. The scratching-restricted subgroups are designated with the "@" symbol after the drug names in the figures. Measurement of spontaneous scratching behavior AD-like skin lesions and scratching behavior were induced by repeated application of TNCB solution onto the ears and backs of NC/Nga mice as previously described (7,8). Baseline spontaneous scratching movements were obtained by recording the scratching movements on the abdomen for 2 hours after TNCB sensitization (day 6). On days 7 and 10, scratching behavior was monitored by video for 2 hours after 1% TNCB challenge on the ears and back. The number of scratching movements after challenge was compared with baseline measurements. Measurement of total clinical severity score and ear thickness The severity of AD was evaluated by scoring clinical signs and symptoms as previously described (7,8). The total clinical severity score was defined as the sum of the scores for erythema/ hemorrhage, edema, excoriation/erosion, and scaling/dryness on the following scale: 0 (none), 1 (mild), 2 (moderate), and 3 (severe). Assessment was performed by an investigator who was blinded to group assignment. Ear thickness was measured using a dial caliper (Kori Seiki MFG, Tokyo, Japan) on day 0 to ob-tain a baseline measurement and then again on days 7 and 10 after TNCB challenge on the ears and back. Measurement of transepidermal water loss (TEWL) TEWL was assessed on the dorsal skin of the NC/Nga mice using a VapoMeter instrument (Delfin Technologies, Kuopio, Finland). TEWL was measured at baseline (day 0) and on days 7 and 10. Tissue samples and histopathologic examination On days 7 and 10, 4 mm punch biopsies were performed on the backs of mice and tissue samples were also excised from each ear. The samples were fixed in 4% paraformaldehyde (PFA) for hematoxylin and eosin (H & E) and toluidine blue staining. Sections were evaluated at × 400 magnification in 3 consecutive microscopic fields for quantitative analysis. Enzyme-linked immunosorbent assay (ELISA) Blood samples were collected from the retro-orbital plexus using heparinized glass capillary tubes at the end of the experiment. Serum was obtained by centrifugation and stored at −20°C until use. Serum concentrations of total IgE and IL-31 were determined using a mouse IgE ELISA kit (Bethyl, Montgomery, TX, USA) and a mouse IL-31 ELISA kit (R & D Systems, Minneapolis, MN, USA), respectively, according to the manufacturers' instructions. Statistical analysis The significance of differences among experimental groups was examined using Student's t-test and analysis of variance (ANO-VA). Treatment group comparisons were evaluated using ANO-VA. For subgroup comparison in each group and comparison of the KRG group and other groups we used the Student's t-test. All calculations were performed in SPSS v19.0 (SPSS Inc., Chicago, IL, USA) and all values are expressed as mean ± standard error of the mean (SEM). P values less than 0.05 were considered statistically significant. Ethics statement The animal care, handling and experimental procedures were carried out in accordance with a protocol approved by the Animal Care and Use Committee of Catholic University in Korea (CIMH-2013-010). RESULTS KRG, EPO, and cyclosporine effectively suppressed AD-like symptoms are evoked by TNCB challenge AD-like symptoms such as erythema, edema, excoriation, dryness, and scaling appeared after sensitization and percutaneous challenge with TNCB on the backs and ears of NC/Nga mice. Total clinical severity scores of the KRG, EPO, cyclosporine, and AD control groups were significantly increased compared with those of the sham group on days 7 and 10 (P < 0.001). Moreover, total clinical severity scores were significantly elevated in the AD control group compared with the sham, KRG, EPO, and cyclosporine groups on day 7 and compared with the sham, KRG, and cyclosporine groups on day 10 (P < 0.001; Fig. 3A). System-ic administration of KRG, effectively suppressed clinical symptoms compared with the EPO group on day 7 (3.00 ± 0.26 vs. 4.00 ± 0.37, P = 0.049) and compared with the EPO and cyclosporine groups on day 10 (7.00 ± 0.37 vs. 10.00 ± 0.37, P < 0.001 and 7.00 ± 0.37 vs. 8.00 ± 0.26, P = 0.049, respectively). Clinical severity scores were also significantly higher in the scratchingpermitted subgroups on day 10 in both the KRG and EPO groups (7.67 ± 0.33 vs. 6.33 ± 0.33, P = 0.047 and 10.67 ± 0.33 vs. 9.33 ± 0.33, P = 0.047, respectively). In general, clinical severity scores tended to be higher in the scratching-permitted subgroups than in the scratching-restricted subgroups, although these differences were not always statistically significant. KRG treatment reduces scratching behaviors in TNCBinduced NC/Nga mice We monitored the scratching behavior of each group to determine the effect of each agent on pruritus. To clinically evaluate the extent of pruritus, we monitored scratching movements at baseline before TNCB challenge (day 6) and for 2 hours after TNCB challenge (days 7 and 10). This approach was based on a previous study by Inagaki et al. (10) showing that most scratching movements occurred in the first hour after initial 2,4-dinitrofluorobenzene (DNFB) challenge and in the first 2 hours after the ninth challenge. The number of scratching movements increased over time Treatment with KRG and cyclosporine significantly reduces ear thickness The mean ear thickness increased over time from day 0 to day 10 after initial TNCB application. Ear thickness of the KRG (0.27 ± 0.01 cm) and cyclosporine (0.27 ± 0.02 cm) groups was significantly decreased on day 7 compared with the AD control group (0.34 ± 0.03 mm; P = 0.017 and P = 0.020, respectively; Fig. 3C). Thus, both KRG and cyclosporine exert anti-inflammatory effects when administered systemically. Significant differences were also observed between the 2 AD subgroups on day 7 (scratching-permitted subgroup 0.32 ± 0.01 mm vs. scratching-restricted subgroup 0.36 ± 0.01 mm, P = 0.031). No significant differences were observed between the scratching-permitted and scratching-restricted subgroups within the sham, KRG, EPO, and cyclosporine groups on day 7 and within all groups on day 10, presumably because the ears were not protected from scratching. KRG treatment prevented TEWL increase We measured TEWL on the backs of NC/Nga mice on day 0 (baseline), day 6 (before the first TEWL challenge), day 7 (2 hours after the first TEWL challenge), and day 10 (2 hours after the second TEWL challenge) using a VapoMeter instrument (Delfin Technologies). TEWL increased over time from day 0 to day 10 after initial TNCB application. The KRG group (48.02 ± 1.26 g/ m 2 h) exhibited significantly reduced TEWL compared with the AD control group (60.17 ± 5.29 g/m 2 h) on day 10 (P = 0.049; Fig. 3D). EPO and cyclosporine treatments did not have any significant effect on TEWL increase. No significant differences were observed in TEWL between the scratching-permitted and scratching-restricted subgroups. Histopathologic analysis of KRG treatment on TNCBinduced AD-like skin lesions After repeated application of TNCB, excessive epidermal hyperplasia, hyperkeratosis, epidermal parakeratosis, marked leukocyte infiltration in the dermis, and dermal edema were observed on the backs and ears of mice in the AD control groups. Oral administration of KRG, EPO, and cyclosporine effectively reduced these epidermal changes and dermal infiltrations (Fig. 4). The KRG (19.67 ± 1.71) and cyclosporine (18.00 ± 2.27) groups exhibited significantly reduced mast cell infiltration on the back compared with the AD control group (28.00 ± 2.50; P = 0.043 and P = 0.011; Table 1). EPO mice (20.67 ± 0.80) exhibited less mast cell infiltration on the back compared with the AD control group, but this was not significant (P = 0.092). Cyclosporine was more effective for reducing mast cell infiltration than either KRG or EPO, but this difference was not significant (P = 0.973 and P = 0.868, respectively). No significant differences were observed between scratching-permitted and scratching-restricted subgroups. Effects of KRG on TNCB-induced upregulation of TNF-α, IFN-γ, TSLP, and IL-31 mRNA expression in AD-like skin lesions TNF-α is a proinflammatory cytokine that initiates inflammation in the acute phase of AD by recruiting inflammatory cells to the lesion (11). mRNA expression of TNF-α was reduced by KRG (7.47 ± 3.35), EPO (16.88 ± 7.98), and cyclosporine (11.50 ± 5.45) treatment compared to the AD control group (22.84 ± 3.66). KRG treatment significantly suppressed mRNA expression of TNF-α compared to AD control group (P = 0.011; Table 1). TNF-α mRNA expression was lower in the scratching-restricted subgroups, although these differences were not significant. TSLP is produced by keratinocytes and activates DCs to stimulate naïve T cells to produce Th2 cytokines (12). TSLP mRNA expression was reduced by EPO treatment (8.64 ± 4.00) compared with the AD control group (35.89 ± 4.98) to a greater extent than it was by KRG (19.30 ± 7.09) and cyclosporine (21.97 ± 15.75) treatments, but these differences were not significant (P = 0.182; Table 1). TSLP mRNA expression tended to be lower in the scratching-restricted subgroups, although this difference was only significant in the AD control group (46.25 ± 3.44 vs. 25.53 ± 2.20, P = 0.007). KRG and EPO prevents the elevation of systemic IgE levels in mice with TNCB-induced AD Decreased systemic IgE elevation was observed in the KRG (102.90 ± 8.79), EPO (109.51 ± 3.97), and cyclosporine (132.23 ± 9.08) treatment groups compared with the AD control group (168.30 ± 9.98; P < 0.001, P < 0.001, and P = 0.002, respectively; Table 1). The KRG treatment more effectively suppressed systemic IgE elevation than cyclosporine treatment (P = 0.043). No significant differences were observed between any of the scratchingpermitted and -restricted subgroups. KRG effectively prevents the elevation of systemic IL-31 levels in mice with TNCB-induced AD IL-31 is mainly produced by Th2 cells, and some evidence indicates that IL-31 production correlates with the severity of pruritus and inflammation in the context of AD (13,15). We evaluated serum IL-31 levels using ELISA and IL-31 mRNA expression levels in cutaneous tissue via RT-PCR. The increase in systemic IL-31 was significantly reduced by oral administration of KRG (1,524.181 ± 75.490) and EPO (1,602.99 ± 113.67) compared with the AD control group (2,241.29 ± 80.41; P < 0.001 and P = 0.001, respectively), although cyclosporine (2,041.25 ± 147.39) did not have a significant effect (Table 1). KRG and EPO controlled serum IL-31 elevation more effectively than cyclosporine (t-test, P = 0.010 and P = 0.035, respectively), The overall systemic IL-31 levels in the scratching-restricted subgroups were lower than those in the scratching-permitted subgroups, but this difference was not significant. DISCUSSION A range of studies have shown the diverse biological effects of Panax ginseng (5)(6)(7)(8)(9)16). We previously examined the anti-inflammatory effects of systemic administration of KRG on a TNCBinduced model of AD in NC/Nga mice and found that KRG pre- vented the development of acute AD-like lesions and late-onset chronic AD-like lesions (7,8). As an extension of this previous research, we evaluated the anti-pruritic effect as well as the anti-inflammatory effect of systemic KRG in a murine model of AD. Clinical severity score, number of scratching movements, ear thickness, and TEWL were measured to evaluate the macroscopic effects of KRG, EPO, and cyclosporine. Systemic administration of KRG, EPO, and cyclosporine significantly reduced TNCBinduced clinical signs and symptoms. KRG was more effective than EPO on day 7 and more effective than EPO and cyclosporine on day 10. Compared to the AD control group, KRG and cyclosporine mitigated ear swelling and ear thickness up to day 7. These results indicate the clinical anti-inflammatory effects of systemic KRG. The KRG group exhibited significantly less TEWL elevation compared with the AD control group on day 10. This was consistent with our previous study (8). Cho et al. (8) showed that 6 hours after TNCB application, KRG-and cyclosporine-treated groups had lower TEWL values compared to the AD control group. In this study, we observed a correlation between scratching movements and TEWL changes. KRG, EPO, and cyclosporine treatment all reduced the number of scratching movements, with KRG being the most at day 10. Thus, we presumed that TEWL changes correlated with protection against mechanical stimuli (scratching) and anti-the inflammatory effects of systemic KRG. Samukawa et al. (16) demonstrated that topical application of tacrolimus and red ginseng extract (RGE) significantly inhibited scratching behavior in a DNFB induced AD mouse model, whereas dexamethasone did not. Topical RGE also significantly reduced DNFB-induced nerve growth factor expression and nerve fiber extensio. In addition, topical application of RGE suppressed histamine-induced scratching behavior (16). In the present study as well as our previous studies, KRG was administered via a systemic method, not topically. Our previous demonstrated that oral administration of KRG suppresses the expression of protein gene product 9.5 and substance P in nerve fibers (7). In this study, the systemic KRG, EPO, and cyclosporine groups all exhibited reduced scratching behavior compared with the AD control group after initial TNCB challenge. KRG suppressed scratching behaviors more effectively than EPO on day 7 and more effectively than EPO and cyclosporine on day 10. This finding suggests that systemic administration of KRG has an anti-pruritic effect in this mouse model of AD. In the present study, we divided each study group into 2 subgroups to evaluate the clinical and systemic effects of the itching-scratching cycle in our AD mouse model. Clinical severity scores were higher in the scratching-permitted subgroups on day 10, especially in the KRGand EPO-treated groups. These results imply that the itchingscratching cycle exacerbates skin lesions in AD and restriction of scratching behavior might be an effective approach for preventing this vicious cycle in patients with AD. In the present study, histopathological analysis also revealed that KRG administration markedly reduced the extent of epidermal hyperplasia, hyperkeratosis, parakeratosis, and dermal infiltration of leukocytes. In addition, KRG and cyclosporine treatment reduced mast cell infiltration in the back and cyclosporine treatment was most effective. This finding is consistent with a previous study reporting that cyclosporine inhibits mast cell degranulation and the release of inflammatory mediators (17). In AD, mast cells are sensitized by IgE and modulate the differentiation of naïve T cells and induce Th2 polarization (11). Elevated levels of IgE autoantibodies also correlated with early onset of disease, intense pruritus, and bacterial infection relapse (18,19). TNCB -induced elevation of serum IgE was also significantly lowered by KRG, EPO and cyclosporine treatment. KRG was more effective than cyclosporine in suppressing total IgE elevation. These results were suggesting that systemic administration of KRG has anti-allergic effects and oral KRG might reduce the severity of AD lesions and prevent progression to respiratory atopy. TNF-α is one of proinflammatory cytokines and responses to environmental damage and inflammation (20). Systemic administration of KRG significantly suppressed TNF-α mRNA expression compared to the AD control group. Kim et al. (6) showed that topical KRG as well as 0.05% betamethasone significantly suppressed TNCB-induced TNF-α mRNA expression in NC/Nga mice. Other previous studies showed that systemic KRG treatment also effectively suppressed TNF-α mRNA expression in atopic mouse model (7,8). Thus, systemic administration of KRG was effective to reduce production of TNF-α from keratinocyte. EPO inhibited TSLP mRNA expression most effectively, although this was not significant. This result is similar with our previous study, which showed that oral administration of EPO effectively reduced total IgE level and mRNA expression of TSLP (8). EPO is a natural source of gamma-linolenic acid (GLA), and GLA is converted to dihomo-GLA (DGLA) in humans. Dietary DGLA supplementation induced prostaglandin D1 (PGD1) production by mast cells in a mouse study (21). PGD1 suppressed IgE mediated degranulation and TSLP gene expression in keratinocytes and topical application of PGD1 also reduced scratching behavior in an AD mouse model (21). These results confirm those of our previous study and this study, in that oral administration of EPO was effective for reducing clinical symptoms, scratching movements, serum IgE, and IL-31 levels in a mouse model of AD. KRG and EPO significantly reduced the serum level of IL-31 compared to the AD control group. IL-31 secretion was similar to the observed scratching patterns (14), suggested that KRG and EPO exerted more potent anti-pruritic effects, as demonstrated by their dampening of the scratching behavior and systemic IL-31 levels compared with cyclosporine. Oral administration of cyclosporine also had an anti-pruritic effect on day 7, but it did not significantly alter the serum IL-31 level. We hyhttps://doi.org/10.3346/jkms.2017.32. 4.679 pothesize that the anti-pruritic effect of cyclosporine is mainly due to its suppression of histamine release by mast cells. Since the mechanism of action of KRG is different from immunosuppressants such as cyclosporine, concurrent use of KRG and cyclosporine could be synergistic. Systemic IL-31 levels in the scratching-restricted subgroups were lower than those in the scratching-permitted subgroups. This finding suggests that scratching behavior and serum IL-31 levels may have a cyclic feed-forward relationship. There are several limitations to our study. First, this is a pilot study evaluating the anti-pruritic and anti-allergic effects of systemic KRG for AD. Thus, future animal studies with larger sample sizes and human research will be needed to confirm these findings. Second, we focused on mouse scratching behavior than previous our study, and used a novel chamber for scratching restriction. There have been several observational studies about scratching behavior in mice, but there have been few experiments using scratching restriction. In addition, we used indirect values such as scratching movement counts and serum IL-31 levels for objective evaluation of pruritus, which have been used in prior studies (13)(14)(15)22). To evaluate the role of mechanical injury due to scratching in AD and its relation to anti-pruritic effects in humans, further studies are needed. In summary, we confirmed the anti-pruritic and anti-allergic effects of KRG in a NC/Nga mice model. KRG decreased clinical signs and symptoms and the number of scratching movements; the mechanism of these effects may be a reduction in mast cell infiltration and/or suppression of systemic IL-31 and IgE. Second, KRG protected skin barrier function as indicated by TEWL elevation. Third, KRG exert anti-inflammatory effects by reducing the mRNA levels of TNF-α and by histopathologic findings such as a markedly diminished extent of epidermal hyperplasia, hyperkeratosis, parakeratosis, and dermal infiltration of leukocytes and mast cells. Thus, proactive systemic administration of KRG may effectively reduce early allergic sensitization and suppress the Th2 response. Restriction of scratching in the early stages could also help prevent the itching-scratching cycle, thus reducing clinical and systemic inflammation.	2017-10-04T08:00:57.342Z	2017-01-31T00:00:00.000	{ "year": 2017, "sha1": "b1f6330a374c222dec1c4efc3b24503a116db07d", "oa_license": "CCBYNC", "oa_url": "https://doi.org/10.3346/jkms.2017.32.4.679", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "b1f6330a374c222dec1c4efc3b24503a116db07d", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
21462936	pes2o/s2orc	v3-fos-license	Adenosine 2B Receptor Expression Is Post-transcriptionally Regulated by MicroRNA We have reported that epithelial adenosine 2B receptor (A2BAR) mRNA and protein are up-regulated in colitis, which we demonstrated to be regulated by tumor necrosis factor α (TNF-α). Here, we examined the mechanism that governs A2BAR expression during colitis. A 1.4-kb sequence of the A2BAR promoter was cloned into the pFRL7 luciferase vector. Anti-microRNA (miRNA) was custom-synthesized based on specific miRNA binding sites. The binding of miRNA to the 3′-untranslated region (UTR) of A2BAR mRNA was examined by cloning this 3′-UTR downstream of the luciferase gene in pMIR-REPORT. In T84 cells, TNF-α induced a 35-fold increase in A2BAR mRNA but did not increase promoter activity in luciferase assays. By nuclear run-on assay, no increase in A2BAR mRNA following TNF-α treatment was observed. Four putative miRNA target sites (miR27a, miR27b, miR128a, miR128b) in the 3′-UTR of the A2BAR mRNA were identified in T84 cells and mouse colon. Pretreatment of cells with TNF-α reduced the levels of miR27b and miR128a by 60%. Over expression of pre-miR27b and pre-miR128a reduced A2BAR levels by >60%. Blockade of miR27b increased A2BAR mRNA levels by 6-fold in vitro. miR27b levels declined significantly in colitis-affected tissue in mice in the presence of increased A2BAR mRNA. Collectively, these data demonstrate that TNF-α-induced A2BAR expression in colonic epithelial cells is post-transcriptionally regulated by miR27b and miR128a and show that miR27b influences A2BAR expression in murine colitis. There are 4 adenosine receptors: A 1 AR, A 2A AR, A 2B AR, and A 3 AR. A 2B AR 3 is the predominant adenosine receptor that is expressed in human colonic epithelial cells and colonic epithelial-derived cell lines, such as T84 (1). In intestinal epithelial cells, A 2B AR couples with G␣ s and activates adenylate cyclase. In the colonic epithelium, A 2B AR regulates vectorial electrogenic ion secretion, a secretory pathway that results in movement of isotonic fluid into the lumen (2)(3)(4). Up-regulation of ion secretion during inflammation is believed to be an impor-tant process in inflammation-associated diarrhea (5,6). In addition to its effect on ion transport, A 2B AR mediates luminal secretion of interleukin-6 by polarized colonic epithelial cells, which results in neutrophil activation (7). Similarly, adenosine induces apical secretion of fibronectin, which potentiates the adhesion and invasion of Salmonella typhimurium through chemokine secretion (8). We have observed that A 2B AR mRNA and protein expression are up-regulated in human and animal models of inflammatory bowel disease (9). We have demonstrated that deletion of the A 2B AR gene protects against murine colitis (10). In addition, A 2B AR antagonists mitigate colonic inflammation in murine models of colitis (11), suggesting that A 2B AR is a therapeutic target in inflammatory bowel disease and other intestinal inflammatory diseases. In this study, we examined the regulation of A 2B AR expression in the colon by TNF-␣, a critical proinflammatory cytokine that plays an important role in the pathogenesis of inflammatory bowel disease and whose antagonism constitutes an effective therapeutic strategy for inflammatory bowel disease (12). Prediction of miRNA Targets for A 2B AR Gene-To identify the targets of human miRNAs for the A 2B AR receptor, we used PicTar software, an algorithm that identifies miRNA targets. The miRNA target database contains all animal miRNA sequences from miRBase. The sequence database was scanned against the 3Ј-untranslated regions (3Ј-UTRs) that have been predicted in all available species in Ensemble. The interface links each miRNA to a list of predicted gene targets. A 2B AR has four putative miRNA target sites (miR27a, miR27b, miR128a, and miR128b) in the 3Ј-UTR of A 2B AR mRNA. Fig. 3 shows the binding sites and target miRNA alignments for the A 2B AR 3Ј-UTR. Cloning of 3Ј-UTR of A 2B AR with Predicted miRNA Target Sequences-The 402-bp 3Ј-UTR of A 2B AR, encompassing the predicted miRNA sequences, was inserted into the multiple cloning site of the pMIR-REPORT luciferase miRNA expression reporter (Applied Biosystems catalogue no. AM5795) with HindIII and SpeI. miRNA Reverse Transcription-Total RNA was extracted using TRI reagent (MRC, Inc.). Reverse transcription was performed per the Invitrogen Ncode miRNA first-strand cDNA module kit protocol. Nuclear Run-on Assay-Nuclear run-on assay was performed according to the protocol of Dr. Yan (14). Briefly, nuclei were isolated from T84 cells that were untreated or treated with 10 ng/ml TNF-␣ for 2-4 h. Cells (5 ϫ 10 7 ) were pelleted in a test tube and washed twice with ice-cold phosphate-buffered saline. Cells were then lysed on ice for 10 min in lysis buffer, containing 0.3 M sucrose, 0.4% (v/v) Nonidet P-40, 10 mM Tris-HCl at pH 7.4, 10 mM NaCl, and 3 mM MgCl 2 . After centrifugation (15 min at 500ϫ g), the nuclear pellet was resuspended and subjected to an additional 5-min lysis to remove any remaining intact cells. Following centrifugation, the nuclei were purified by centrifugation through a 2.0 M sucrose cushion. The nuclei were resuspended in 300 l of transcription buffer (50 mM Tris-HCl (pH 8.0), 150 mM KCl, 5 mM MgCl 2 , 0.5 mM MnCl 2 , 1 mM dithiothreitol, 0.1 mM EDTA, 10% glycerol). After sequential pretreatment with 1 l of 50 g/ml RNase A and 2.5 l (100 units) of RNasin (Promega, Madison, WI), the in vitro elongation reaction was initiated with 0.25 mM each ATP, GTP, CTP, and UTP. The reaction proceeded for 25 min at 30°C. After incubation with RNase-free DNase, RNA was extracted with phenol-chloroform, precipitated with ammonium acetate and isopropyl alcohol, washed with 70% ethanol, and dissolved in water. cDNA was synthesized with the Fermentas First-strand synthesis system and amplified by realtime PCR using A 2B AR-specific primers and 18S as an internal control. Real-time PCR-Real-time PCR was performed using the iCycler sequence detection system (Bio-Rad). Briefly, 50 ng of reverse-transcribed cDNA was amplified (40 cycles of 95°C for 15 s and 55°C for 1 min), using 10 M gene-specific primers. Experimental Animals-The Institutional Animal Care and Use Committee of Emory University approved all procedures that were performed on the animals. In addition, all experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, published by the U.S. Public Health Service. Six-week-old C57BL/6 mice were purchased from Jackson Laboratories (Bar Harbor, ME) and bred at our facility. They were maintained on a 12-hour light-dark cycle and allowed free access to pelleted diet and tap water at controlled temperatures (25 Ϯ 2°C). Induction of DSS-Colitis-Colitis was induced in male C57BL/6 mice by ad libitum oral administration of 3% (w/v) DSS in tap water for 6 days. Age-matched male wild-type C57BL/6 mice that received tap water served as controls. Mice were observed daily and evaluated for changes in body weight and the development of clinical symptoms. Transient Transfection of Pre-miR and Anti-miR-To study the regulation of A 2B AR by miRNA, pre-miR27b and/or pre-miR128a (A 2B AR target miRNAs) and a plasmid that contained the associated miR27b and miR128a binding sites in A 2B AR (pMIR-3Ј-UTR) were co-transfected into subconfluent Caco2-BBe cells. The binding of miRNAs to A 2B AR 3Ј-UTR mRNA was reflected by pMIR-REPORT luciferase activity. Anti-miR27b and/or anti-miR128a was introduced into HT29 cells to study the blockade of miR27b and miR128a. Endogenous A 2B AR mRNA levels were measured by real-time PCR 48 h after transfection. The positive control for blocking efficiency was anti-miR let 7c, which regulates high mobility group A2 (HMGA-2) mRNA. cAMP Measurement-HT29-Cl.19A cells were washed and equilibrated with Hanks' solution; then, the cells were untreated or treated with 100 M adenosine (basolateral). cAMP measurements were made in whole cell lysates in the presence of 1 mM 3-isobutyl-1-methylxanthine using a competitive cAMP immunoassay kit according to the manufacturer's protocol. Luminescence was read on a Luminoscan Ascent (Thermo Labsystems, Needham Heights, MA). To study the suppression of endogenous miRNA function, anti-miR27b and anti-miR128a (Ambion AM17000, 17100) were introduced into HT29 cells, and A 2B AR signaling was measured by estimating the levels of adenosine-mediated cAMP production with and without transfection of the blockers. Statistical Analysis-All data are expressed as means Ϯ S.E. Statistical analysis was performed by Student's t test or analysis of variance (Prism 4.0; GraphPad, San Diego, CA). A p value Ͻ 0.05 was regarded as significant. TNF-␣ Does Not Affect A 2B AR Promoter Activity-Our previous studies demonstrated that A 2B AR mRNA and protein were up-regulated in cells that were pretreated with TNF-␣. To understand better the regulation of A 2B AR, we performed experiments using a luciferase reporter construct that contained the A 2B AR promoter (pFRL7 1.4A 2B AR). The 5Ј-flanking region of A 2B AR was obtained from Homo sapiens chromosome 17S from the PAC (P1-derived artificial chromosome) RPCI-1 121M24 map 17p11 (a gift from Dr. R. Reinhardt, Max Planck Institute for Molecular Genetics, Berlin, Germany) by PCR, yielding a 1.4-kb fragment of the region. The human A 2B AR promoter sequence was submitted to NCBI GenBank under accession number DQ 351340. The isolated 1.4-kb and 2.5-kb fragments were subsequently cloned into the pFRL7 luciferase vector (13) at the KpnI and NcoI restriction sites to yield the pFRL7 1.4 A 2B AR plasmid. The transcriptional start site, as determined by rapid amplification of cDNA ends using the Gene Racer kit (Invitrogen), was 119 bp upstream of the translational start site. COS-7 or Caco2-BBe cells were transfected with pFRL7 1.4A 2B AR and treated 72 h later with TNF-␣ for 6 h. As shown in Fig. 1, the relative luciferase activity in pFRL7 1.4 A 2B AR total lysates (11.6 Ϯ 1.3) was ϳ10 times greater than that in empty vector transfectants (1.9 Ϯ 0.13). TNF-␣, however, did not increase A 2B AR promoter activity (11.01 Ϯ 0.3), despite the increase in A 2B AR mRNA levels (9). Also, TNF-␣ had no effect on luciferase activity from a longer 2.5-kb fragment of the A 2B AR promoter. TNF-␣ Does Not Increase mRNA Transcription-To clarify the effect of TNF-␣ on A 2B AR transcription, we performed nuclear run-on assays using nuclei from vehicle-or TNF-␣-treated T84 cells. After in vitro elongation with dNTPs, the RNA was extracted, and novel A 2B AR mRNA synthesis was quantified by real-time PCR using A 2B AR-specific primers. There was no significant difference in A 2B AR mRNA levels between vehicle-and TNF-␣ treated nuclei. These data indicate that TNF-␣ does not affect A 2B AR transcription and suggest that TNF-␣ regulates A 2B AR by a post-transcriptional mechanism. This experiment was repeated three times, each yielding similar results. TNF-␣ Down-regulates miR27b and miR128a-One of the primary mechanisms by which gene expression is post-transcriptionally regulated is miRNA. We determined whether A 2B AR was regulated by miRNA. First, we performed an online search of miRNAs using the pictar and targetscan databases and identified four putative miRNA target sites (miR27a, miR27b, miR128a, and miR128b) in the 3Ј-UTR of A 2B AR mRNA (Fig. 2). All four miRNAs were expressed in T84 cells. To determine the effect of TNF-␣ on miRNA and A 2B AR mRNA levels, we pretreated T84 cells for various times with TNF-␣ (10 ng/ml). Pretreatment of T84 cells with TNF-␣ increased A 2B AR mRNA levels by ϳ35-fold, (p Ͻ 0.03, n ϭ 6) (Fig. 3A) compared with untreated cells. Pretreatment with TNF-␣ reduced the levels of miR27b and miR128a by 60% (p Ͻ 0.001, n ϭ 6) (Fig. 3B), whereas miR27a and miR128b levels were unaffected (data not shown). These data establish a direct correlation between A 2B AR mRNA and the miR27b and miR128a miRNAs. Pre-miR27b and -miR128a and Anti-miR27b and -miR128a Regulate A 2B AR mRNA Levels and Function-Pre-miRNA precursors are small, partially double-stranded RNAs that mimic endogenous precursor miRNAs. To examine the regulation of A 2B AR by miR27b and miR128a, we transfected T84 cells with pre-miR27b or pre-miR128a. Forty-eight hours after transfection, the cells were harvested, and A 2B AR mRNA levels were determined by real-time PCR. As shown in Fig. 4A, transfection of pre-miR27b or pre-miR128a decreased A 2B AR mRNA levels by 50% compared with untransfected cells (pre-miR27b 43.4 Ϯ 4.83, p Ͻ 0.001; pre-miR128a 30.3 Ϯ 6.1, p Ͻ 0.003), whereas 18S mRNA (control) was unaffected by pre-miR27b or pre-miR128a transfection. A commercial negative control pre-miRNA, containing a nontargeting sequence that had no homology to A 2B AR, had no effect on A 2B AR mRNA levels. These results demonstrate that miR27b and miR128a influence A 2B AR mRNA levels. Next, we sought to determine the effect of inhibition of miR27b or miR128a on A 2B AR mRNA. Cells were transfected with custom-synthesized blockers of the miR27b or miR128a binding sites, and A 2B AR mRNA levels were measured by realtime PCR. As shown in Fig. 4B, anti-miR27b and anti-miR128a increased A 2B AR mRNA levels by 5.3 Ϯ 0.9-fold (p Ͻ 0.01) and 2.2 Ϯ 0.02-fold (p Ͻ 0.005), respectively, compared with vector-transfected cells. 18S mRNA did not change, demonstrating the specificity of the miR blockers for A 2B AR mRNA. These data support the results of the pre-miRNA experiments and further demonstrate that miR27b and miR128a regulate A 2B AR mRNA levels. To determine whether the effects of miR27b and miR128a on A 2B AR mRNA levels were reflected in functional alterations in A 2B AR, we measured adenylate cyclase activity in HT29-Cl.19A cells that were transfected with anti-miR27b or anti-miR128a. Forty-eight hours after transfection, cells were treated FIGURE 3. TNF-␣ pretreatment down-regulates miR27b and miR128a levels. A, T84 cells were pretreated with or without TNF-␣. Cells were harvested after 2 h, and total RNA was extracted from the cells. A 2B AR mRNA levels were determined by real-time PCR. Data represent the mean Ϯ S.E. (error bars) of two independent experiments; n ϭ 6. B, T84 cells were pretreated with or without TNF-␣, and total RNA was extracted. miR27b and miR128a levels were measured using specific primers and normalized to 5.8S. Data represent mean Ϯ S.E. of two independent experiments; n ϭ 6. basolaterally with adenosine (100 M), and cAMP levels were measured. As shown in Fig. 4C, cAMP levels were significantly higher in miR27b (0.07 Ϯ 0.01, p Ͻ 0.02) and miR128a transfectants (0.08 Ϯ 0.01, p Ͻ 0.016) compared with untransfected cells (0.04 Ϯ 0.004), suggesting that the up-regulation of A 2B AR mRNA in anti-miR27b-and anti-miR128a-transfected cells effects an increase in A 2B AR levels. Elevations in A 2B AR mRNA Levels during Colitis Correlate with Decreased miR27b Levels-We have shown that A 2B AR mRNA and protein levels are increased during colitis. To deter-mine whether miR27b or miR128a levels correlate inversely with A 2B AR mRNA during colitis, we measured miR27b, miR128a, and A 2B AR mRNA levels by real-time PCR in a DSS model of colitis. All mice that were fed 3% DSS for 7 days developed signs of severe colitis (clinical score: 9.4 Ϯ 1.0; histological score: 9.1 Ϯ 2.3), presented as bloody diarrhea and weight loss. Consistent with our previous observations (9), A 2B AR mRNA levels increased 2.3 Ϯ 0.3-fold (p Ͻ 0.05), compared with control mice that received water (Fig. 5A). Notably, miR27b levels decreased 33.4 Ϯ 6.6fold (p Ͻ 0.005) in colitic mice compared with control mice that received water (Fig. 5B). There was no change in miR128a levels (data not shown). These data demonstrate that there is an inverse correlation between miR27b and A 2B AR mRNA, supporting evidence for a post-transcriptional mechanism of A 2B AR regulation by miRNA. Pre-miR27b and Pre-miR128a Decrease A 2B AR 3Ј-UTR Luciferase Activity-To confirm that miRNA functions through an interaction with the A 2B AR 3Ј-UTR, we co-transfected pre-miR27b and/or pre-miR128a with the A 2B AR 3Ј-UTR luciferase reporter construct pMIR-REPORT. As shown in Fig. 6, A and B, introduction of pre-miR27b or pre-miR128a significantly inhibited 3Ј-UTR-dependent A 2B AR luciferase activity in a dosedependent manner (25-100 nM, n ϭ 8, *p Ͻ 0.0001). Further, co-transfection of pre-miR27b (50 nM) and pre-miR128a (50 nM) produced a robust additive inhibitory effect on luciferase activity (Fig. 6C). We performed the luciferase assay with an additional group, in which we co-transfected the A 2B AR 3Ј-UTR and the A 2B AR promoter. We observed no additional increase in luciferase activity beyond that induced by the 3Ј-UTR alone (Fig. 6D), suggesting that A 2B AR expression is regulated by miRNA. DISCUSSION miRNAs are short RNAs in introns and intergeneic regions that regulate gene expression. A large primary miRNA (pre-miRNA) is transcribed and processed to a pre-miRNA hairpin, which is converted in the cytoplasm into a mature miRNA that FIGURE 5. Increased levels of A 2B AR mRNA during colitis correlate with decreased miR27b levels. Ageand gender-matched C57BL/6 mice were fed with 3% DSS for 7 days, after which mice were killed and examined for severity of colitis. Total RNA was extracted from the colons of these mice. A 2B AR mRNA (A) and miR27b (B) levels were measured compared with control mice. Data are represented as A 2B AR mRNA normalized to 18S, and miR27b is normalized to SnoRNA55. Data are represented as mean Ϯ S.E. (error bars) of two independent experiments; n ϭ 3. WT, wild type. FIGURE 6. Pre-miR27b and pre-miR128a decrease A 2B AR 3-UTR luciferase activity, and addition of A 2B AR promoter to 3-UTR does not influence luciferase activity. The 3Ј-UTR of the A 2B AR-predicted miRNA binding target was inserted into the multiple cloning site of the pMIR-REPORT luciferase miRNA expression reporter vector. Pre-miR27b and pre-miR128a were co-transfected with A 2B AR 3Ј-UTR pMIR-REPORT. Dose-dependent effect of pre-miR27b (A), dose-dependent effect of pre-miR128a (B), and additive effect of pre-miR27b and pre-miR128a (C) were demonstrated by luciferase assay. Data are represented as mean Ϯ S.E. of two independent experiments; n ϭ 12. D, transfection of 3Ј-UTR of A 2B AR alone, pFRL7 1.4 A 2B AR promoter alone, or 3Ј-UTR ϩ pFRL7 1.4 A 2B AR promoter together were transfected into Caco2-BBe cells with or without TNF-␣ treatment (20 ng/ml) for 6 h. Cells were lysed, and luciferase activity was measured. Data represent mean Ϯ S.E. (error bars); n ϭ 6. targets genes that harbor complementary sites in their 3Ј-or 5Ј-UTRs (18,19). miRNAs are up-and down-regulated in several disease conditions. We have demonstrated that A 2B AR protein levels increase after treatment with proinflammatory cytokines but that A 2B AR promoter activity is unaltered. We also examined miRNAs in bioinformatics databases that were complementary to the A 2B AR 3Ј-UTR to determine whether miR27b and miR128a regulated the levels of A 2B AR mRNA and protein. In our previous studies, we observed the up-regulation of A 2B AR by TNF-␣ in colonic epithelial cells and in the DSS mouse model of colitis (9). Using a pFLR7 promoter-reporter construct that contained either the 1.4-kb and 2.5-kb fragments of the A 2B AR promoter, we found that TNF-␣ did not increase A 2B AR promoter activity (Fig. 1). A 2B AR, however, has been shown to be transcriptionally regulated during hypoxia through hypoxia inducible factor (HIF)-binding elements (20) and by b-myb (21) in endothelial cells and vascular smooth muscle cells, respectively. Our luciferase assays were performed with whole cell lysates, which should have harbored all of the transcription factors that regulate the promoter. Studies by our laboratory and others have noted the transcriptional regulation of genes by TNF-␣ by 4 -6 h, suggesting that the transcriptional mechanism is active in our luciferase assay (22)(23)(24). Similarly, by nuclear run-on assay, we observed no increase in A 2B AR transcription in response to TNF-␣. Although the involvement of other cis-or trans-elements and signaling pathways that regulate A 2B AR transcription further upstream of the 2.5-kb element cannot be ruled out with these assays, these data strongly suggest that TNF-␣ up-regulates A 2B AR through a post-transcriptional mechanism. Thus, we speculated that A 2B AR regulation was mediated by A 2B AR-specific miRNAs. A search of bioinformatics databases identified miR27b and miR128a as containing target sequences in the 3Ј-UTR of A 2B AR. Treatment of colonic epithelial cells with TNF-␣ inhibited miR27b and miR128a by 60%, which was associated with a 5-fold increase in A 2B AR mRNA levels (Fig. 4, A and B). Colonic tissue from mice with DSS-induced colitis experienced a 50% reduction in miR27b levels (Fig. 5B). These findings implicated miR27b and miR128a in the regulation of A 2B AR mRNA. Our studies were unable to determine the relative contribution of miR27b and miR128a because the consensus binding sites for them in the A 2B AR 3Ј-UTR overlap significantly (Fig. 2). miRNAs constitute a prominent class of regulatory genes in animals, representing ϳ1% of all predicted genes, suggesting that sequence-specific, post-transcriptional regulatory mechanisms that are mediated by these small RNAs are ubiquitous (25)(26)(27). We tested this hypothesis using synthetic pre-miRNAs (miRNA precursor molecules) and a reciprocal set of anti-miRNAs. Pre-miRNAs are small, chemically modified, double-stranded RNA molecules that mimic endogenous mature miRNA molecules. These molecules are incorporated into the RNA-induced silencing complex, which is also responsible for miRNA activity. In contrast to miRNA expression vectors, pre-miRNAs can be used in dose-response studies because they can be introduced directly into the cell by transfection. Inside cells, pre-miRNAs up-regulate miRNA activity (28), enabling one to analyze miRNA function. Transfection of pre-miRNA reduced A 2B AR mRNA levels in T84 cells (Fig. 4A). In reciprocal experiments, blockade of miR27b and miR128a, as a result of the introduction of specific anti-miRNAs, increased A 2B AR mRNA levels (Fig. 4B). Anti-miRNAs also enhanced adenosine-mediated cAMP production (Fig. 4C), indicating an increase in A 2B AR function and directly linking miR27b and miR128a to the regulation of A 2B AR expression (29). miRNAs have important functions in the regulation of genes. Several studies have demonstrated that miRNAs govern basic cellular functions. Studies by Care et al. (30,31) noted that miR133 and miR1 regulate cardiomyocyte hypertrophy. In our experiments, co-transfection of pre-miR27b or pre-miR128a with an A 2B AR 3Ј-UTR reporter construct increased luciferase activity in a dose-dependent manner (Fig. 6, A and B). Co-transfection of pre-miR27b and pre-miR128a resulted in the additive inhibition of luciferase activity (Fig. 6C). By luciferase assay, co-tranfection of the 3Ј-UTR of A 2B AR and the A 2B AR promoter resulted in no additional increase in luciferase activity beyond that induced by 3Ј-UTR alone, suggesting that miRNA plays the significant role in the regulation of A 2B AR expression (Fig. 6D). Recent bioinformatic and experimental evidence suggests that a substantial proportion of genes (Ͼ30%) are subject to miRNA-mediated regulation (32). Our studies on the possible function of miRNAs in colonic inflammation demonstrate that miR27b and miR128a are critical regulators of A 2B AR expression. Because A 2B AR antagonism ameliorates colitis, our results suggest that the regulation of A 2B AR expression by targeting miRNAs might be an effective therapeutic strategy for colitis.	2018-04-03T02:01:10.393Z	2010-04-13T00:00:00.000	{ "year": 2010, "sha1": "eed5adafe528ed0567af65d10346e2f1b9c631a6", "oa_license": "CCBY", "oa_url": "http://www.jbc.org/content/285/24/18184.full.pdf", "oa_status": "HYBRID", "pdf_src": "Highwire", "pdf_hash": "89a9e264bf4601ce3026f7603dbf799b526dbf6d", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
7965679	pes2o/s2orc	v3-fos-license	Regulation of Yersinia Protein Kinase A (YpkA) Kinase Activity by Multisite Autophosphorylation and Identification of an N-terminal Substrate-binding Domain in YpkA* Background: The catalytic mechanism of the Yersinia protein kinase YpkA is poorly understood. Results: Multiple N-terminal autophosphorylation sites regulate YpkA activation and residues 40–49 of YpkA contribute to Gαq binding and phosphorylation. Conclusion: The N-terminal domain of YpkA plays a role in autophosphorylation and substrate binding. Significance: Elucidating how type III bacterial effectors are regulated is essential to our understanding of infectious diseases. The serine/threonine protein kinase YpkA is an essential virulence factor produced by pathogenic Yersinia species. YpkA is delivered into host mammalian cells via a type III secretion system and localizes to the inner side of the plasma membrane. We have previously shown that YpkA binds to and phosphorylates the α subunit of the heterotrimeric G protein complex, Gαq, resulting in inhibition of Gαq signaling. To identify residues in YpkA involved in substrate binding activity we generated GFP-YpkA N-terminal deletion mutants and performed coimmunoprecipitation experiments. We located a substrate-binding domain on amino acids 40–49 of YpkA, which lies within the previously identified membrane localization domain on YpkA. Deletion of amino acids 40–49 on YpkA interfered with substrate binding, substrate phosphorylation and substrate inhibition. Autophosphorylation regulates the kinase activity of YpkA. To dissect the mechanism by which YpkA transmits signals, we performed nano liquid chromatography coupled to tandem mass spectrometry to map in vivo phosphorylation sites. Multiple serine phosphorylation sites were identified in the secretion/translocation region, kinase domain, and C-terminal region of YpkA. Using site-directed mutagenesis we generated multiple YpkA constructs harboring specific serine to alanine point mutations. Our results demonstrate that multiple autophosphorylation sites within the N terminus regulate YpkA kinase activation, whereas mutation of serine to alanine within the C terminus of YpkA had no effect on kinase activity. YpkA autophosphorylation on multiple sites may be a strategy used by pathogenic Yersinia to prevent inactivation of this important virulence protein by host proteins. The serine/threonine protein kinase YpkA is an essential virulence factor produced by pathogenic Yersinia species. YpkA is delivered into host mammalian cells via a type III secretion system and localizes to the inner side of the plasma membrane. We have previously shown that YpkA binds to and phosphorylates the ␣ subunit of the heterotrimeric G protein complex, G␣q, resulting in inhibition of G␣q signaling. To identify residues in YpkA involved in substrate binding activity we generated GFP-YpkA N-terminal deletion mutants and performed coimmunoprecipitation experiments. We located a substrate-binding domain on amino acids 40 -49 of YpkA, which lies within the previously identified membrane localization domain on YpkA. Deletion of amino acids 40 -49 on YpkA interfered with substrate binding, substrate phosphorylation and substrate inhibition. Autophosphorylation regulates the kinase activity of YpkA. To dissect the mechanism by which YpkA transmits signals, we performed nano liquid chromatography coupled to tandem mass spectrometry to map in vivo phosphorylation sites. Multiple serine phosphorylation sites were identified in the secretion/translocation region, kinase domain, and C-terminal region of YpkA. Using site-directed mutagenesis we generated multiple YpkA constructs harboring specific serine to alanine point mutations. Our results demonstrate that multiple autophosphorylation sites within the N terminus regulate YpkA kinase activation, whereas mutation of serine to alanine within the C terminus of YpkA had no effect on kinase activity. YpkA autophosphorylation on multiple sites may be a strategy used by pathogenic Yersinia to prevent inactivation of this important virulence protein by host proteins. YpkA is a serine/threonine protein kinase that phosphorylates actin and the small heterotrimeric G protein subunit G␣q (20 -21). The C terminus of YpkA interacts with members of the Rho family of small GTPases, RhoA and Rac1 (22)(23). The deubiquitinating enzyme otubain 1 (OTUB1) was initially identified as a substrate for YpkA; however more recent studies have linked otubain 1 to the RhoGDI domain of YpkA (24 -25). The 729-amino acid YpkA protein is composed of multiple domains (Fig. 1). Residues 1-77 (Sec/Trans) mediate type III secretion and translocation of YpkA into a target cell (26). This region coincides with a chaperone-binding domain (CBD), amino acids 20 -77 (27). Once inside the cell, a membrane localization domain (MLD) consisting of amino acids 20 -90 localizes YpkA to the inner side of the plasma membrane where it is in close proximity to signaling proteins involved in transducing external signals inside the cell (27)(28). Residues of a conserved aspartic acid (Asp-267) and a lysine (Lys-269) residue with alanine results in a catalytically inactive kinase (21). Y. pseudotuberculosis mutant strains expressing catalytically inactive YpkA variants are markedly attenuated in virulence in mouse infection studies (19). In cell culture infection assays, the enzymatic activity of YpkA was necessary for inhibition of host cell bacterial internalization (29 -31). A region within the C-terminal domain (residues 431-612, RhoGDI) of YpkA possesses Rho GTPase binding guanine nucleotide dissociation inhibitor (GDI)-like activity and has been shown to be important for inactivation of the small Rho GTPases, RhoA and Rac1 (32). The GDI-like activity interferes with phagocytosis by disrupting the host actin cytoskeleton (33). Substitution of three amino acids (Y591A, N595A, E599A) in the GDI-like domain interferes with Rho GTPase binding (32). The last 21 amino acids (residues 709 -729) are involved in actin binding and subsequent autoactivation of YpkA kinase activity (21). Residues serine 90 and serine 95 were reported as autophosphorylation sites required for efficient activation and phosphorylation of exogenous substrates by YpkA (30). Both kinase and guanine nucleotide dissociation inhibitor domains of YpkA are important in the activity of full length YpkA (19,(31)(32). The kinase activity of YpkA is dependent on its association with actin (21,30). Although YpkA has been shown to phosphorylate actin and otubain 1 in vitro, the physiological importance of these findings is unclear (21,24,25). We previously reported that YpkA interacts with and phosphorylates the heterotrimeric G protein G␣q, although the involvement of additional components remains to be determined (20). YpkA-mediated phosphorylation of Ser-47 on G␣q impairs guanine nucleotide binding and subsequently inhibits G␣qmediated signaling pathways (20). G␣q belongs to the family of heterotrimeric G proteins that couple with G protein-coupled receptors (GPCRs) to transduce signals from a myriad of extracellular agents and play a central regulatory role in a number of cellular activities (34 -35). G proteins are divided into four families based on sequence similarities of the ␣ subunits: G␣s, G␣i/o, G␣12/13, and G␣q. Members of the G␣s and G␣i families are known to activate and inhibit adenylyl cyclase, respectively. Members of the G␣12/13 family regulate the small G protein RhoA, while G␣q family members stimulate phospholipase C-␤ (PLC-␤), leading to the hydrolysis of phophatidyl-4,5-bisphosphate and the production of inositol triphosphate (IP3) and diaccylglycerol (DAG). In addition, G␣q family members have also been shown to activate RhoA-mediated pathways. The importance of heterotrimeric G protein ␣ subunits in eukaryotic defense responses is underscored by the observation that a number of bacterial pathogens have evolved toxins that specifically target their activity (36). We have previously shown that the N-terminal 430 amino acids of YpkA are essential for substrate binding (20). As a first step toward elucidating the mechanism of substrate recognition mediated by the YpkA N-terminal domain, we have identified residues 40 -49 that are critical for YpkA-mediated inhibition of G␣q signaling. The efficiency of substrate-binding and -phosphorylation by YpkA is diminished by deletion of residues 40 -49 of YpkA, suggesting that they are important for substrate recognition. Trasak et al. proposed a model in which actin binding induces autophosphorylation of YpkA on serine 90 and serine 95 (30). Using an in vivo labeling assay we demonstrated that a YpkA S90A/S95A mutant undergoes autophosphorylation and demonstrates substrate phosphorylation activity, indicating the presence of additional autophosphorylation sites. Here, we report that multiple autophosphorylation sites within the N terminus of YpkA regulate its kinase activity. These findings further our understanding of the molecular mechanism used by Yersinia type III effectors to circumvent host defenses. EXPERIMENTAL PROCEDURES Cell Culture, Transfection, and Reagents-Human embryonic kidney cells (HEK293A) were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum, 0.1 mM non-essential amino acids, and 2 mM L-glutamine. Cells were cultured in a humidified atmosphere of 5% CO 2 at 37°C. The TransIT-LT1 Transfection Reagent (Mirus) or the FuGENE 6 transfection reagent (Roche Molecular Biochemicals) was used according to the manufacturer's recommendations. All reagents were from Fisher Scientific, Sigma-Aldrich, Invitrogen, or New England Biolabs unless otherwise noted. All oligonucleotide primers were from Integrated DNA Technologies. Construction of Plasmids-The Y. enterocolitica YpkA ORF (YopO) was isolated by PCR using the plasmid pYV80811 (a generous gift from James Bliska, The State University of New York at Stony Brook). Full-length YpkA and its various mutants were cloned in-frame into the pEGFP-C3 (Clontech), FLAGtagged pcDNA3.1 (Invitrogen), or GST-tagged pGEX-6P-2 vectors following standard protocols. YpkA internal deletion mutants were generated using the In-Fusion HD Cloning Kit (Clontech) following the manufacturer's instructions. All point mutations were introduced by using the QuikChange II Sitedirected Mutagenesis Kit (Agilent) following the manufacturer's recommendations. All expression constructs were verified by sequencing. Regulation of YpkA Activation by Multisite Autophosphorylation incubated with 0.2% Triton X-100 prior to pelleting the bacterial debris. The GST-tagged protein was affinity-purified on glutathione-Sepharose 4B beads (GE Healthcare) according to the manufacturer's instructions. Bead-bound proteins were pelleted, washed with GST lysis buffer, and eluted with 200 mM reduced glutathione (in GST lysis buffer, pH 7.4). Immunoprecipitation and Immunoblotting-After transfection for 16 -20 h, cells were lysed in modified RIPA buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 1% Triton X-100, 10% glycerol, 10 mM sodium fluoride, and 0.4 mM EDTA) containing protease inhibitors (Halt Protease Inhibitor Mixture; Thermo Scientific) and 1 mM PMSF. The homogenate was centrifuged (14,000 ϫ g, 4°C, 20 min), and the supernatants were incubated with anti-FLAG M2 agarose beads (Sigma) or with the indicated antibodies bound to protein G beads overnight at 4°C with gentle rotation. After incubation, immunoprecipitates were washed extensively with ice-cold modified RIPA buffer. Proteins bound to the beads were eluted by heating at 70°C for 10 min in LDS-PAGE sample loading buffer. The eluted proteins were separated by SDS/PAGE, transferred to a PVDF membrane, and probed with the specified antibodies followed by chemiluminescence detection. Whole cell lysates were separated by SDS/ PAGE and subjected to immunoblotting as described above. The following antibodies were used: anti-actin (20 -33) (Sigma), anti-GFP (JL-8) (Clontech), anti-FLAG-M2 (Sigma), and anti-G␣q (E-17) (Santa Cruz Biotechnology). Generation of G␣q Phosphospecific Antibodies-Polyclonal affinity purified phosphospecific peptide antibodies were generated by 21 st Century Biochemicals (Marlboro, MA). The G␣q peptides used were: TGESGK[pS]TFIKQMC and CGTGES-GK[pS]TFIKQM. To selectively purify antibodies with phosphospecificity from the anti-sera, a two-stage affinity purification was taken. First, each anti-serum was negatively purified by exposure to an affinity column containing the nonphosphorylated peptide. Second, the flow-through from the negative purification was positively purified on a column containing the phosphorylated peptide. Immunofluorescence-Semiconfluent HEK293A monolayers were grown overnight on 22-mm-diameter glass coverslips in DMEM supplemented with 10% fetal bovine serum. Monolayers were transfected with plasmid DNA as described above. For all carbachol (Calbiochem) experiments, HEK293A cells were also transfected with HA-M1 muscarinic receptor (M1R) cDNA expression plasmid to make these cells responsive to carbachol since they do not express endogenous M1Rs. Where indicated cells were stimulated with 200 M carbachol for 60 min. For indirect immunofluorescence, samples were fixed, permeabilized, and stained as previously described (21). Proteins were visualized by direct fluorescence of GFP-or mCherry-containing proteins, or, where indicated, with anti-FLAG M2-Cy3 (Sigma), and the appropriate secondary antibody conjugated to Alexa Fluor 488 (Invitrogen). Nuclear staining was achieved by staining with Hoescht stain (Invitrogen). Rhodamine-phalloidin (Invitrogen) was used to stain the actin cytoskeleton. Images were acquired by epi-fluorescence microscopy with the ϫ60 apochromat objective lens using a Nikon Eclipse 80i fluorescence microscope. For subcellular localization of the YpkA variants an Olympus FV1000 laser scanning confocal microscope was used to image the cells. In Vivo Labeling with [ 32 P]Orthophosphate-Labeling experiments were performed as described previously (20). For all labeling experiments HEK293A cells were cultured and transfected as described above. Twenty-four hours after transfection, the cells were washed with phosphate-free DMEM. Following a 2-h incubation with the same medium containing [ 32 P]orthophosphate (150 Ci/ml), the cells were lysed with TNN lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 2 mM EDTA, and 50 mM NaF) supplemented with Halt Protease Inhibitor Mixture (Thermo Scientific) and 1 mM PMSF. Whole-cell lysates were harvested and split equally. One-half was incubated with anti-FLAG agarose beads. The remaining lysate was used in an immunoprecipitation with a polyclonal antiserum bound to protein A-Sepharose that recognizes G␣q. Bead-bound proteins were washed with lysis buffer, heated in LDS sample buffer at 70°C for 10 min, and separated by SDS-PAGE. Protein phosphorylation was visualized by autoradiography. For plasmid expression, an immunoblot was performed as described above on whole-cell lysates. Densitometry was performed using the ImageJ analysis software (NIH) as per the developer's recommendations. Phosphorylation Site Identification by Nano Liquid Chromatography Tandem Mass Spectrometry-In vivo phosphorylation of FLAG-YpkA was performed as described above in the absence of [␥-32 P]ATP. After purification by affinity chromatography, SDS-PAGE, and Coomassie blue staining, bands corresponding to YpkA were excised from the gel. YpkA was prepared for MS analysis using methods adapted from standard reduction, alkylation, and tryptic digestion procedures (37). Peptides were dried down in a vacuum concentrator after digestion, then resolubilized in 2% acetonitrile/0.1% trifluoroacetic acid for LC-MS/MS analysis. Digested peptides were analyzed by LC-MS/MS on an LTQ-FT with Michrom Paradigm LC and CTC Pal autosampler. Peptides were directly loaded onto a Agilent ZORBAX 300SB C 18 reversed phase trap cartridge, which, after loading, was switched in-line with a Michrom C 18 column connected to the Thermo-Finnigan LTQ-FT mass spectrometer through a Michrom Advance Plug and Play nano-spray source. The nano-LC column (Michrom 3 200Å MAGIC C18AQ 200 ϫ 150 mm) was used with a 90-min gradient (2-10% buffer B in 5 min, 10 -35% buffer B in 65 min, 35-70% buffer B in 5 min, hold at 70% buffer B for 1 min, then down to 2% buffer B in 1 min, holding at 2% buffer B for 13 min) at a flow rate of 2 l min Ϫ1 for the maximum separation of tryptic peptides. MS and MS/MS spectra were acquired using a top 4 method and an MS survey scan was obtained for the m/z range 400 -1300. An isolation mass window of 2 Da was for the precursor ion selection, and a normalized collision energy of 35% was used for the fragmentation. Tandem mass spectra were extracted by Xcalibur version 2.0.7. Charge state deconvolution and deisotoping were not performed. All MS/MS samples were analyzed using X! Tandem (The GPM, thegpm.org; version CYCLONE (2013.02.01.1)). X! Tandem was set up to search a Yersinia database (9010 entries) assuming the digestion enzyme trypsin. X! Tandem was searched with a fragment ion mass tolerance of 0.40 Da and a SEPTEMBER 19, 2014 • VOLUME 289 • NUMBER 38 parent ion tolerance of 20 PPM. Carbamidomethyl of cysteine was specified in X! Tandem as a fixed modification. Glu-Ͼpyro-Glu of the n-terminus, ammonia-loss of the n-terminus, deamidation of asparagine and glutamine, oxidation of methionine and tryptophan, acetyl of the N-terminus and phosphorylation of serine, threonine, and tyrosine were specified in X!Tandem as variable modifications. Scaffold (version Scaffold_4.0.1, Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if they met or exceeded 95% probability in Scaffold. Protein identifications were accepted if they were greater than or equal to 95% probability determined in Scaffold, and they contained at least 2 identified peptides. Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters. Two LC-MS/MS runs on the LTQ-FT were used to create the list of possible phosphorylation sites. Regulation of YpkA Activation by Multisite Autophosphorylation In Vitro Kinase Assays-Kinase reactions were performed with ϳ2 g of the indicated recombinant GST-YpkA protein in 30 l of kinase buffer containing 20 mM Hepes, pH 7.4, 10 mM MgC 2 H 3 O 2 , and 1 mM DTT supplemented with 0.5 mM ATP, 2 Ci [␥-32 P]ATP and incubated with 1 g of actin (Sigma). The reaction was incubated at 30°C on the thermomixer for 30 min. The reaction was terminated by the addition of 4ϫ LDS sample buffer and heating for 10 min at 70°C. A third of the reaction was separated by SDS/PAGE gel and transferred onto a PVDF membrane. Phosphorylated proteins were visualized by autoradiography. One-third of the reaction was separated by SDS/ PAGE and stained with Coomassie Blue. Densitometry was performed using the ImageJ analysis software (NIH) as described above. RESULTS Identification of a Novel Substrate-binding Domain at the N Terminus of YpkA-We previously demonstrated that the YpkA N-terminal residues 1-430 mediate interaction with G␣q (20). To further define the substrate-binding domain we generated GFP-tagged N-terminal YpkA truncations and performed immunoprecipitation assays. As previously shown both full-length YpkA and a YpkA variant containing the kinase domain (amino acids 1-430) interacted with G␣q, whereas a C-terminal YpkA variant containing the GDI-like domain (amino acids 399 -729) was unable to interact with G␣q ( Fig. 2A). Furthermore, whereas the N-terminal deletion mutant YpkA 40 -729 associated with G␣q ( Fig. 2A, lane 5), further deletion at the N terminus of YpkA resulted in loss of binding to G␣q ( Fig. 2A, lanes 6 -8). Thus, amino acids 40 -88 of YpkA are required for efficient binding of G␣q, since deletion of the first 88 amino acids prevented G␣q binding. To further define the region responsible for substrate binding we created additional N-terminal YpkA deletion mutants (Fig. 2B). As compared with the efficient substrate binding by YpkA 40 -729 (Fig. 2B, lane 5), removal of an additional ten amino acids (YpkA 50 -729 ) was sufficient to disrupt G␣q binding (Fig. 2B, lanes 6 -10). Controls in this experiment included full-length YpkA and YpkA 1-430 that bound to G␣q (Fig. 2B, lanes 2-3), as opposed to YpkA 399 -729 that did not (Fig. 2B, lane 4). N-terminal YpkA peptides fused to GFP did not associate with G␣q suggesting that additional residues play a role in substrate binding (Fig. 2C, lanes 5-7). Amino acids 40 -50 of YpkA Are Required for Efficient Inhibition of G␣q Signaling-YpkA binds and phosphorylates the heterotrimeric G protein G␣q interfering with GTP binding and activation (20). We examined the effect of the YpkA N-terminal deletion mutants on G␣q signaling using a tubby nuclear translocation assay to monitor activation of G␣q (20). The transcription factor tubby is involved in maturity-onset obe-sity in mice and is known to be a downstream target of G␣q (38). Tubby localizes to the plasma membrane by binding phosphatidylinositol 4,5-bisphosphate. Receptor-mediated activation of G␣q is thought to release tubby from the plasma membrane through the activity of phospholipase C-␤, triggering translocation of tubby to the nucleus (38). HEK293A cells were transfected with a mCherry-tagged tubby plasmid and either GFP, GFP-YpkA, GFP-YpkA D267A or a GFP-YpkA N-terminal deletion mutant. After overnight transfection, HEK293A cells were stimulated with carbachol for 60 min. Then cells were washed, fixed, permeabilized, and visualized via immunofluorescence. Expression of GFP in control cells did not interfere with G␣q-mediated nuclear localization of tubby (Fig. 3A, panel A). As previously shown, expression of YpkA interfered with the nuclear translocation of tubby after carbachol stimulation (Fig. 3A, panel B), whereas the catalytically inactive YpkA mutant, YpkA D267A , did not prevent localization of tubby to the nucleus confirming that the kinase activity of YpkA is required for inhibition of G␣q signaling (Fig. 3A, panel C). Although YpkA 1-430 binds G␣q, it does not interfere with its activation (Fig. 3A, panel D). The GDI-like domain was not sufficient to prevent the nuclear localization of tubby (Fig. 3A, panel E). Importantly, YpkA 40 -729 interacts with G␣q and inhibits its activation (Fig. 3A, panel F). However, YpkA 50 -729 did not inhibit G␣q mediated nuclear translocation of tubby, nor did further N-terminal deletion mutants (Fig. 3A, panels G-M). Therefore, residues 40 -50 of YpkA appear critical for inhibition of G␣q. Since autophosphorylation is a requirement for YpkA kinase activation and subsequent inhibition of G␣q signaling, we confirmed the ability of these N-terminal YpkA mutants to autophosphorylate in an in vitro kinase assay (Fig. 3B) (30). Bacterially expressed GST-YpkA proteins were generated and subjected to an in vitro kinase assay. Autophosphorylation was observed in wild-type YpkA and YpkA 100 -729 . In contrast, autophosphorylation was not observed with either the kinase inactive mutant, YpkA D267A , or the N-terminal deletion mutant YpkA 151-729 , suggesting that amino acids 100 -150 are required for autophosphorylation, since deletion of amino acids 1-99 did not affect YpkA kinase activity. Thus, the inability of YpkA 50 -729 to interfere with G␣q signaling is likely due to reduced substrate binding rather than improper folding of the protein (Fig. 3B). We next examined the effect of deleting amino acids 40 -50 of YpkA on G␣q binding and phosphorylation. Immunoprecipitates prepared from HEK293A cells transfected with vector, FLAG-YpkA, FLAG-YpkA ⌬40 -50 , or FLAG-YpkA 1-430 were analyzed by SDS/PAGE and immunoblotting. YpkA and YpkA 1-430 associated with G␣q, whereas YpkA ⌬40 -50 had minimal binding to G␣q (Fig. 4A). To detect phosphorylation of G␣q by YpkA we used an antibody generated against phospho-G␣q (Ser-47). Phosphorylated G␣q was detected in lysates of YpkA-transfected HEK293A cells. However, phosphorylation of G␣q was significantly reduced in cell lysates expressing YpkA ⌬40 -50 (Fig. 4B). We further tested the effect of YpkA ⌬40 -50 on the nuclear translocation of tubby upon carbachol stimulation (Fig. 4C). Tubby nuclear translocation was seen in control cells as well as in cells expressing the catalytically deficient mutant, YpkA D267A . Wild-type YpkA interfered with tubby nuclear localization, whereas, YpkA ⌬40 -50 lost ability to inhibit G␣q mediated translocation of tubby to the nucleus, suggesting that amino acids 40 -50 on YpkA enhance the efficiency of substrate phosphorylation. The Substrate-binding Domain Resides within the Membrane Localization Domain of YpkA-Upon translocation into a target cell by the Yersinia T3SS, YpkA localizes to the inner side of the plasma membrane (28). Previous studies reported that the MLD of YpkA resides within amino acids 20 -90. Since the substrate-binding domain lies within this region we examined the effect of a YpkA ⌬20 -90 internal deletion on G␣q binding, phosphorylation and signaling. As shown in Fig. 5A, YpkA ⌬20 -90 was deficient in G␣q binding and phosphorylation. Additionally, YpkA ⌬20 -90 did not interfere with G␣q-mediated nuclear translocation of tubby (Fig. 4C). To determine if deletion of amino acids 40 through 50 interfered with membrane localization we transfected HEK293A cells with vector, FLAG-YpkA, FLAG-YpkA 1-430 , FLAG YpkA ⌬40 -50 , and FLAG-YpkA ⌬20 -90 and performed immunofluorescence microscopy. After overnight transfection, cells were fixed, washed and permeablized. YpkA was detected by FLAG-Cy3 visualization using a confocal microscope. YpkA, YpkA 1-430 , and YpkA ⌬40 -50 localized to the plasma membrane, however YpkA ⌬20 -90 did not, and was detected in the cytoplasm (Fig. 5B). Thus, in addition to a secretion/translocation and membrane localization domain, the N terminus of YpkA also contains a substrate-binding domain comprising of amino acids 40 -49 that are essential for efficient substrate binding and phosphorylation. (30). A YpkA S90A/S95A mutant was reduced in autophosphorylation and phosphorylation of exogenous substrates in an in vitro kinase assay (30). To assess whether autophosphorylation on serine 90 and serine 95 were required for phosphorylation of G␣q we examined the affect of a YpkA S90A/S95A mutant on G␣q activation. HEK293A cells were transfected with GFP-tubby and either vector, YpkA, YpkA D267A , or YpkA S90A/S95A and a tubby nuclear translocation assay was performed. After overnight transfection, the cells were stimulated with carbachol for 60 min and processed for immunofluorescence microscopy. Tubby nuclear localization was evident in cells expressing vector and YpkA D267A only (Fig. 6A). Wildtype YpkA and YpkA S90A/S95A inhibited the nuclear translocation of tubby after carbachol stimulation (Fig. 6A). To confirm that YpkA S90A/S95A underwent autophosphorylation we performed metabolic labeling experiments to assess YpkA autophosphorylation and G␣q phosphorylation by YpkA in vivo. HEK293A cells were transiently transfected with G␣q Q209L in the presence of either vector, wild-type YpkA, YpkA D267A , or YpkA S90A/S95A . Following a 2-h incubation with [ 32 P]orthophosphate, YpkA and G␣q were immunoprecipitated from cell lysates with anti-FLAG and anti-G␣q antibodies, respectively. Phosphorylated proteins were separated by SDS-PAGE and visualized by autoradiography. As shown in Fig. 6B, incorporation of [ 32 P]orthophosphate into the YpkA protein was observed in wild-type YpkA and YpkA S90A/S95A , but not in vector controls or upon exposure of equal amounts of the inactive kinase-deficient YpkA D267A . Additionally, G␣q was phosphorylated by all YpkA variants except in vector controls and YpkA D267A lysates. Thus, autophosphorylation on serine 90 and serine 95 on YpkA are not critical for G␣q phosphorylation. Taken together, these results allude to the presence of additional autophosphorylation sites on YpkA and underscore the complexity of this Yersinia effector. Mapping Autophosphorylation Sites in YpkA-To identify additional autophosphorylation sites in vivo we used a mass spectrometric approach. FLAG-YpkA was isolated from transiently transfected HEK293A cells by binding to anti-FLAG agarose beads. Following gel electrophoresis, immunoprecipitated FLAG-YpkA was extracted from the gel followed by nano liquid chromatography tandem mass spectrometry. We identified nineteen autophosphorylation sites, two of which were previously reported (Fig. 7A) (30). These residues were located in the secretion/translocation region, kinase domain, and RhoGDI domain of YpkA. We initially generated a full-length YpkA mutant containing serine to alanine point mutations in all nineteen residues, YpkA SA (Fig. 7A). FLAG-YpkA, FLAG-YpkA D267A and FLAG-YpkA SA were immunoprecipitated from transiently transfected 293A human embryonic kidney cells and autophosphorylated in vitro in a kinase buffer containing [␥-32 P]ATP. Wild-type YpkA underwent autophosphorylation, whereas, YpkA D267A and YpkA SA were deficient in their ability to incorporate radioactivity (Fig. 7B). We next determined whether YpkA SA was affected in its ability to phosphorylate G␣q by performing a Western blot using a phospho-G␣q antibody. G␣q phosphorylation was significantly reduced in HEK293A lysates expressing YpkA SA compared with lysates expressing YpkA (Fig. 7C, second panel). Lack of G␣q phosphorylation was not due to its inability to bind to G␣q as shown in Fig. 7C (first panel). These results imply that YpkA SA may harbor critical residues required for YpkA autophosphorylation. We next generated four additional YpkA serine to alanine mutants: YpkA SA1-150 contains S52A, S55A, S90A, S95A, S102A, S144A, and S147A; YpkA SA1-400 contains S52A, S55A, S90A, S95A, S102A, S144A, S147A, S164A, S303A, S317A, S320A, S327A, S353A, and S389A; YpkA SA150 -400 contains S164A, S303, S317A, S320A, S327A, S353A and S389A; YpkA SA436 -710 contains S496A, S520A, S529A, S534A, and S620A. FLAG-tagged proteins were expressed in HEK293A cells and immunoprecipitated with anti-FLAG. Immunoprecipitates were subjected to an in vitro kinase assay and followed by SDS-PAGE. The level of autophosphorylation was similar for YpkA and YpkA SA436 -710 suggesting that the indicated point mutations within the RhoGDI domain did not affect YpkA autophosphorylation (Fig. 7D). As we previously observed, YpkA SA had minimal levels of autophosphorylation. The YpkA SA1-400 mutant spanning the secretion/translocation and kinase domains was significantlyimpairedinitsautophosphorylationactivity.Thelevelsof autophosphorylation in the YpkA SA1-150 and YpkA SA150 -400 variants were .52 and .26, respectively, relative to wild type YpkA. We examinedtheaffectoftheYpkAvariantsonG␣qsignalinginacellular environment by assessing their ability to interfere with the nuclear translocation of GFP-tubby upon carbachol stimulation. Empty vector control cells displayed nuclear localization of tubby after stimulation with carbachol (Fig. 7E). YpkA expressing cells inhibited the nuclear accumulation of tubby in the nucleus, whereas cells expressing the catalytically inactive YpkA D267A mutant showed tubby nuclear localization. The lack of YpkA SA and YpkA SA1-400 to inhibit the nuclear translocation of tubby shows a direct correlation with the level of autophosphorylation activity. In the presence of YpkA SA1-150 , YpkA SA150 -400 and YpkA SA436 -710 , tubby did not accumulate in the nucleus. Altogether, our results suggest that multisite autophosphorylation is an important determinant in the regulation of YpkA kinase activity. DISCUSSION Identification of a Novel Substrate-binding Site at the N Terminus of YpkA-Virulence in Y. pseudotuberculosis depends upon the translocated virulence factor YpkA, a protein that disrupts the actin cytoskeleton and inhibits phagocytosis (13). Despite the known importance to virulence, little information has been forthcoming to elucidate the mechanism of activity of YpkA. This has been particularly true of the N-terminal domain of the protein, which, while known to have serine/threonine kinase activity, has remained enigmatic in terms of function. In addition to a kinase domain, the N-terminal domain of YpkA also contains a membrane localization domain, amino acids 20 -90, and a chaperone-binding domain, amino acids 20 -77 (27)(28). In this study, we have determined by deletion experiments that a region between residues 40 -49 of YpkA is critical for efficient substrate-binding and phosphorylation, and inhibition of G␣q signaling. The low levels of G␣q observed in Fig. 2B, lanes 6 -10, suggest that, although, amino acids 40 -49 on YpkA enhance the efficiency of substrate binding additional residues may also contribute to substrate binding. Our results using YpkA peptides indicate that the substrate-binding domain is likely "discontinuous" and is dependent on the secondary and tertiary structure of YpkA. Thus, the N terminus of YpkA serves multiple functions for this type III effector. Multifunctional domains of T3SS effector proteins are an emerging theme in Yersinia. For example, the Yersinia T3SS effector YopH is a 468-amino acid protein tyrosine phosphatase, responsible for disruption of focal adhesions and inhibition of integrin-mediated bacterial phagocytosis (9 -11). The N-terminal 129 amino acids of YopH comprise chaperone-and substrate binding activities (39). Additionally, the first 100 residues of YopT, a Yersinia cysteine protease that affects the host actin cytoskeleton by targeting Rho GTPases, contain a chaperone binding site and are essential for binding to Rho GTPases (40). Thus, Yersinia has evolved bacterial virulence factors to exploit the multifunctionality of a single modular domain, as is the case for the YpkA, YopH, and YopT N-terminal domains. Membrane Localization and Substrate Specificity-Yersinia effector proteins are translocated to different subcellular locations within the target cell, emphasizing the complexity of the strategy used by Yersinia to neutralize host defenses. YopE, a Yersinia GTPase-activating protein for RhoA, Rac1 and Cdc42, is targeted to the perinuclear region of a cell, where all three GTPases are localized (28). YopH localizes to focal adhesion complexes, where it is in close proximity to its substrates (28). Yersinia-delivered YopT is targeted to the plasma membrane, where RhoA is located (7). Upon translocation into the host cell, the N terminus of YpkA localizes the protein to the inner surface of the host cell plasma membrane, where it is in close proximity to key proteins involved in transducing extracellular signals into eukaryotic cells. Groves et al. demonstrated the importance of YpkA subcellular localization to the plasma membrane for RhoGTPase binding (33). YpkA was shown to selectively inhibit Rac-dependent Fc␥ receptor-mediated phagocytosis by specifically targeting endogenous membranebound Rac isoforms in cells (33). More importantly, an overexpressed YopO localization deficient mutant (YopO ⌬20 -77 ) had significantly reduced anti-phagocytic ability upon challenge with IgG-sRBC (33). Our findings provide strong evidence for the importance of plasma membrane localization for interaction of YpkA with its cognate host targets. Heterotrimeric G proteins are covalently modified at or near their N termini by covalent attachment of the fatty acids myristate and/or palmitate (41). Palmitoylation on C9 and C10 of G␣q is required for membrane attachment. YpkA interferes with G protein-coupled receptor signaling by inactivating the heterotrimeric G protein G␣q. YpkA binds and phosphorylates G␣q on a critical serine residue preventing GTP binding (20). We have deter- FIGURE 7. Multiple N-terminal autophosphorylation sites stimulate YpkA kinase activity. A, summary of the mass spectrometry results. FLAG-YpkA was immunoprecipitated from transfected HEK293A cells with anti-FLAG agarose beads. After purification by affinity chromatography, SDS-PAGE, and Coomassie Blue staining, bands corresponding to YpkA were excised from the gel and subjected to nano liquid chromatography tandem mass spectrometry as described in "Experimental Procedures." Nineteen potential autophosphorylation sites were identified, including two previously reported (Ser-90 and Ser-95) (30). The YpkA serine to alanine (YpkA SA ) mutants were generated using overlap extension PCR. B, autophosphorylation of wild type and mutant YpkA proteins. Immunoprecipitated FLAG-YpkA constructs were subjected to an in vitro kinase assay, run on SDS-PAGE and blotted onto PVDF membrane. Shown are an autoradiogaph (upper panel) and an immunoblot probed with an anti-FLAG antibody (lower panel). C, YpkA SA is deficient in G␣q phosphorylation. FLAG-tagged YpkA proteins and G␣q were expressed in HEK293A cells, and cell lysates were immunoprecipitated with anti-FLAG. Immunoprecipitates were subjected to SDS-PAGE and immunoblotted with antibodies anti-G␣q (first panel) and anti-FLAG (fourth panel). Whole cell lysates were probed with phospho-G␣q (second panel) and G␣q (third panel) antibodies. (*, nonspecific band). D, autophosphorylation of YpkA constructs harboring the indicated serine to alanine point mutations. HEK293A cells were transfected with YpkA or the indicated YpkA variants. Cells were lysed, and anti-FLAG immunoprecipitates were subjected to an in vitro kinase assay. Samples were analyzed by autoradiography (upper panel) and anti-FLAG immunoblotting (lower panel). The numbers indicate levels of autophosphorylation of the YpkA constructs relative to YpkA (control). E, N-terminal serine residues are required for in vivo YpkA activity. Merged image of HEK293A cells transfected with GFP-tubby (green) and vector or indicated YpkA variants. Hoescht staining (blue) was used to detect nuclear staining, and rhodamine phalloidin was used to detect actin (red). mined that in addition to G␣q, YpkA associates with other members of the G␣q family, but not with members of the G␣12/13, G␣i or G␣s families (data not shown). Thus, YpkA exhibits substrate specificity and acts only on a defined subset of cellular targets. Our results demonstrate that substrate binding is not a requirement for appropriate subcellular localization of YpkA. The molecular mechanism used for membrane localization by YpkA is unknown. Recently, Salomon et al. identified a conserved bacterial phosphoinositide-binding domain (BPD) present in type III effectors of both animal and plant pathogens, including YpkA. However, a YpkA truncation mutant containing a mutation of a conserved tyrosine residue (Tyr41 in Y. pseudotuberculosis and Tyr-38 in Y. enterocolitica) within the BPD could not be examined for membrane localization due to lack of expression (42). We speculate that additional YpkA kinase substrates will be targeted to the plasma membrane. Thus, the membrane localization domain mediates the full effect of YpkA function by contributing to both the kinase activity and the GDI activity of the C-terminal domain. Multisite Phosphorylation Regulates Kinase Activity-Our data indicate that YpkA kinase activity is regulated by multisite autophosphorylation within its N-terminal domain (amino acids 1 to 400). We observed the level of autophosphorylation for the YpkA SA1-400 and YpkA D267A mutants to be similar. Based on our results it seems likely that serine residues within the first 150 amino acids of YpkA (YpkA SA1-150 ) and within amino acids 150 to 400 are necessary for full activation of YpkA kinase activity. Additional one, two and three point mutations within these regions to narrow down the critical sites reduced the level of YpkA autophosphorylation, but had no effect on tubby nuclear translocation (data not shown). What advantages are offered by the complex regulation of YpkA involving multisite phosphorylation? It is conceivable that the requirement for multiple phosphorylation sites may impose a certain threshold of YpkA autophosphorylation that must be obtained for activation of YpkA. In this way, a low level of YpkA autophosphorylation would allow the kinase to phosphorylate its substrate(s), and thus, interfere with host signaling. Interestingly, YpkA is secreted in lower amounts relative to other Yersinia effectors (12). Thus, multisite phosphorylation could introduce the potential for sophisticated control over the dephosphorylation and inactivation of YpkA by host phosphatases and influence the strength and duration of YpkA kinase activity within the target cell. The multisite phosphorylation of proteins is indeed an extremely common mechanism for greatly increasing the regulatory potential of proteins. For example, the protein kinase MAPK-activated protein kinase-2 plays important roles in protecting cells against cell-damaging agents and infection. Its activation by stress-activated protein kinase-2 (SAPK2, also called p38) is accompanied by the phosphorylation of three residues, namely Thr-221, Ser-272, and Thr-334 (43). Phosphorylation of any one residue is insufficient for activation, whereas maximal activation is achieved if any two of the three sites are phosphorylated. Despite the extensive analysis on YpkA, its mechanism of kinase activation remains enigmatic. Obviously, it is not clear how multisite phosphorylation orchestrates the activity of YpkA, and it will be of considerable interest to determine whether phosphorylation at each site is critical. Structural analysis of full-length YpkA would provide significant insights into the mechanism of YpkA kinase activation, as was reported for the C terminus of YpkA (32).	2018-04-03T05:45:00.152Z	2014-08-01T00:00:00.000	{ "year": 2014, "sha1": "7673caa06f3a1e6aadafa09968fdd2c6ab20c710", "oa_license": "CCBY", "oa_url": "http://www.jbc.org/content/289/38/26167.full.pdf", "oa_status": "HYBRID", "pdf_src": "Highwire", "pdf_hash": "bb85a52d0fe5d1dc148e333050b2cda11125de70", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
253509771	pes2o/s2orc	v3-fos-license	Shapeshifting tau: from intrinsically disordered to paired-helical filaments Abstract Tau is an intrinsically disordered protein that has the ability to self-assemble to form paired helical and straight filaments in Alzheimer’s disease, as well as the ability to form additional distinct tau filaments in other tauopathies. In the presence of microtubules, tau forms an elongated form associated with tubulin dimers via a series of imperfect repeats known as the microtubule binding repeats. Tau has recently been identified to have the ability to phase separate in vitro and in cells. The ability of tau to adopt a wide variety of conformations appears fundamental both to its biological function and also its association with neurodegenerative diseases. The recently highlighted involvement of low-complexity domains in liquid–liquid phase separation provides a critical link between the soluble function and the insoluble dysfunctional properties of tau. Introduction Tau is a microtubule binding protein and an important component of the neuronal cytoskeleton as well as playing a role in the nucleus [1,2]. Humans have six major spliced isoforms in the central nervous system from MAPT, a single gene located on chromosome 17 (cytogenetic location 17q21.1), which range from 352 to 441 residues [3]. The longest form contains an N-terminal projection domain (residues 1-165), a proline rich region (residues 166-242), a microtubule binding region (MTBR) (residues 243-367) and a C-terminal (residues 368-441) and is referred to as 2N4R tau ( Figure 1). The MTBR consists of four imperfectly repeated sequences: R1 (243-273), R2 (274-304), R3 (305-335) and R4 (336-367) [4] (Figure 1). Isoforms can differ in the number of microtubule-binding imperfect repeats and can exist in three or four repeat forms (3R and 4R, respectively) with there being equal amounts in the human cortex [5] ( Figure 1). The primary sequence of tau has a relatively low proportion of hydrophobic amino acids (mean hydrophobicity of 0.4) [4,6] and has a high-number of polar and charged amino acid residues (net charge at neutral pH is +2), which results in it being highly soluble in water and natively unfolded. A large number of charged residues are found in the MTBR resulting in a local net-charge of +9 [4]. The primary structure of tau can contain an array of post-translational modifications of specific residues including: phosphorylation, acetylation, deamidation, methylation, O-glycylation or ubiquitination, many of which may impact on the structural organisation of the protein [6,7]. Truncation can also cause the loss of the N-terminal and/or C-terminal regions [8,9]. Intracellular accumulations of tau in tauopathies [10,11] arise when tau misfolds and self-assembles into filaments with the classic amyloid-like, cross-β structure [12]. The tauopathies include Alzheimer's disease (AD), Pick's-type frontotemporal dementia, cortical basal degeneration, chronic traumatic encephalopathy and progressive supranuclear palsy [13]. In AD, the filaments contain both 3R and 4R tau, while in Pick's disease the filaments are composed of 3R only and in cortical basal degeneration, only 4R is deposited [13]. Despite the well-known association of tau misfolding with a large number of neurodegenerative diseases, it remains unclear how tau is involved in disease causation and progression and what may trigger protein misfolding. One potential area of interest is liquid-liquid phase separation (LLPS) . Tau can exist with three or four repeats (3R or 4R, respectively) and R2 is missing from 3R tau. R2 and R3 contain hexapeptide motifs that have been highlighted in driving amyloidogenic propensity (VQIINK and VQIVYK). There are multiple possible phosphorylation sites on serine, threonine and tyrosine in tau and some of these are highlighted (purple p). NMR studies have shown that regions of the protein have β-sheet propensity (denoted by green arrows). Full-length tau 2N4R is shown. where tau may be present at higher concentrations and this could give rise to the ideal conditions for abnormal protein aggregation [14]. In this mini-review, we discuss the shape-shifting nature of tau where it can interconvert from soluble and natively unfolded, to phase separated and cross-β assemblies. Here, we have mainly focussed on studies that did not include heparin to initiate assembly. The biological function and conformation of tau Tau is best known for its association and stabilisation of microtubules, predominantly in neuronal cells [15]. It binds to the tubulin dimers via the MTBR. The cryo-electron microscopy (cryoEM) structure of tau associated with microtubules revealed the high-resolution structure of R1-R2 regions organised like a staple across the tubulin molecules to stabilise the growing microtubules (6CVK .pdb) while R3-R4 and the proline-rich region were less well defined in the cryoEM map [16] (Figure 2). Modelling and experimental analysis of the full-length tau in association with microtubules revealed that post-translational modifications including phosphorylation and acetylation modulate the ensemble conformation resulting in altered binding affinities (7QPC) [17]. Although tau is named for its microtubule binding properties, it has been shown to bind to multiple proteins including actin and protein phosphatase 2A [18], various membrane proteins and α-synuclein [19]. More recently, tau has been identified in the nucleus [20] and further localised to the nucleolus where it appears to participate in chromatin remodelling [21]. Interestingly, the post-translational modifications on tau appear to regulate both its distribution, binding partners and also its conformation. Phosphorylation plays an important role in conformation of tau and nuclear tau is predominantly non-phosphorylated [1. 21] Tau has no well-defined secondary or tertiary structure and is classified as an intrinsically disordered protein (IDP). Solution state NMR of full-length tau has been hampered by the large size and intrinsically disordered nature of the protein [22]. Circular dichroism studies conducted at different pH, temperatures and ionic strengths showed very little differences in the majority random coil secondary structure for 2N4R tau [4]. NMR studies [22] indicate that short regions of the tau sequence possess β-structure propensity, identifying previously known regions 274-284 and 305-315 which include the amyloidogenic VQIINK and VQIVYK motifs [23] with much of the rest of the protein being random coil in character (343-411) [23]. Serine, threonine or tyrosine phosphorylation have been suggested to modulate the amyloidogenicity of tau and most of the phosphorylation sites are found in the N-terminal portion of the protein mainly within the proline rich domains [22]. However, some potential phosphorylation sites are found in the repeat regions and phosphorylation of Y310 might be expected to affect self-assembly propensity since tyrosine 310 is found within the key amyloidogenic driving region. Self-assembly of tau The mechanism of self-assembly of full-length tau into form amyloid fibrils remains unclear given the difficulty of enabling assembly in vitro without the contribution of additives. In vitro, full-length tau is robustly soluble, requiring additives to initiate self-assembly in contrast with the amyloidogenic nature of amyloid-β [24]. Additives that have been used include heparin [25], RNA [26] and lipids [27]. However, studies utilising the kinetics analysis in the absence of heparin pioneered by Linse et al. [28] have indicated that under quiescent conditions at pH8, tau 304-380(C322S) fragment assembles via a slow primary nucleation phase followed by a dominant secondary nucleation phase fuelled by the additional of monomers to the filament surfaces and the production of further oligomers arises from secondary nucleation events [28]. In general, self-assembly of tau peptides follows the expected secondary structural transition from random coil to β-sheet by circular dichroism measurements with time [29] although clear observation of the β-sheet signal contribution requires resuspension of a pellet due to the dominant nature of the random coil signal in solution [29]. This appears to indicate that assembly is slow and that there is a dynamic equilibrium that exists between fibrillar and soluble tau. Historically, regions covering R1-R4 have been investigated, referred to as K18 (244-372) and K19 (244-372 lacking R2 275-305) using heparin to template aggregation [30,31], and short hexapeptides with sequences 306 VQIVYK 311 [32] and 275 VQIINK 280 [33] have been shown to form highly ordered fibrils in vitro. However, the inclusion of heparin appears not to result in filaments that resemble those extracted from AD brain [34,35] and isolated short peptides Left and middle panel show immunogold labelling electron microscopy of tau filaments in AD brain tissue (tau filaments identified using antibodies against total tau (A) and tau oligomers (T22) (B) while the right panel (C) shows PHF formed in vitro from the tau fragment 297-391, dGAE [ 29]. Inserts show enlarged images to highlight the paired-helical appearance. Scale bars are shown. Early work isolated a fragment of tau from AD brain which was resistant to degradation and this was mapped to 297-391 and named dGAE for the last three amino acids of the C-terminus of the fragment [8,9]. dGAE is able to self-assemble in vitro without any additives [29]. Under certain conditions, dGAE can form filaments that appear to twist and resemble PHFs by electron and atomic force microscopy and they share the amyloid cross-β structure by X-ray fibre diffraction [29,37] (Figure 3). Monitoring assembly using circular dichroism shows a gradual loss of random coil content and a concurrent increase in β-sheet signal [29]. dGAE contains a single cysteine residue at 322 and oxidation to form disulphide linked dimers hinders assembly [29]. In contrast, previous studies using different tau models, such as full-length tau and K18/K19 fragments, indicated that these disulphide linked dimers are essential for tau self-assembly [38][39][40]. These contrasting findings illustrate the importance of truncation and the influence of the environment on tau self-assembly, and highlights the need for a reliable model in tau self-assembly. The region first observed by cryoEM of Alzheimer's derived tau filaments, region 306-378 [41], is also able to self-assemble in vitro and to seed the assembly of full-length tau [42]. Following oxidation of dGAE, one of the resulting modifications is the formation of dityrosine cross-links. These cross-links appear to 'freeze' or halt further self-assembly of dGAE and prevent further elongation [43,44]. Dityrosine cross-linking is observed in neurofibrillary tangles where it may stabilise the deposits protecting them from disassembly, while only soluble forms of tau appear to be susceptible to cross-linking [44]. Early fluorescence resonance energy transfer (FRET) studies of tau in solution have suggested interactions between the N and C-termini of tau suggesting the formation of a 'hairpin' [45]. More recent studies have identified two stable forms of tau from brain and recombinant sources, one of which is inert and the other is assembly competent [46]. Although no obvious differences in secondary structure were apparent, the two forms differed in intermolecular contacts involving lysine residues located in R1 and R2 and modelling studies supported by protease digestion experiments implicated the exposure of the assembly driving regions VQIINK and VQIVYK in repeat 2 and repeat 3, respectively (Figure 1) in the assembly competent form, while it was buried in the assembly inert form [46]. Interesting, the inert form could be converted to assembly competent form in the presence of heparin. This highlights the capacity of tau to adopt alternative structural ensembles capable of multiple functions, from molecular interactions to self-assembly. The pathological structures of tau Electron microscopy of tissue sections clearly shows the paired helical appearance of AD filaments with a regular repeat distance of 73 nm (Figure 3). AD tissue extracted filaments provided more detail with a C-shaped cross-section and the identification of a fuzzy coat which could be removed using proteases [47]. Early work using X-ray fibre diffraction revealed that paired helical filaments (PHF) isolated from AD brains shared the cross-β structure that characterises amyloid fibrils [12] demonstrating that the tau PHF share the hallmark cross-β amyloid structure and [48] (6HRF), filaments from cortical basal degeneration [49] (6TJX), (C) filaments from Pick's disease [50] (6GX5), (D) filaments from chronic traumatic encephalopathy [51] (6NWQ), (E) paired helical filaments from AD brain ( [48] (6HRE) and (F) shows a predicted structure of 2N4R using alphafold (https://alphafold.ebi.ac.uk) [52]. leading to the classification of tau filaments as amyloid fibrils. More recently, major advances in cryoEM techniques have resulted in a gallery of exciting details of the different architectures of disease associated tau filaments. All the structures are cross-β, with a distance of 4.7Å between stacked β-strands but the organisation of each individual monomeric cross-section differs. Starting with AD PHF and straight filaments (SF), it was shown that the core structure was made up of repeats 3R-4R, residues 306-378 [41] later extended to 304-380 [48]. It was confirmed that the filament cross-section is made up of two C-shaped protofilaments that associate back-to-back and have slightly different protein-protein interactions for PHF versus SF (Figure 4). Chronic traumatic encephalopathy extracted filaments share a similar structure to those of AD, while Pick's disease filaments are quite different and arise from only 3R tau (Figure 4). On the other hand, cortical basal degeneration filaments are made of only 4R tau and give another architecture again (Figure 4). Since these early structures were solved, many more tauopathy filaments have been provided all with similarities and differences raising the pressing question of what drives and then determines the specific amyloid fold and how much the conformation and architecture impact on the characteristic symptoms of each disease. Solid-state NMR of filaments formed from full-length tau 2N4R in the absence of heparin revealed that the region P270-S400 was highly structured in the core and not observed by 1 H-15 N INEPT, which allows observation of dynamic regions [49], suggesting a larger core region formed by this isoform in vitro that those derived from AD tissue and more similar to those observed for filaments extracted from cortical basal degeneration tissue [22,49,50]. Recent advances have been made in analysis of atomic force microscopy images allowing single filaments to be examined and reconstructed paving the way for the potential to interrogate the structure of individual filaments within a sample [53]. This approach was used to investigate the structure of PHFs generated without additives from dGAE in vitro (Figure 3), indicating that these filaments shared a structure with those extracted from AD brain [37,48,53]. More recently, Lovestam et al. confirmed that the dGAE has the ability to form AD-like PHF, solving the structure by cryoEM [54] and revealing multiple different filament conformations ( Figure 5). Interestingly, small changes in the assembly conditions can modulate the conformation and architecture of the filaments including an influence of shaking speed and addition of various salts ( Figure 5). These results point to the important contribution of environmental conditions in generating disease specific tau filament structures and highlighting the range of structural strains possible [54]. Extension of the peptide at the N-and C-termini resulted in inhibition of assembly suggesting that these regions are important in maintaining the solubility of full-length tau. The effect of pseudophosphorylation (Ser/Thr>Asp) was also explored and it was observed that pseudophosphorylation at C-terminal sites (396-404) overcame the inhibitory effect of the C-terminal region resulting in formation of PHFs [54]. Liquid-liquid phase separation: connecting the dots? The mechanism by which tau transitions from a highly soluble, flexible protein to an ordered amyloid fibril remains unclear. Tau sequence is defined by the primary sequence which contains low-complexity domains which are often enriched in polar and charged amino acids (Gln, Ser, Pro, Glu and Lys) [55]. They also contain a low number of hydrophobic amino acids (Val, Leu, Ile, Met, Phe, Trp and Tyr) which would normally form the core of a globular protein via hydrophobic collapse [55]. Tau shares some sequence characteristics with prion-like domains and proteins implicated in LLPS, such as the RNA binding proteins FUS, TDP43 and hnRNPA1. LLPS is an emerging phenomenon which results in dynamic and reversible structures that are known as membraneless organelles, condensates, liquid droplets and coacervates since the phase separation is driven by interactions within the droplet that result in a different phase from the surrounding solvent environment. Tau has been observed to associate with phase separated organelles in the cytoplasm and in the nucleus [56] and recently LLPS has been implicated in the mechanism of nucleation of microtubules [57]. Tau has also been observed in stress-granules in association with T-cell intracellular antigen (TIA1), which maintain partially translated mRNA under conditions of stress [58] and in other membraneless organelles [59,60]. The ability of tau to associate with nucleic acids may arise from its function observed in the nucleolus [21]. TIA1 is important in generation of stress granules and tau has been reported to bind TIA1 [61,62], while reduction of TIA1 has been shown to reduce tau-mediated neurodegeneration in vivo suggesting a key important interaction occurs on the pathological pathway in stress granules [63]. Furthermore, TIA1 has been implicated in the formation and propagation of toxic oligomeric tau [64]. In vitro studies of tau have revealed the ability to form LLPS in association with crowding agents or RNA [65,66] (Figure 6). The conformation of tau within the condensates remains unclear since these membraneless organelles are thought to behave as hydrogels. Under normal conditions, phase separation is highly dynamic and fully reversible. However, a transition to the (largely irreversible) insoluble amyloid state has been observed for other neurodegenerative disease related proteins such as FUS in amyotrophic lateral sclerosis [67] and increasingly discussed in relation to other neurodegenerative disease proteins [68,69]. Tau has been shown to associate with RNA, a cofactor known to increase aggregation [26], and phase separate into a complex coacervate that is dynamic allowing monomeric tau to move freely [65] and reversibly [66]. In cells, GFP labelled tau441 formed intraneuronal droplets which show fluorescence recovery after photobleaching (FRAP) implying a liquid state of the structures [73]. The GFP-tau441 was expressed in a phosphorylated form or incubated in crowding agents and was then shown to form liquid droplets in cells [73]. In vitro phosphorylated tau droplets matured with time and formed fibrillar structures providing some support for the idea that LLPS may be an initiator of transition to the insoluble amyloid state [70]. Phase separation of tau can be optimised by titration of RNA and salt (NaCl) concentrations and examination of the regions of tau that participate in RNA-binding centred within proline rich and MTB domains [62]. Furthermore, the inclusion of TIA1 resulted in LLPS of tau without addition of crowding agents and this appeared to support further self-assembly to form oligomeric and fibrillar tau [62]. The conditions that result in conversion of tau from LLPS to filaments remains to be further explored in detail. Structural details of tau in LLPS and during transition are still lacking. However, evidence is accumulating for LLPS providing the ideal environment for the conversion of other neurodegenerative disease proteins from soluble to amyloid-structured and this may provide the missing piece of the puzzle for tau aggregation in vivo. Whether LLPS is a required intermediate or just another potential pathway is yet to be shown but it may unlock future therapeutics and advances in the treatment of tauopathies and other neurodegenerative diseases. Summary • Tau is a protein that is central to a large number of neurodegenerative diseases collectively known as tauopathies and understanding the conformational space occupied by tau is essential for deciphering the mechanisms of pathological tau in tauopathies. • Membraneless organelles arising from liquid liquid phase separation in cells may provide a suitable environment for initiation of aggregation prone conformations of tau resulting in misfolding, aggregation and deposition. Competing Interests The authors declare that there are no competing interests associated with the manuscript. Open Access Open access for this article was enabled by the participation of University of Sussex in an all-inclusive Read & Publish agreement with Portland Press and the Biochemical Society under a transformative agreement with JISC.	2022-11-15T06:17:33.719Z	2022-11-14T00:00:00.000	{ "year": 2022, "sha1": "3f803dabc1646b635879ee5e2a56524ecce2d759", "oa_license": "CCBY", "oa_url": "https://portlandpress.com/essaysbiochem/article-pdf/doi/10.1042/EBC20220150/939451/ebc-2022-0150c.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "68e28a8899b2d78061ec11b35c5c79945267c8c1", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
261100621	pes2o/s2orc	v3-fos-license	A fully-coupled nonlinear magnetoelastic thin shell formulation A geometrically exact dimensionally reduced order model for the nonlinear deformation of thin magnetoelastic shells is presented. The Kirchhoff-Love assumptions for the mechanical fields are generalised to the magnetic variables to derive a consistent two-dimensional theory based on a rigorous variational approach. The general deformation map, as opposed to the mid-surface deformation, is considered as the primary variable resulting in a more accurate description of the nonlinear deformation. The commonly used plane stress assumption is discarded due to the Maxwell stress in the surrounding free-space requiring careful treatment on the upper and lower shell surfaces. The complexity arising from the boundary terms when deriving the Euler-Lagrange governing equations is addressed via a unique application of Green's theorem.The governing equations are solved analytically for the problem of an infinite cylindrical magnetoelastic shell. This clearly demonstrates the model's capabilities and provides a physical interpretation of the new variables in the modified variational approach. This novel formulation for magnetoelastic shells serves as a valuable tool for the accurate design of thin magneto-mechanically coupled devices. Introduction Large deformation of thin structures made from soft rubbers or elastomers is critical for numerous engineering components, including tyres, airbags, air springs, buffers, pneumatic actuators, and soft grippers (Galley et al., 2019;Hao et al., 2017). The analysis of slender structures undergoing large deformation, such as rods, membranes, plates, and shells, in which one or more characteristic dimensions are negligible compared to the others, is challenging. They exhibit both material and geometric nonlinearities often leading to instabilities. Slender structures are generally modelled as lower-dimensional manifolds embedded in three-dimensional space with appropriate kinematic simplifications (Niordson, 1985;Simo and Fox, 1989). Novel developments in smart materials with multi-physics coupling have led to a dramatic increase in technological applications in soft robotics, actuators, and sensors. These materials often rely on a non-mechanical stimulus from electric, magnetic, thermal or chemical fields (Jolly et al., 1996;McKay et al., 2010;Kim et al., 2012;Sheng et al., 2012) and are difficult to model due to their complex physics. Of particular relevance is magneto-mechanical coupling in thin structures due to the ability to produce extremely large reversible deformations in a short time-scale. The presence of strong magneto-mechanical coupling in some manufactured materials, such as magnetorheological elastomers (MREs) (Jolly et al., 1996), has the potential to underpin future engineering and technological applications, for example, in micro-robotics (Hu et al., 2018;Ziyu et al., 2019), as sensors and actuators (Böse et al., 2012;Psarra et al., 2017), in active vibration control (Ginder et al., 2000), and as waveguides (Saxena, 2018;Karami Mohammadi et al., 2019). Magnetoelastostatics concerns the analysis of suitable phenomenological models to describe the equilibrium of deformable solids associated with multifunctional processes involving magnetic and elastic effects. The main constituent of the theory is the coupling between elastic deformation and magnetisation in the presence of externally applied mechanical and magnetic force fields. The magnetoelastic coupling occurs in response to a phenomenon involving reconfigurations of small magnetic domains. This is observable as a continuum vector field emerging from an averaging of microscopic and distributed subfields. Thus, the imposition of a magnetic field also induces a deformation of the material in addition to the magnetic effects caused by the traditional mechanical forces. With a rich history spanning six decades (Tiersten, 1964;Brown, 1966;Maugin and Eringen, 1972;Maugin, 1988;DeSimone and James, 2002;Kankanala and Triantafyllidis, 2004;Dorfmann and Ogden, 2004;Keip and Sridhar, 2019;Sharma and Saxena, 2020;Moreno-Mateos et al., 2023a), the mathematical and computational modelling of magnetoelasticity continues to be an active area of research. The coupling between magnetic fields and mechanical deformations of a shell structure introduces additional complexities compared to purely mechanical or electromagnetic analyses, making the modelling and solution process considerably more challenging. Magneto-active soft materials are broadly divided into two sub-classes based on the type of embedded particles: soft-magnetic soft materials (SMSMs) and hard-magnetic soft materials (HMSMs). SMSMs contain particles with low coercivity, such as iron or iron oxides, and their magnetization vector varies under external magnetic loading. They are often modelled as three-dimensional solid continua (Danas et al., 2012;Saxena et al., 2013;Ethiraj and Miehe, 2016;Mehnert et al., 2017;Mukherjee et al., 2020;Bustamante et al., 2021;Akbari and Khajehsaeid, 2020;Hu et al., 2022). HMSMs consist of particles with high coercivity, such as CoFe 2 O 4 or NdFeB. The magnetisation vector, or remnant magnetic flux, of HMSMs remains unchanged over a wide range of applied external magnetic flux (Lee et al., 2020;Schümann et al., 2021;Moreno-Mateos et al., 2023b). The viscoelastic material behaviour of HMSMs significantly affects the magnetic actuation behaviour of hard-magnetic soft actuators (Lucarini et al., 2022;Nandan et al., 2023;Stewart and Anand, 2023). Motivated by the need to model thin magnetoelastic structures, Steigmann (2004) presented a dimensionally reduced-order model for thin magneteolastic membranes. Barham et al. (2008) investigated the limit point instability for a finitely deforming circular magnetoelastic membrane in an axisymmetric dipole field under oneway magneto-mechanical coupling. This analysis was extended by Saxena (2017, 2018); Saxena et al. (2019); Ali et al. (2021); Mishra et al. (2023) to study wrinkling, bifurcation, and limit point instabilities in axisymmetric inflating magnetoelastic membranes. However, a shell theory for fully-coupled magnetoelasticity that can account for bending resistance is still lacking. An overview of the classical shell theory is given, for example, in Simo and Fox (1989); Cirak et al. (2000); Kiendl et al. (2009) or in the books by Basar and Krätzig (1985); Niordson (1985); Blaauwendraad and Hoefakker (2014). When modelling physical phenomena on curved surfaces, defining geometric quantities (normal vectors, curvatures, etc.) and differential surface operators (gradients, divergence, etc.) is crucial (Steinmann, 2015). Reduced-order theories for hard-magnetic linear and nonlinear beams (Wang et al., 2020;, and rods (Sano et al., 2022) have been derived based on the three-dimensional model presented in Zhao et al. (2019). These studies involved a dimensional reduction procedure on the three-dimensional magneto-elastic energy, assuming reduced kinematics based on the Kirchhoff-Love assumptions (Niordson, 1985). Green and Naghdi (1983) focused on the nonlinear and linear thermomechanical theories of deformable shell-like bodies, considering electromagnetic effects. The development was carried out using a direct approach, utilising the two-dimensional theory of directed media known as Cosserat surfaces. Yan et al. (2020) studied linear elastic magneto-active axisymmetric shells made of HMSMs. They leveraged the coupling between mechanics and magnetism to tune the onset of instability of shells undergoing pressure buckling. Magnetoelastic shell models for axisymmetric deformation and geometrically exact strain measures were compared with experimental results. Their findings demonstrated that the magnetic field can control the critical buckling pressure of highly sensitive spherical shells with imperfections (Hutchinson, 2016;Hutchinson and Thompson, 2018). performed a dimensional reduction of the three-dimensional magneto-elastic energy contribution presented in Zhao et al. (2019) by assuming a reduced kinematics according to the Kirchhoff-Love assumptions, and focussing specifically on hard-magnetic, linear-elastic shells. Models for non-axisymmetric deformations of magnetoelastic shells have been derived for shallow shells (Seffen and Vidoli, 2016;Loukaides et al., 2014). Dadgar-Rad and Hossain (2023) proposed a micropolar-based shell model to predict the deformation of thin HMSMs, incorporating a ten-parameter formulation that considers the micro-rotation of the microstructure with the enhanced assumed strain method to alleviate locking phenomenona. Lee et al. (2023) have presented a direct two-dimensional formulation to couple non-mechanical stimuli with large deformation of shells. However, despite this wealth of research, models for general deformation cases in the context of SMSM shells have received limited attention with the form of the coupling between magnetism and mechanics remaining an open question. To address the aforementioned shortcoming, a theory is derived to model large deformations of soft-magnetic hyperelastic thin shells using the Kirchhoff-Love assumption for mechanical deformation and a linearly varying magnetic scalar potential across the thickness of the structure. The salient features of the theory are the following: 1. In the present work, a derived theory approach is adopted for SMSM shells by considering the total energy of a three-dimensional incompressible magnetoelastic body and its surrounding space as the starting point. A two-dimensional system is derived based on appropriate approximations for a thin shell by incorporating a new set of generalised solution variables in a modified variational setting. The magnetic potential in the free-space is treated as an independent solution variable to capture the underlying physics and strongly couple the magnetoelastic interactions between the shell and the free space. This approach is required to formulate appropriate boundary and interface conditions and ensure consistency in the mathematical modelling of the system. 2. In numerical simulations involving thin structures, the common practice is to apply external hydrostatic pressure at the mid-surface. In the present derived theory approach, a distinction is made between the applied pressures at the top and bottom surfaces. The implications of this departure from the conventional practice are discussed. 3. In the present context, obtaining a dimensionally reduced-order theory entails linearly approximating the variation of the magnetic potential across the shell's thickness. This leads to the total potential in the magnetoelastic body adopting a form similar to the Kirchhoff-Love assumption used for the mechanical behaviour of hyperelastic thin shells. Such an approximation is well-suited for modelling the physics of thin magnetoelastic shells and facilitates mathematical simplifications. 4. The plane-stress assumption, commonly employed in structural mechanics, assumes negligible stresses in the thickness direction of thin plates or shells. However, it is not directly applicable to magnetoelastic shells due to the coupling between magnetic field and mechanical deformation. This coupling results in threedimensional stress and strain states, exemplified by the study of an inflating soft magnetoelastic cylindrical shell presented here. The plane-stress assumption fails to consider magnetic field-induced stresses in the thickness direction, leading to inaccuracies. 5. The physical three-dimensional shell is conceptualised as a stack of surfaces. Thereby, the overall deformation of the shell structure is described by the deformed mid-surface position vector, augmented by a term accounting for the through-thickness stretch and the deformed normal. Introducing the first variation of the thickness stretch and the deformed normal in the modified variational format adds richness and complexity to the derivation of the shell system of equations. Notably, when obtaining a reduced-order model for the thin soft magnetoelastic shell, a unique application of Green's theorem is required. The present work addresses this complexity and provides a suitable generalisation by deriving a system of partial differential equations with boundary terms that encompass these effects. Outline The paper is organised as follows: In Section 2, the mathematical preliminaries and the fundamentals of nonlinear magnetoelastostatics are introduced. Sections 3.1 and 3.2 define the geometry and kinematics of nonlinear magnetoelastic thin shells, respectively. In Section 3.3, the expressions for the divergence of the total stress tensor and the magnetic induction vector in the shell are provided. These are essential for deriving the equilibrium equations. Section 3.4 presents the interface condition on the magnetic potential, which is imposed by the continuity of the tangent space components of the magnetic field at the shell boundaries. Section 4 introduces the variational formulation accounting for the magnetoelastic body and the corresponding free space under suitable loading situations. Then, in Section 5, a new set of generalised solution variables for a modified variational format suitable for deriving the shell system of equations is introduced. Sections 5.2, 5.3, and 5.4 demonstrate the contributions of the stress tensors, magnetic induction vector, and external loads to the modified variational setting, respectively. Section 6 is dedicated to obtaining the governing equations for the system, and in Section 7, an example of an inflating magnetoelastic thin cylinder is illustrated to derive the response equations for a given boundary-value problem using the derived equations. Finally, in Section 8, concluding remarks are presented. Notation A variable typeset in a normal weight font represents a scalar. A bold weight font denotes a first or second-order tensor. A scalar variable with superscript or subscript indices normally represents the components of a vector or second-order tensor. Latin indices i, j, k, . . . vary from 1 to 3 while Greek indices α, β, γ, . . . , used as surface variable components, vary from 1 to 2. Einstein summation convention is used throughout. e i represent the basis vectors of an orthonormal and orthogonal system in three-dimensional Euclidean space with x, y and, z as its components. The three covariant basis vectors for a surface point are denoted as a i , where a α are the two tangential vectors and a 3 as the normal vector with θ α and η as the respective coordinate components. The comma symbol in a subscript represents the partial derivative with respect to the surface parameters, for example, A ,β is the partial derivative of A with respect to θ β . The scalar product of two vectors p and q is denoted p · q, and the tensor product of these vectors is a second-order tensor H = p ⊗ q. Operation of a second-order tensor H on a vector p is given by Hp. The scalar product of two tensors, H and G, is denoted H : G. The notation ∥·∥ represents the usual (Euclidean) norm For a second-order tensor in its component form Nonlinear Magnetoelastostatics A brief review of the key equations of nonlinear magnetoelastostatics is provided, see Dorfmann and Ogden (2014); Pelteret and Steinmann (2020) for further details. Either the magnetic field, magnetic induction, or the Figure 1: Schematic of a thin shell that occupies regions B 0 and B in its reference and current configurations, respectively. The body is embedded in volumes V 0 and V in the two configurations connected by a deformation map χ. The two-dimensional parametric coordinate system is denoted by P and the local triads in the two configurations are also shown. magnetisation vector can be selected as the independent magnetic variable. The present work is established on the variational formulation based on the magnetic field. Consider a magnetoelastic body that occupies the regions B 0 ∈ R 3 and B ∈ R 3 in its reference and deformed configurations, respectively, with corresponding boundaries denoted as ∂B 0 and ∂B. A point X B ∈ B 0 is related to a point x B ∈ B through a one-to-one map χ B (X B ) : B 0 → B. The three-dimensional region B is enclosed within the region V, as schematically shown in Figure 1, so that the surrounding free space is with the Jacobian, J = detF > 0, such that dv = JdV, (2.2) where dV and dv are the volume elements in the reference and deformed configuration, respectively. The right Cauchy-Green tensor is defined by To facilitate the description of fields exterior to the body. Consider a fictitious deformation map χ F as χ F (X F ) : The boundaries ∂V 0 and ∂V coincide, implying The magnetostatic problem is governed by the Maxwell's equations involving the spatial magnetic induction b and magnetic field h given by divb = 0 and curlh = 0, in V, (2.6) along with the boundary (or jump) conditions, b · n = 0 and h × n = 0, on ∂B, where • represents jump in a quantity across the boundary with a unit outward normal vector n. Equation (2.6) 2 motivates the introduction of a magnetic scalar potential ϕ such that h = −gradϕ. (2.13) Using the transformations (2.10) in the constitutive relation (2.9), one obtains (2.14) Since, m and M vanish in the vacuum, the constitutive relation simplifies to Coupled magnetoelastic constitutive relations in the body are established by assuming a free energy density function per unit reference volume, Ω, that is of the form Ω = Ω (F , H). Objectivity and isotropy require that the free energy take the form Ω =Ω (C, H) = Ω (I 1 , I 2 , I 3 , I 4 , I 5 , I 6 ) , (2.16) where I 1 , I 2 , I 3 are scalar invariants of C, that is (2.18) Incompressibility requires J ≡ 1 and the energy density function is further simplified to Ω (I 1 , I 2 , I 3 , I 4 , I 5 , I 6 ) =Ὼ (I 1 , I 2 , I 4 , I 5 , I 6 ) . (2.19) The total Piola stress tensor is given for the case of incompressible solid as where p is the Lagrange multiplier due to the incompressibility constraint. The constitutive relation for the magnetic induction is given as The Maxwell stress tensor outside the body is given by where 1 is the spatial identity tensor. Using the Piola transform P M = Jσ M F −T , this can be written in the reference configuration as where 1 0 is the two-point identity tensor. Geometry Consider the magnetoelastic body to be a thin shell. Each point X B ∈ B 0 is mapped from the parametric domain defined by the coordinate system {θ 1 , θ 2 , η}. The Kirchhoff-Love hypothesis states that for thin shell structures, lines perpendicular to the mid-surface of the shell remain straight and perpendicular to the midsurface after deformation (see e.g. Niordson, 1985). Hence, assuming the shell has a thickness T (θ α ) in the reference configuration, the point X B can be defined using a point on the mid-surface S m of the shell, R ∈ S m , and the associated unit normal vector N as The points on the mid-surface in the deformed configuration denoted by r. The midsurface point in the deformed configuration corresponds to the mid-surface point in the reference configuration after a motion as shown in Figure 1, and can be expressed as where u denotes the mid-surface displacement. A point x B ∈ B can therefore be expressed as where d = λn, and λ is the through thickness stretch for a finitely deformed shell defined by where t(θ α ) is the shell thickness after deformation. Further, in B 0 , one can assume a form for the magnetic potential as implying a linear variation of the magnetic potential along the thickness of the thin shell. Therefore, the higherorder terms vanish, allowing one to express the Kirchhoff-Love assumption as which is similar in form to Equation (3.1). Table 1 provides a list of surface parameters used to describe the geometry of the shell and Table 2 presents the surface and volume elements of the shell. The expressions and associated derivations are elaborated on in Appendix A. The boundaries, ∂B 0 and ∂B, can be written as ∂B 0 = S t ∪ S b ∪ S ℓ , and ∂B = s t ∪ s b ∪ s ℓ , where the subscripts, t, b, and ℓ, represents the top, bottom, and lateral surfaces in the two configurations, and the top surface is the side of the boundary that is reached along the unit outward normal vector. The incorporation of the variation of the through-thickness stretch and the deformed normal in obtaining the reduced-order model for the soft thin magnetoelastic shell requires the evaluation of the integral: for an arbitrary T α θ β , as discussed in Sections 5.1 and 5.3. This integral can be expressed as: where C m represents the boundary of the curved mid-surface S m . This is elaborated upon further in Appendix B. Kinematics The deformation gradient and its inverse for a shell-point can be written so as to separate the thickness variable from the surface parameters as and Here B α β and b α β are the components of the curvature tensors K and κ, respectively. Further, using the relation, a α · n = 0, the right Cauchy-Green deformation tensor can be written as (3.14) The mid-surface right Cauchy-Green tensor is defined by are the components of the three-dimensional covariant and contravariant metric tensors on the mid-surface, respectively, with where the surface stretchâ is defined in Table 2. Divergence of the total stress tensor and magnetic induction vector The divergence of the total Piola stress tensor, as well as the magnetic induction vector, enters the governing equations for the Kirchhoff-Love magnetoelastic thin shell arising from the variational formulation involving the mechanical deformation and an independent field representing the magnetic component, that is, the magnetic field vector. The divergence of the total stress tensor can be expressed as Similarly, for the magnetic induction vector at a shell-point, The total Piola stress tensor and magnetic induction vector at the top and bottom boundaries are obtained by setting η = ±T /2 in their respective through-thickness expansions. Interface condition on magnetic field The continuity of magnetic field components projected onto the tangent space across the boundaries of the thin shell and the surrounding space is implied by Equation (2.12) 2 . Thereby, at the interfaces, by equating these components and using Equation (A.40), the resulting expression can be written as where GradΦ is evaluated in the free space. This imposes a constraint on the potential at the top, bottom, and lateral surfaces of the shell. Note, to the continuity of the potential across the boundaries of the thin shell with the surrounding space, this constraint is explicitly imposed and is not obtained from the modified variational setting while deriving the reduced-order theory, as discussed in Section 5. Variational formulation in three dimensions Definingχ = {χ, Φ, p} as the generalised set of the solution variables, the total potential energy of the system is written as (Dorfmann and Ogden, 2014): The external spatial magnetic induction is denoted as b e . Its normal component is prescribed on ∂V and its counterpart in the reference configuration is denoted as B e . The fourth term in Equation (4.1) representing the work done by the external magnetic induction is expressed in the current configuration, and using Equation (2.5) can be rewritten in the reference configuration as with the associated unit normals on the outer boundary of the free space, denoted by N ′ and n ′ in the reference and deformed configurations, respectively. The body force field per unit reference volume is B while t ℓ is the applied traction at C m . Also, C u m are the parts of the boundary where displacements are specified. p t (θ α ) and p b (θ α ) are the magnitudes of external pressure at the top and bottom surfaces of the shell, respectively, such that with n is the mid-surface unit normal in the current configuration. Let the set δχ = {δχ, δΦ, δp}. In the subsequent calculations, refer to Appendix C for details of the variation of key variables. From Equations (2.2) and (2.10), the first variation of the total energy is given by The Euler-Lagrange equations are obtained by setting δΠ = 0. • The first and second terms in Equation (4.4) can be combined as and taking into account the incompressibility condition (J = 1), and using Equations (2.20), (2.21), and (B.11), the expression reduces to On an application of the divergence theorem, (4.7) • The fourth term in Equation (4.4) can be written as Applying the divergence theorem, one obtains (4.10) From Equation (2.5) it follows that the second term is zero. The exterior magnetic induction is denoted as B ′ . Further, at the top and bottom boundaries, B ′ is denoted as B ′ t and B ′ b , respectively. Similarly, the Maxwell stress tensors at the top and bottom surfaces of the shell are given by P Mt and P Mb , respectively. For the lateral surface, the expressions for P Mℓ and B ′ ℓ are as follows: (4.11) • Since, B e is the applied magnetic induction, the fifth term in Equation (4.4) can be written as (4.12) The remaining terms in Equation (4.4), that is, the virtual work done by the dead load traction, body force and pressures are dealt with in Section 5.3, where their contributions to a modified variational form for a Kirchhoff-Love thin shell are discussed. Two dimensional variational formulation for magnetoelastic shells The following discussions outline the key steps involved in deriving the Kirchhoff-Love shell equations. It is important to note that the derived equations can achieve accuracy up to the linear order of the through-thickness parameter. The generalised set of solution variables is now extended to χ = {r, χ F , Φ 0 , Φ ′ , p 0 }, and define Here, the magnetic potential in B ′ 0 is denoted as Φ ′ , and the Lagrange multiplier is expressed as follows: The contribution of each integral in the first variation (4.4) to the above modified format is discussed in detail in the following subsections. Integrals related to the total Piola stress tensor Taking into account the definition of χ B , and following Equations (A.43) and (A.44), the first term in Equation (4.7) which is the domain term related to the total Piola stress tensor can be written as Here, M is defined in Table 2, which outlines the surface and volume elements of the shell. From Equations (3.18) and (A.46), and noting that, where The boundary term contribution related to the total Piola stress tensor at the top surface in Equation (4.7) can be expressed as Similarly, for the bottom surface, In Equations (5.6) and (5.7), the second and third integrals can be rewritten with the help of Equation (A.23) as Further, using Equations (A.62) and (A.65), the integral for the lateral boundary in Equation (4.7) can be represented as follows: (5.9) Integrals related to the Maxwell stress tensor Following eqn. (4.10), the terms concerning the Maxwell stress tensor at the inner boundaries of the free space can be written for the modified format as follows: (5.10c) Contribution arising from both stress tensors Now, considering the fictitious map χ F , the net contribution due to the Maxwell and total stress tensors to the modified variational form can be written as shown below: T δλP N · nA 1/2 dP = − P T λ P N · n a α · δa α A 1/2 dP, (5.13) with C α = T λ P N · n a α · δr, and the Christoffel symbols of the second kind Γ is defined in Table 1. Similarly, Equation (5.12) 2 can be written as with D α = T λ P N · a α n · δr. Moreover, following Equation ( Integrals related to the magnetic induction vector inside the shell In Equation (4.7), using Equation (3.20), the domain term involving the magnetic induction vector can be written as (5.16) From Equations (3.6) and (3.7), the corresponding boundary term at the top surface is given by The continuity of the magnetic potential at the shell boundaries is enforced, allowing Φ 1 to be expressed as as the potential at the top and bottom boundaries, respectively. The first integral can be rewritten as follows: and from Equation (A.52), noting that, Incorporating Equations (5.18) and (5.20), and omitting the subscripts t and b for the exterior magnetic potential, one obtains Similarly, for the bottom surface of the shell, Note that the effect of the external field on the response of the soft magnetoelastic thin shell, leads to integrals over the top, bottom, and mid-surfaces during the derivation of the reduced-order model. In Equation (4.7), the contribution corresponding to the lateral surface of the shell can be expressed as (5.23) Contribution arising from the magnetic induction vector In Equation (4.10), for the exterior magnetic induction, the integral over the lateral surface can be written as follows: The total contribution resulting from the magnetic induction vector in the shell and free space in the modified variational form can now be expressed as Contribution to the first variation due to external loads The terms related to the external mechanical loads, as they appear in Equation (4.4), will now be expanded upon. Integrals related to externally applied pressure In the present work, a noteworthy aspect is the differentiation of the applied pressures at the top and bottom surfaces of the shell structure, instead of directly considering them on the mid-surface during the derivation of the shell system of equations. From Equations (A.52) and (A.53), the virtual work due to the external pressure at the top surface of the shell is given by Using the expression for µ, and taking into account that n · n = 1, n · δn = 0, and λ a 1/2 = 1, Equation (5.26) can be rewritten as Similarly, the virtual work due to the external pressure at the bottom surface of the thin shell can be expressed as Therefore, using Equation (A.44), with p = p t + p b 2 , and the integral over the mid-surface excluding the higher-order terms can be further simplified as Integral related to the dead load traction For the lateral surface of the shell, the contribution due to the dead load traction applied at the bounding curve of the mid-surface can be written as (5.32) Integral related to the body force The virtual work due to the body force per unit volume can be further simplified as Net contribution due to the external loads The applied magnetic induction, as defined by Equation (4.12), along with the overall role of external mechanical loads on the modified variational framework, is considered. Consequently, the combined effect of the external stimulus can be expressed as (5.34) Contribution to the first variation due to the incompressibility constraint For the volume-preserving magnetoelastic body, when expanding the Lagrange multiplier along the thickness of the thin shell and considering only the first-order terms with respect to the through-thickness parameter in the modified variational form, the contribution arising from the incompressibility constraint can be expressed as follows: (5.35) 6 Governing equations for the Kirchhoff-Love magnetoelastic shell and accompanying free space The equations for a nonlinear magnetoelastostatic Kirchhoff-Love thin shell are now derived using the modified variational form. In Section 5, the contributions of the stress tensors, external loads, and magnetic induction vector to the modified variational format were determined. By adding Equations (5.11), (5.25), (5.34), and (5.35), the variation of the total potential energy of the system is obtained. The state of magnetoelastic equilibrium is obtained by considering the variable δ χ as arbitrary, which must correspond to an extremum of δΠ. In other words, the first variation of the potential energy functional must be zero. Now, by the arbitrary variation δr, the shell-system of equations are obtained as follows: Also, considering the arbitrary variations δΦ 0 and δΦ ′ in the shell and the free space, respectively, the equations obtained are given by (6.2d) and (6.3b) The arbitrary variation δχ F and δp 0 leads to Equation (6.4) 2 returns the incompressibility relation (3.17) at the mid-surface, as derived in Section 3 . Furthermore, by neglecting the higher-order terms, the condition on the magnetic potential at the surfaces of the shell structure, as given by Equation (3.22), can be rewritten as: Finite inflation and magnetisation of a long cylindrical shell The main objective of this section is to illustrate via an example how the equilibrium equations for a Kirchhoff-Love magnetoelastic thin shell, as introduced in Section 6, can be used to derive the response equations for the boundary-value problems at hand. Consider the problem of an inflating infinite magnetoelastic thin cylindrical shell. The body is subjected to two loading situations, as depicted in Figure 2. The first scenario is purely mechanical case where external pressures are applied at the inner and outer surfaces of the shell structure. The second scenario is the magnetoelastic case, with a wire carrying a current i along the axis of the thin cylinder. The inner boundary of the free space is at an infinitesimal distance ∆ from the wire while the outer boundary extends to infinity, as shown in Figure 2. The axisymmetric deformation of an infinite cylinder under a unit axial stretch is given by Here, θ and z are the deformed coordinates corresponding to their azimuthal and axial counterparts in the reference configuration (Θ and Z, respectively). The unit vectors along the axial and radial directions are denoted by e z and e ρ , respectively. Additionally, R and r represent the radius at the mid-surface of the cylindrical shell in the two configurations. The displacement vector is given by u = u(ρ)e ρ . The covariant and contravariant vectors at the mid-surface in the two configurations, as well as the reference and deformed normals, are given by = a 1 = re θ , a 2 = e z , a 1 = 1 r e θ , a 2 = e z , where e θ is the azimuthal unit vector. The normal vectors in both the configurations coincide for the deforming cylinder, implying δn = 0. The components of the covariant and contravariant metric tensors at the mid-surface in the reference configuration are respectively and similarly, in the deformed configuration, a αβ = r 2 0 0 1 and a αβ = r −2 0 0 1 , (7.4) along with the determinant of the covariant metric tensors at the mid-surface given by A = R 2 , and a = r 2 . (7.5) From Equation (6.4), one obtains with λ θ = r/R as the azimuthal stretch. The non-zero components of the curvature tensor at the mid-surface are Furthermore, A generalised neo-Hookean constitutive relation for magnetoelasticity (Dorfmann and Ogden, 2014) is chosen where and β = nα and µ s is the shear modulus of the material. The constants α and β must be negative to ensure stability. Therefore, for convenience, α = −1, and n ∈ R + . From Equation (2.20), the total Piola stress can be calculated as and from Equation (2.21), the magnetic field induction vector in the reference configuration of the shell is Now, the zeroth and first-order terms along the thickness of the shell of the total Piola stress are given as (7.12b) and similarly, the components of the magnetic induction vector are Here the applied magnetic field in the spatial configuration at a shell-point is h = i 2π [r + ηλ] 2πR 2 e θ with i ∈ R + , and from the relation, H = F T h = H 0 + ηH 1 , the following expressions are obtained: (7.14) The components of the deformation gradient and its inverse are calculated using Equations (3.9) and (3.11) as The expressions for the zeroth and first-order components of the deformation gradient and its inverse are required for evaluating the total Piola stress and the referential magnetic induction vector. Therefore, considering the shell system of equations, specifically equations (6.1a), (6.1c), and (6.1d), for the purely mechanical loading of the infinite soft cylinder in the absence of body force and pressure at the outer boundary, the response equation is given by This relationship is plotted in Figure 3 for different shell thickness values ( T = T /R) and external pressure values ( p = p t /µ s ). As the pressure difference (∆p = [p b − p t ]/µ s ) between the inner and the external shell surface increases, the stretch λ θ increases monotonically until a critical value of ∆p corresponding to a limit point instability is reached. At this point massive changes in inflation occur for a minor change in the applied pressure. Similar limit point instabilities have been observed for inflation of thin hyperelastic shells as well as soft cylindrical cavities (Kiendl et al., 2015;Cheewaruangroj et al., 2019;Mehta et al., 2022). The critical limit point pressure reduces as the shell thickness is reduced. We further demonstrate the distinction between considering the pressure on top and bottom surfaces of the shell separately as opposed to the common convention of considering a pressure difference on the mid-surface. The shell's response to applied pressure difference ∆p can significantly change by varying the pressure on the external surface p t /µ s . Reducing the shell thickness brings these response curves closer together, as observed from T /R = 1/30 to 1/10. For the magnetoelastic deformation of the cylindrical shell due to an applied current along its axis, the equilibrium Equations (6.2a), (6.2c), and (6.2d), governing the magnetic induction vector are trivially satisfied. Furthermore, by considering the shell equilibrium Equations (6.1a), (6.1c), and (6.1d), the following response relation for the system is obtained. Since, β = −n and λ < 1, it is evident that the condition, (7.18) must be satisfied by the constitutive parameter n to ensure a physical deformation. This is further elaborated by plotting n against the azimuthal stretch for multiple T /R values in Figure 4(a). Based on this analysis, n > 0.02 is necessary for an inflating cylindrical shell to ensure that the condition (7.18) is satisfied for all deformation states. The deformation of the magnetoelastic cylinder based on Equation (7.17) is shown in Figure 4(b) for different T /R and n values. Application of magnetic field via the conductor causes the cylinder to inflate and the amount of inflation is higher for larger values of the coupling parameter n. For a given n ̸ = 0, there is a critical value of applied current (i/R µ 0 /µ s ) beyond which the cylinder experiences rapid inflation, akin to a limit point instability (Barham et al., 2008;Reddy and Saxena, 2017). For n = 1.5, this critical value is i/R µ 0 /µ s ≈ 3.64, while for n = 0.5, this is close to 6.36. Notably, when n = 0 for T /R = 1/30, the cylinder exhibits slower inflation due to a weak magnetoelastic coupling, eventually saturating at λ θ ≈ 1.5. This behaviour is alternatively explained from the requirement of the radial expansion to satisfy the imposed condition (7.18) on n, as presented in Figure 4(a). Furthermore, reducing the cylinder's thickness below a certain magnitude has a negligible effect for a given n, as indicated by the overlapping response curves for T /R = 1/20 and 1/30 at n = 0.5. The constitutive relation (7.13b) 1 for the azimuthal component of the magnetic induction at the shell's midsurface can be expressed as The above indicates that ζ 0 remains positive as the cylinder deforms. Additionally, the azimuthal components of the exterior magnetic induction at the outer and inner boundaries of the cylindrical shell, denoted as B ′ tθ and B ′ bθ , respectively, are given by (7.20b) The above expressions, along with the azimuthal magnetic induction at the mid-surface of the shell, are graphically presented in Figure 4(c) for T /R = 1/30. Since T /R = 1/30 ≪ 1 and λ < 1, ζ t ≈ ζ b and the two curves overlap, hence, only ζ t is plotted. The plot demonstrates that the magnetic induction at the mid-surface increases monotonically as the applied magnetic field (via the electric current) is increased. Beyond a critical value of the applied current, the magnetic induction ζ 0 increases abruptly similar to a limit point instability for the mechanical problem. In the surrounding space, the magnetic induction at the shell boundaries exhibits an initial monotonic increase, followed by a gradual decrease, culminating in a sharp decline in magnitude. This decline occurs at relatively higher operational currents for smaller values of n, with slightly higher magnetic induction observed at this juncture for lower n. Additionally, for a specific n, the magnitude of the exterior magnetic induction at the inner surface is marginally greater than that at the outer boundary, although this difference diminishes as the cylinder inflates. The radial, azimuthal, and axial components of the zeroth-order term of the total Piola stress (P 0 ) are The radial component of the Maxwell stress tensor in the surrounding space can be determined at the shell boundaries using Equation (2.23) and is given by (7.23) at the outer and inner surfaces of the cylinder, respectively. Figure 4(d) presents the principal components of P 0 for n = 0.5 and T /R = 1/30. The axial stress shows a monotonic increase with the applied magnetic loading until the limit point at i/R µ 0 /µ s = 6.36. The radial component of the total Piola stress remains negative suggesting compression. It drops to a minimum value of −0.26 before increasing slightly until the limit point. The azimuthal component displays a non-monotonic behaviour. It initially increases and remains positive for low magnetic loads but then starts to decrease and becomes negative for i/R µ 0 /µ s > 5.2. It is notable that the radial stress is non-negligible compared to the other components, indicating the presence of inaccuracies when employing plane stress assumptions in modelling soft magnetoelastic shells. Also, the zeroth-order component of the Lagrange multiplier is plotted, and it decreases as the magnetoelastic cylindrical shell inflates. The presence of compressive stresses in the azimuthal direction indicates a possibility of wrinkling instability, the analysis of which will form part of a future study. Concluding Remarks In this work, The governing equations for the large deformation of Kirchhoff-Love magnetoelastic thin shells have been rigorously derived. The free space in which the magnetostatic energy is bounded to finite volumes is accounted for. The equilibrium equations have been obtained using the derived theory approach. This point of departure was a variational form for a three-dimensional continuum magnetoelastic body involving mechanical deformation, magnetic field, and a Lagrange multiplier in the presence of body force, dead-load traction along the bounding curve of the mid-surface, external pressures at the top and bottom surfaces, and an external magnetic field. Treating the shell as a stack of surfaces, the general deformation map in the body has been restated in terms of a point on the deformed mid-surface. This requires an additional term that incorporates the through-thickness stretch and the deformed normal (i.e., the first director). By defining a new set of generalised solution variables and thereby modifying the variational form, the shell equilibrium equations have been obtained. The thickness variable has been separated from the surface parameters, and the field variables expanded along the thickness of the thin shell. Additionally, the governing equations for the corresponding three-dimensional free space are derived. The new formulation relies on a Kirchhoff-Love type kinematic assumption for the magnetic scalar potential thereby ensuring a consistent derivation of the governing equations. The top and bottom surfaces of the shell are considered in addition to the mid-surface to consistently account for the magnetic field in vacuum. This leads to a departure from the commonly used plane-stress assumption in thin shell theory. Furthermore, a distinction between the hydrostatic pressure applied on the top and bottom surfaces has been considered. Variations of the through-thickness stretch and the deformed normal introduces richness into the formulation together with additional complexity through the distribution of the parametric derivative of the mid-surface position vector to the lateral bounding curve. The novel magnetoelastic shell theory and implications of the factors have been illustrated by analysing the inflation of a cylindrical magnetoelastic shell. Capabilities of the present theory to model large deformation and limit point instabilities have been demosntrated. The possibility of wrinkling instabilities due to the presence of compressive in-plane stresses in the shell have been detailed. The present analysis provides a new perspective into a strongly-coupled shell system of equations, which is challenging to obtain due to strong kinematic and constitutive nonlinearities. The geometrically exact formulation ensures a high level of accuracy. The focus here is on formulating and demonstrating the capabilities of the derived equations. The derivation from a variational formulation ensures that the theory is amenable for numerical implementation via the finite element method. Details of the numerical implementation will be presented in a future contribution. A.2 The unit alternator and permutation symbol In general, for a surface tensor Q = Q αβ A α ⊗ A β , the surface inverse Q −1 defined from where Q = det Q αβ , and the so-called unit alternator given as Further, the permutation tensor is defined by A.3 The natural basis at a shell-point A point x B ∈ B can be written as where d = λn and λ = t T . The covariant basis vectors at a point X B in the shell are given by Again, the covariant basis vectors at a point x B in the shell are Here the long-wave assumption has been considered (Kiendl et al., 2015;Liu et al., 2023). That is, While this assumption is strong, it is reasonable since the thickness of the shell is typically very small, resulting in negligible out-of-plane shearing and localised necking. Therefore, where i = a β ⊗ a β = a β ⊗ a β denotes the projection onto the tangent plane of s m , the deformed counterpart of S m . Also, Again, N ,β and n ,β is given by with the surface curvature tensors in the two configurations defined by where B βδ = N · A β,δ , and b βδ = n · a β,δ , (A.31) and further, For a point in the shell, the components of the covariant and contravariant metric tensors in the reference configuration are G αβ = G α · G β and G αβ = G α · G β , (A.34) with their deformed counterparts as g αβ = g α · g β and g αβ = g α · g β , (A.35) where G α and g α denotes the contravariant basis vectors in the two configurations defined by G α · G β = δ α β , and g α · g β = δ α β . and µ −1 can be expanded as A.4 The volume and surface elements The volume element in the reference configuration can be expressed as If the bounding curve C m of the mid-surface S m is characterised by the arc-length parameter l, then the infinitesimal length dl between two points R(θ 1 , θ 2 ) and R(θ 1 + dθ 1 , θ 2 + dθ 2 ) is given by dl = R(θ 1 + dθ 1 , θ 2 + dθ 2 ) − R(θ 1 , θ 2 ) = A γ dθ γ = A α dθ α · A β dθ β = A αβ dθ α dθ β . (A.54) The tangent vector at a point R on C m is defined as implying that τ is a unit tangent vector. Further, define and following the relation, E αβ E ηδ δ η α = δ α η δ β δ − δ α δ δ β η δ η α = δ β δ , (A.60) implying that ν is the in-plane unit normal to τ on C m , and ν = τ × N . (A.61) An elemental area, dS ℓ , at a point X B on the lateral surface is given by This establishes a relation between the integral over the parametric domain and the line integral along the boundary of the curved mid-surface.	2023-08-25T06:42:09.846Z	2023-08-18T00:00:00.000	{ "year": 2023, "sha1": "9f91fa8da119d085c7b036bc4fcd39502c57625c", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "9f91fa8da119d085c7b036bc4fcd39502c57625c", "s2fieldsofstudy": [ "Physics", "Engineering" ], "extfieldsofstudy": [ "Physics" ] }
202408132	pes2o/s2orc	v3-fos-license	A Smart Tongue Depressor-Based Biosensor for Glucose The development of new bioelectronic platforms for direct interactions with oral fluid could open up significant opportunities for healthcare monitoring. A tongue depressor is a widely used medical tool that is inserted into the mouth, where it comes into close contact with saliva. Glucose is a typical salivary biomarker. Herein, we report—for the first time—a tongue depressor-based biosensor for the detection of glucose in both phosphate buffer and real human saliva. Carbon nanotubes (CNTs) are attractive electronic materials, with excellent electrochemical properties. The sensor is constructed by printing CNTs and silver/silver chloride (Ag/AgCl) to form three electrodes in an electrochemical cell: Working, reference, and counter electrodes. The enzyme glucose oxidase (GOD) is immobilized on the working electrode. The glucose detection performance of the sensor is excellent, with a detection range of 7.3 μM to 6 mM. The glucose detection time is about 3 min. The discretion between healthy people’s and simulated diabetic patients’ salivary samples is clear and easy to tell. We anticipate that the biosensor could open up new opportunities for the monitoring of salivary biomarkers and advance healthcare applications. Introduction The development of new types of biosensors has attracted considerable interest in various healthcare, environmental monitoring, and defense applications [1][2][3]. Saliva is easy to obtain using noninvasive methods, and it contains a variety of biomarkers related to diseases or health statuses [4][5][6][7]. Thus, the development of a simple method for the detection of salivary biomarkers is highly desirable. Glucose is a typical salivary compound, and it can be used as a typical analyte for the construction of biosensors [8][9][10]. Many biosensors on conventional substrates, such as ceramics and glass, have been developed [11][12][13]. Recently, the potential utility of several types of nonconventional substrates for the detection of glucose have been studied. These include cotton fabric [14], plastic film [15], plastic centrifuge tubes [14], and contact lenses [16]. A variety of electronic materials have also been studied for the construction of sensing electrodes. These materials include carbon graphite [14,17,18], carbon nanotubes (CNTs) [19,20], graphene [21], polymers [22,23], and indium tin oxide (ITO) [24]. Many of these materials have shown excellent biosensing performances. For example, studies have shown that CNTs, a typical type of electronic material, have excellent electrochemical properties and that they can be used for the construction of amperometric biosensors [25,26]. A tongue depressor is a widely used medical tool that is inserted into the mouth, where it comes into close contact with saliva [27,28]. It is cost-effective, lightweight, and convenient to use for frequent examinations [29]. A tongue depressor could provide a unique platform for building biosensing devices. However, to the best of our knowledge, there are no reports on the construction of any medical-related devices on a tongue depressor. In this work, we present-for the first time-a tongue depressor-based amperometric biosensor for the determination of glucose. Since a tongue depressor is a commonly used oral medical tool, it can directly contact and detect human saliva. For example, for diabetic patients, the glucose concentration can be tested by placing the tongue depressor-sensor in the mouth instead of using other invasive methods. The sensor is constructed based on printing CNTs and Ag/AgCl to form three electrodes in an electrochemical cell: A CNT working electrode, an Ag/AgCl reference electrode, and a CNT counter electrode. Glucose oxidase (GOD) is immobilized on the working electrode. The reference electrode provides a stable potential. The working electrode and the counter electrode form a circuit. A constant potential is applied between the working electrode versus the reference electrode, and the current in the circuit is measured. The oxidation of glucose is catalyzed by GOD to produce H 2 O 2 . H 2 O 2 generates a current response, as the detectable signal, via the oxidation reaction at the working electrode. A calibration curve is plotted between the current reponse and the concentration of glucose. We also tested real human saliva with and without added glucose to prove our sensor's capability in distinguishing between healthy people and those who suffer from diabetes. Sensing Electrode Preparation First, three parallel rectangles were drawn on the tongue depressor to mark the positions of three electrodes, each with a length of 20 mm, a width of 2 mm, and 2 mm apart from each other. The tongue depressor was then covered with a thin layer of epoxy and dried at room temperature (27 • C) for 24 h to ensure that it was totally waterproof. The working and counter electrodes were prepared manually with CNT (0.1 mg/mL in toluene; CNT: electrical conductivity >100 s/cm, optical density: 10-20 nm, length: 10-30 µm, ash: <1.5 wt%, specific surface area: >200 m 2 /g, purity: >95%). The reference electrode was prepared with Ag/AgCl ink (50:50, Gwent Electronics Materials Ltd.). Scotch tapes were used to cover the area between and around the three rectangles, leaving out the space for the electrodes. The CNT solution was carefully dropped on the locations for the working and counter electrodes with an autopipette. Then, the Ag/AgCl ink was painted on the location for the reference electrode with a foam-tipped swab. The printed electrodes were dried in an oven at 80 • C for half an hour, and then cooled at room temperature. A chamber was casted with a glue gun at the front end of the electrodes where solution would be added. The device was further treated with a UV ozone cleaner for 20 min to change the hydrophobic surface to be hydrophilic. Enzyme Functionalization A 2% glutaraldehyde and a 10 U µL −1 GOD (from Toyobo Co., Ltd.) were mixed in a volume ratio of 1:1. Then, 3 µL of the mixture was placed evenly on the surface of the front end of the working Sensors 2019, 19, 3864 3 of 10 electrode, with an area of 2 mm in width and 10 mm in length. The sensor was then dried in a refrigerator at 4 • C overnight. On the second day, it was taken out of the refrigerator and incubated with buffer at room temperature for about 1 h before sensing measurement. Sensing Measurements The measurement of the tongue depressor sensor for H 2 O 2 or glucose was performed using a potentiostat CHI660e at room temperature. A constant potential of 0.6 V was applied between the working versus the reference electrode. First, 90 µL of phosphate buffer (50 mM, pH 7.0) was placed on the sensor, and 10 µL of H 2 O 2 or glucose was injected into the buffer droplet. A series of different concentrations were injected sequentially. The current-versus-time curves were recorded, and a calibration curve between the steady-state current response and the concentration of H 2 O 2 or glucose was plotted. The cyclic voltammetry of the sensor was studied in a buffer solution containing 2 mM of H 2 O 2 or glucose, or without H 2 O 2 or glucose. A small volume of glucose in water with a high concentration was added into the artificial saliva to mimic the saliva environment of diabetic patients. It was noticed that the slight dilution of the saliva with water had almost no change on the current baseline, compared with that of saliva.The rise in current was then recorded. For the comparision of the sensor with a spectrophotometric method, saliva was added to the solution of glucose catalytic enzyme, catalase, and 4-aminoantipyrine at room temperature for 15 min. After the reaction, a colored compound was formed, and the characteristic absorption peaks at 505 nm were detected by a spectrophotometer. Figure 1 shows a schematic illustration and images of the tongue depressor-based sensor. Figure 1a shows a schematic illustration for the sensor preparation. A tougue depressor was first painted with epoxy for insulation from water (1). Then, CNTs were printed on the substrate (2), followed by the printing of Ag/AgCl ink (3). Coppor wires were used to connect the electodes with the potentiostat (4). The sensor was then treated with UV ozone to change the surface property from being hydrophobic to be hydrophilic (5). After this, enzymes were immobilized on the working electrode (6). As can be seen in Figure 1b, all three electrodes were printed onto the tongue depressor in parallel and covered by a buffer droplet. The analytes were added into the buffer droplet and diffused to the electrode surfaces to induce the oxidation, reduction, and electron transfer on the electrodes. The camera image ( Figure 1c) shows the working, reference, and counter electrodes successfully painted onto the surface of the tongue depressor in parallel. The electrodes are thin, straight, and have relatively sharp edges. The two electrodes in black color are the CNT electrodes that function as the working and counter electrodes. The grey color is the Ag/AgCl electrode that functions as the reference electrode. The optical images show the boundary between the wooden substrate and the CNT electrodes (Figure 1d), and the boundary between the wooden substrate and the Ag/AgCl electrode ( Figure 1e). These images show that all the electrodes had uniform geometric shapes, and were successfully printed on the surface of the tongue depressor. Results and Discussion The detection mechanism is based on an enzymatic reaction. Glucose is oxidized by O 2 to produce glucolactone H 2 O 2 by GOD. H 2 O 2 reacts at the working electrode at a positive potential: From the amount of generated H 2 O 2 , the concentration of glucose can be determined. To explore the detection performance of H 2 O 2 or glucose at different potentials, cyclic voltammetry was studied for the tongue depressor-based sensors, as shown in Figure 2. Figure 2a presents the cyclic voltammogram of the sensor for 2 mM H 2 O 2 in the buffer. Figure 2b shows the cyclic voltammogram without H 2 O 2 in the buffer. Figure 2c shows the cyclic voltammograms of GOD-immobilized sensor for 2 mM glucose in buffer. Figure 2d shows the cyclic voltammogram without glucose in the buffer. GOD catalyzed the oxidation of glucose to generate H 2 O 2 , which was then oxidized/reduced at the working electrode to generate a current change. From the comparison of Figure 2a,b, and from the comparision of Figure 2c,d, we can see that in the presence of H 2 O 2 or glucose, the currents are higher than that without H 2 O 2 or glucose at the same potential. The patterns for curves under 30 mV/s to 100 mV/s scanning rates were slightly different, as shown in both Figure 2a,c. This may be due to the different oxidation/reduction processes caused by different scanning rates, and the current is proportional to the squre root of the scanning rate. Comparing the current responses at a same potential in Figure 2a,c, it exhibits that glucose generated a smaller response than that of H 2 O 2 . This was expected, as H 2 O 2 was oxidized/reduced directly on the electrode, whereas glucose went through the enzymatic reaction to generate H 2 O 2 , and H 2 O 2 was oxidized/reduced on the electrode afterwards. The patterns of the curves are associated with the properties of the electronic inks, and these curves have similar patterns. As can be seen from the curves, with the increase of the potential, the current became larger. A potential of 0.6 V was chosen in this study, as this was sufficient for the oxidation or reduction process of H 2 O 2 . Enzyme Functionalization A 2% glutaraldehyde and a 10 U μL −1 GOD (from Toyobo Co., Ltd.) were mixed in a volume ratio of 1:1. Then, 3 μL of the mixture was placed evenly on the surface of the front end of the working electrode, with an area of 2 mm in width and 10 mm in length. The sensor was then dried in a refrigerator at 4 °C overnight. On the second day, it was taken out of the refrigerator and incubated with buffer at room temperature for about 1 h before sensing measurement. Sensing Measurements The measurement of the tongue depressor sensor for H2O2 or glucose was performed using a the reaction, a colored compound was formed, and the characteristic absorption peaks at 505 nm were detected by a spectrophotometer. Figure 1 shows a schematic illustration and images of the tongue depressor-based sensor. Figure 1a shows a schematic illustration for the sensor preparation. A tougue depressor was first painted with epoxy for insulation from water (1). Then, CNTs were printed on the substrate (2), followed by the printing of Ag/AgCl ink (3). Coppor wires were used to connect the electodes with the potentiostat (4). The sensor was then treated with UV ozone to change the surface property from being hydrophobic to be hydrophilic (5). After this, enzymes were immobilized on the working electrode (6). As can be seen in Figure 1b, all three electrodes were printed onto the tongue depressor in parallel and covered by a buffer droplet. The analytes were added into the buffer droplet and diffused to the electrode surfaces to induce the oxidation, reduction, and electron transfer on the electrodes. The camera image (Figure 1c) shows the working, reference, and counter electrodes successfully painted onto the surface of the tongue depressor in parallel. The electrodes are thin, straight, and have relatively sharp edges. The two electrodes in black color are the CNT electrodes that function as the working and counter electrodes. The grey color is the Ag/AgCl electrode that functions as the reference electrode. The optical images show the boundary between the wooden substrate and the CNT electrodes (Figure 1d), and the boundary between the wooden substrate and the Ag/AgCl electrode (Figure 1e). These images show that all the electrodes had uniform geometric shapes, and were successfully printed on the surface of the tongue depressor. Figure 3a presents the current-versus-time signal response curve of the tongue depressor sensor for detecting H 2 O 2 . When the current baseline with buffer was stable, H 2 O 2 was injected into the buffer droplet. Then it diffused and reached the surface of the electrode, where it was then oxidized and generated a current response. When the entire diffusion and oxidation process became stable, the current arrived at a new steady state. The sensor response for H 2 O 2 was rapid, and the entire detection took less than 5 min. At lower concentrations, the sensor response was even quicker, owing to the shorter diffusion process and reaction time. At C1 and C2, it took less than 1 min for the sensor to reach a stable current. The current response increased as the concentration of H 2 O 2 increased, and it was proportional to the H 2 O 2 concentration. As shown in Figure 3a, from C5 to C8, the current increases to the additions of 1 mM H 2 O 2 were similar. Figure 3b illustrates the calibration curve for sensing H 2 O 2 . As can be seen in the figure, a linear relationship was obtained between the current response and the H 2 O 2 concentration, with a concentration range of 0.1-5 mM, a slope of 3.676 µA mM −1 , and an R 2 of 0.9926. The detection limit of the tongue depressor biosensor for H 2 O 2 was calculated to be 4.7 µM, 6.12 µM, and 4.08 µM (signal-to-noise ratio of 3), with an average of 4.97 µM and a standard deviation of 1.04 µM. The error bars were from repetitive experiments made by three different electrodes. It can be seen that the signal error between different electrodes is less than 15%. The results demonstrate that a tongue depressor-based sensor showed a rapid and sensitive detection of H 2 O 2 . The findings point to the possibility of constructing various biosensors based on tongue depressors via the immobilization of bioreceptors, such as enzymes, peptides, and antibodies. Figure 4 presents the characterization of the GOD-immobilized tongue depressor sensor for the detection of glucose. The immobilized GOD catalyzes the oxidation of glucose to generate H 2 O 2 . Therefore, through the detection of H 2 O 2 , glucose can be detected indirectly. Similarly, a buffer droplet was first placed on the sensor, and after the current baseline achieved a stable status, a series of glucose solutions were added sequentially to result in the increases of current. Figure 4a presents the current-versus-time curve for the detection of glucose. The sensor exhibited a rapid measuring time. Although it took around 5 min to detect 6 mM glucose, it could detect 0.3 mM glucose within 1 min. As the glucose concentration increased, the current response increased accordingly. As shown from C5 to C9 in Figure 4a, the biosensor showed similar current increases when adding several times of 1 mM glucose. Figure 4b presents the calibration curve of the tongue depressor biosensor for detecting glucose. A linear relationship was obtained between the current response and the glucose concentration, with a concentration range of 0.1-6 mM, a slope of 2.281 µA mM −1 , and an R 2 of 0.9968. The signal saturated at a concentration of 15 mM, whicg may be because of the saturation of the enzymatic reaction. The detection limit of the tongue depressor biosensor for glucose was calculated to be 7.3 µM, 10.96 µM, and 8.76 µM (signal-to-noise ratio of 3), with an average of 9.01 µM and a standard deviation of 1.84 µM. The results presented herein demonstrate that the tongue depressor-based biosensor has been successfully constructed and can detect glucose sensitively and rapidly. The sensing performances were comparable to the detection ranges and detection limits from other reports in the literature by using different electronic and substrate materials [30,31]. The detection mechanism is based on an enzymatic reaction. Glucose is oxidized by O2 to produce glucolactone H2O2 by GOD. + ⎯ + H2O2 reacts at the working electrode at a positive potential: H2O2 − 2e -= 2H + + O2↑. From the amount of generated H2O2, the concentration of glucose can be determined. To explore the detection performance of H2O2 or glucose at different potentials, cyclic voltammetry was studied for the tongue depressor-based sensors, as shown in Figure 2. Figure 2a presents the cyclic voltammogram of the sensor for 2 mM H2O2 in the buffer. Figure 2b shows the cyclic voltammogram without H2O2 in the buffer. Figure 2c shows the cyclic voltammograms of GOD-immobilized sensor for 2 mM glucose in buffer. Figure 2d shows the cyclic voltammogram without glucose in the buffer. GOD catalyzed the oxidation of glucose to generate H2O2, which was then oxidized/reduced at the working electrode to generate a current change. From the comparison of Figure 2a,b, and from the comparision of Figure 2c,d, we can see that in the presence of H2O2 or glucose, the currents are higher than that without H2O2 or glucose at the same potential. The patterns for curves under 30 mV/s to 100 mV/s scanning rates were slightly different, as shown in both Figure 2a,c. This may be due to the different oxidation/reduction processes caused by different scanning rates, and the current is proportional to the squre root of the scanning rate. Comparing the current responses at a same potential in Figure 2a,c, it exhibits that glucose generated a smaller response than that of H2O2. This was expected, as H2O2 was oxidized/reduced directly on the electrode, whereas glucose went through the enzymatic reaction to generate H2O2, and H2O2 was oxidized/reduced on the electrode afterwards. The patterns of the curves are associated with the properties of the electronic inks, and these curves have similar patterns. As can be seen from the curves, with the increase of the potential, the current became larger. A potential of 0.6 V was chosen in this study, as this was sufficient for the oxidation or reduction process of H2O2. Figure 3a presents the current-versus-time signal response curve of the tongue depressor sensor for detecting H2O2. When the current baseline with buffer was stable, H2O2 was injected into the buffer droplet. Then it diffused and reached the surface of the electrode, where it was then oxidized and generated a current response. When the entire diffusion and oxidation process became stable, the current arrived at a new steady state. The sensor response for H2O2 was rapid, and the entire detection took less than 5 min. At lower concentrations, the sensor response was even quicker, owing to the shorter diffusion process and reaction time. At C1 and C2, it took less than 1 min for the sensor to reach a stable current. The current response increased as the concentration of H2O2 increased, and it was proportional to the H2O2 concentration. As shown in Figure 3a, from C5 to C8, Figure 5 shows the working stability of a tongue depressor glucose sensor. The sensor was studied for measuring a same glucose concentration for more than 50 times, where 20 µL of 50 mM glucose was added to 180 µL buffer and the current response was recorded. The response of the first detection is regarded as 100%, and the following results are showed in percentages based on the first response. It can be seen that the sensor maintained an excellent response for about 90% of the initial signal after detecting glucose for 50 times. This demonstrates that the sensor has an excellent working stability for the detection of glucose continuously. / 10 μM and a standard deviation of 1.84 μM. The results presented herein demonstrate that the tongue depressor-based biosensor has been successfully constructed and can detect glucose sensitively and rapidly. The sensing performances were comparable to the detection ranges and detection limits from other reports in the literature by using different electronic and substrate materials [30,31]. Figure 5 shows the working stability of a tongue depressor glucose sensor. The sensor was studied for measuring a same glucose concentration for more than 50 times, where 20 μL of 50 mM glucose was added to 180 μL buffer and the current response was recorded. The response of the first detection is regarded as 100%, and the following results are showed in percentages based on the first response. It can be seen that the sensor maintained an excellent response for about 90% of the initial signal after detecting glucose for 50 times. This demonstrates that the sensor has an excellent working stability for the detection of glucose continuously. Figure 6 shows the comparison between tongue depressor amperometric sensors and a spectrophotometer. The actual concentrations were 0.2 mM, 0.5 mM, and 0.8 mM. Figure 6a show the graph of the sensor for detecting the concentration of glucose in saliva. Different concentrations of glucose in 200 μL of saliva were dropped onto the electrode surface. After the addition of glucose and waiting for 5 min to decompose glucose into H2O2, the concentration of H2O2 became uniform on the electrode in saliva, and the current baseline was detected for 100 s at 0.6 V. As shown in the figure, the signal is well correlated with the concentration of glucose. The glucose's concentration in saliva was derived from the calibration curve in Figure 4. Figure 6b shows the comparision of the detection performances by using the sensor (pink) and a spectrophotometer (blue). The blue is the result from a spectrophotometer. As shown in the figure, both methods show comparable results, which are about the same as the standard concentrations of glucose. These results demonstrate that the sensor is an effective detection tool for measuring glucose concentrations in saliva. Figure 6 shows the comparison between tongue depressor amperometric sensors and a spectrophotometer. The actual concentrations were 0.2 mM, 0.5 mM, and 0.8 mM. Figure 6a show the graph of the sensor for detecting the concentration of glucose in saliva. Different concentrations of glucose in 200 µL of saliva were dropped onto the electrode surface. After the addition of glucose and waiting for 5 min to decompose glucose into H 2 O 2 , the concentration of H 2 O 2 became uniform on the electrode in saliva, and the current baseline was detected for 100 s at 0.6 V. As shown in the figure, the signal is well correlated with the concentration of glucose. The glucose's concentration in saliva was derived from the calibration curve in Figure 4. Figure 6b shows the comparision of the detection performances by using the sensor (pink) and a spectrophotometer (blue). The blue is the result from a spectrophotometer. As shown in the figure, both methods show comparable results, which are about the same as the standard concentrations of glucose. These results demonstrate that the sensor is an effective detection tool for measuring glucose concentrations in saliva. (b) The comparison between the sensor and a spectrophotometer for the detection of different concentrations of glucose. Blue is from a spectrophotometer, red is the true concentration in saliva, and pink is from a sensor. Figure 7 presents the current signal after the the addition of saliva containing glucose on the sensing electrode. According to previous research, the average glucose concentration of healthy people is 78.7 ± 9.2 μM for males and 80.4 ± 7.9 μM for females; while for diabetic patients, it is 201.9 ± 34.9 μM for males and 175 ± 22.3 μM for females [32]. We can see that the difference of glucose concentration between the healthy and diabetic groups is around 100 μM. We believe that this test to see current change against a concentration of about 50 μM of glucose in saliva should be able to is 78.7 ± 9.2 µM for males and 80.4 ± 7.9 µM for females; while for diabetic patients, it is 201.9 ± 34.9 µM for males and 175 ± 22.3 µM for females [32]. We can see that the difference of glucose concentration between the healthy and diabetic groups is around 100 µM. We believe that this test to see current change against a concentration of about 50 µM of glucose in saliva should be able to prove our sensor's capability in distinguishing between healthy and diabetic individuals. Saliva (195 µL) was placed onto the sensor to serve as the buffer solution; after the baseline had stabilized, 5 µL of H 2 O 2 or 10 mM glucose was added to simulate the increase in glucose concentration for a diabetic patient, compared with that of a healthy individual. As shown in Figure 7a, we can see a clear rise of the current signal to glucose, and the signal of H 2 O 2 is slightly larger than that of glucose, which may due to the catalytic conversion rate of glucose by the enzyme and the diffusion issue by the enzyme matrix. Figure 7b shows the the detection of different concentrations of glucose, and it can be seen that a higher concentration of glucose shows a much clear larger signal. The results proves that our sensor can distinguish different concentrations of glucose in saliva, and can distinguish between healthy people and diabetic patients easily. (b) The comparison between the sensor and a spectrophotometer for the detection of different concentrations of glucose. Blue is from a spectrophotometer, red is the true concentration in saliva, and pink is from a sensor. Figure 7 presents the current signal after the the addition of saliva containing glucose on the sensing electrode. According to previous research, the average glucose concentration of healthy people is 78.7 ± 9.2 μM for males and 80.4 ± 7.9 μM for females; while for diabetic patients, it is 201.9 ± 34.9 μM for males and 175 ± 22.3 μM for females [32]. We can see that the difference of glucose concentration between the healthy and diabetic groups is around 100 μM. We believe that this test to see current change against a concentration of about 50 μM of glucose in saliva should be able to prove our sensor's capability in distinguishing between healthy and diabetic individuals. Saliva (195 μL) was placed onto the sensor to serve as the buffer solution; after the baseline had stabilized, 5 μL of H2O2 or 10 mM glucose was added to simulate the increase in glucose concentration for a diabetic patient, compared with that of a healthy individual. As shown in Figure 7a, we can see a clear rise of the current signal to glucose, and the signal of H2O2 is slightly larger than that of glucose, which may due to the catalytic conversion rate of glucose by the enzyme and the diffusion issue by the enzyme matrix. Figure 7b shows the the detection of different concentrations of glucose, and it can be seen that a higher concentration of glucose shows a much clear larger signal. The results proves that our sensor can distinguish different concentrations of glucose in saliva, and can distinguish between healthy people and diabetic patients easily. Conclusions In this work, we have shown the development of a tongue depressor-based biosensor for the first time, and its sensitive, rapid, and reliable determination of glucose in both phosphate buffer and real human saliva. The sensor construction method is simple, with CNTs dissolved in toluene, CNT and Ag/AgCl inks printed onto the surface of a tongue depressor, and GOD immobilized on the sensor. The detection process was easy-to-operate, cost-effective, and quick. The sensor showed a bro ad detection range and a rapid measuring time for glucose. The sensor performance was reliable. The low cost and portability of the substrate lend itself to be used in disposable devices, and make the proposed sensor extremely suitable for at-home tests or on-site medical examination. The noninvasive and disposable nature of the biosensor template improve the user experience, eliminate the chance of cross-infection, and also mean that it would be suited for use in examinations where there are concerns about the risk of infectious disease. Its practicability has also been proved in real human saliva test. This work may open up new avenues for developing various other tongue depressor-based biosensors dedicated to healthcare applications.	2019-09-11T13:06:32.908Z	2019-09-01T00:00:00.000	{ "year": 2019, "sha1": "8befc348e641a399c7e21ea0d455ae3eeb72cb7d", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1424-8220/19/18/3864/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "a3b0bef6656c5e0ac570c6a0e173cbe36e7b2c28", "s2fieldsofstudy": [ "Medicine", "Engineering", "Materials Science" ], "extfieldsofstudy": [ "Chemistry", "Medicine", "Computer Science" ] }
219427105	pes2o/s2orc	v3-fos-license	The Impact Of Financial Liberalization On Economic Risk In The Asia Pacific Countries In this study, we analyze the non-linier effect of financial liberalization policy toward the economic risk of the Asia Pacific countries with trade openness as a threshold variable. To do that, we apply panel regression threshold proposed by Hansen (1999) as our method of analysis. Based on yearly data from 1975 – 2015, we find that there is a non-linier effect of financial liberalization on economic risk depending on the certain level of trade openness. Regarding to this finding, we find that when the trade openness is below the threshold value, financial liberalization policy can reduce the economic risk of Asia Pacific countries. However, when the trade openness exceeds the threshold value, financial liberalization will increase the economic risk. So that, we conclude that an open domestic financial market that is followed by high degree of trade openness will tend to create an economic instability. INTRODUCTION Globalization, which is marked by the strengthening of economic relations between countries, has increased the economic risk of countries in the world. In practice, the strengthening of economic relations between countries makes economic shocks that happen in one country can be quickly transmitted to other countries whether through financial or trade channels. As a result, there is no country in the world that can be fully protected from the cycle occured in the world economy (Aizenman et al. 2013). Financial liberalization and the discussion of its effects on global economic risk has become one of the important issues studied in the economic literature. In relation to economic risk, Ozcan et al. (2003) found that financial liberalization policies can reduce a country's economic risk. This is because financial openness can help a country in obtaining additional capital which allows the country to improve its production quality. So that, financial liberalization can maintain economic stability. However, it is still unclear whether financial liberalization really can provide positive benefit for the economy (Kose et al. 2003). This is because several other studies have shown different results compare to Ozcan (2003) and Mirdala et al. (2015). Easterly et al. (2001) which examines factors that determine economic risk, have found that a liberal financial system can actually increase the volatility risk of output growth. In addition to this, Kose et al. (2003) also found that open financial sector has led to an increase in the risk of consumption and income in a country. Neaime (2005), who analyzed the Middle East and North Africa (MENA) countries, found that financial openness has increased the economic risk of countries located in this region. However, it is still unclear whether financial liberalization really can provide positive benefits for the economy (Kose et al. 2003). This is because several other studies have shown different results compare to Ozcan (2003) and Mirdala et al. (2015). Easterly et al. (2001), for example, which examines factors determining economic risk, have found that a liberal financial system can actually increase the volatility risk of output growth. In addition to this, Kose et al. (2003) also found that open financial sector has led to an increase in the consumption risk and income risk in a country. Neaime (2005), who analyzed the Middle East and North Africa (MENA) countries, found that financial openness has increased the economic risk of countries located in this region. The difference in results that is found in the aforementioned studies indicate that there is still no clear conclusion regarding the impact of financial liberalization on economic risk in a country. One of the reasons why this difference could happen is that the effect of financial liberalization toward the economic risk might not be linear as always assumed by aforementioned studies. In response to this gap, the purpose of this study is to analyze the effect of financial liberalization toward economic risks in a non-linear manner. This non-linear aspect is still not widely discussed in previous studies. To achieve the research objective, this study adapted the panel regression threshold approach developed by Hansen (1999). One of the main advantages of this approach is that the threshold value is not determined exogenously, but endogenously determined by the data. In its operations, this study suspects that the non-linear effect Source: External Wealth of Nations, dan Indeks Chinn-Ito (diolah) Picture 1. The Condition of Financial Liberalization in the Asia Pacific Region of financial liberalization on economic risks is due to differences in the characteristics of trade in each country. Financial liberalization followed by trade openness can affect the economic conditions of a country/region (Mirdala et al. 2015). Therefore, in relation to the financial liberalization and economic risk, the nonlinear effect of financial liberalization This study uses countries in the Asia Pacific region as its sample. These Asia Pacific countries were chosen because from 1970, the openness of the financial sector in this region was very accelerated compared to other regions (Chinn et al. 2007). The increasing movement of international capital flows occurring in the Asia Pacific, according to Borensztein (2011), is characterized by the presence of free and open capital mobility. This increasing capital mobility, as a consequence of the implemen-tation of financial liberalization policy, may effect the economic risk of the countries. toward economic risk is observed through trade openness that are treated as threshold variables. This study also includes several variables that are considered important in influencing a country's economic risk as a control variable, which includes inflation volatility, fiscal policy, institutional quality and volatility of terms of trade. Positive impact: if the countries specialized in a LITERATURE REVIEW Studies discussing macroeconomic uncertainty began to develop around 1990. At that time, Ramey and Ramey (1995) had discovered a detrimental effect of the output volatility on economic growth which ended in a decline of economic prosperity. Even in the short term, the economic volatility has the opposite effect and is detrimental to the poor people. Macroeconomic volatility affects expectations of economic actors related to production, consumption, price uncertainty, and unemployment. The negative relationship between volatility and growth that ultimately affects the level of well-being becomes one of the important problems that raises further research on what factors and sources that cause economic volatility to occur. Internally, in the modern view of macroeconomics, according to Easterly et al. (2000), uncertainty in the economy is determined by rational actions between companies and households, policy intervention by the government and even the complexity of collective behavior that brings the economy to quickly return to full employment. In relation to the context of globalization, research on the economic crisis, which is an extreme manifestation of high economic volatility, has begun to be a concern since the 2000s. In this period, Kose et al. (2005) reveal that many crises are caused by external factors. This is because many developing countries open their economies faster than they should for trade and global finance. Theoretically, the effect of increasing trade openness and financial flows on output volatility, as a proxy for economic risk, depends on various factors, including the composition of trade and financial openness, patterns of specialization and other sources of shocks. Kose et al. (2003) revealed that in the relationship between trade openness and risk, increasing specialization of the production structure affects output volatility in the business cycle scheme. This is because the more specialized a country in certain industries, the effect of international shocks will tend to provide negative transmission and increase the economic risk. Related to the financial openness, Kalemli-Ozcan, Sorensen and Yosha (2003) revealed that financial openness has strengths and weaknesses. The advantage of financial openness is that it can help reduce economic risk in developing countries by providing access to capital, so that it can help them diversify the production base. On the contrary, the weakness of financial openness is because it allows the creation of specialization of production so as to make the economy more vulnerable to a specific industrial shock. As important findings from Kose et al. (2003) who found that trade openness increases output volatility in developing economies because trade flows can increase the possibility of risk sharing. Important findings by Kose et al. (2003) regarding how external shocks exposed to the domestic economy are relevant to study. To date, the results of various studies on the effects of financial openness and trade are still debatable. This is because previous studies find both positive relationship and a negative relationship between financial openness and trade toward economic risks. From the point of view of financial openness, the debate arises because there are no clear conclusions and there are still many debates about the relationship between financial liberalization and macroeconomic volatility. The lack of clarity about the relationship is expected to occur because of the two major forces in financial liberalization. These two forces, on the one hand, may reduce economic risks, but on the other hand, they can also increase the economic risk of a country. In this case, international financial openness can reduce the economic risk because of diversification in sharing risks. But at the other side, financial openness can lead to greater specialization and thus increase the risk of the domestic economy. Neaime (2005) found that financial and trade openness in poor regions such as the JBPE Journal of Business and Political Economy, Vol 2 (1), 2020 Middle East and North Africa (MENA) have a positive relationship to economic volatility. That is, the more open and integrated financial and trade systems in MENA to the world economy, the more it will have a detrimental effect on economic risk in terms of output and consumption. Later, Ahmed and Suardi (2009) have examined the effect of trade and financial liberalization on macroeconomic volatility in the Sub-Saharan African region. The results show that there are differences in the effect between financial openness and trade on macroeconomic volatility. Using the Panel-GMM system method, the results show that financial openness has a positive and significant influence on macroeconomic volatility in the Sub-Saharan African region. It means that the more financial openness increases, the more it will increase the macroeconomic uncertainty. On the contrary, trade openness policy has a negative impact on macroeconomic risks. This means that the more liberal and integrated the trade systems in Sub-Saharan Africa, the more it will reduce the macroeconomic volatility. Similarly, Mujahida and Alam (2013) have examined the effects of trade and financial openness toward economic risks in Pakistan. Using the Autoregressive Distributed Lag (ARDL) method, the results show that in the long run trade openness has a negative and significant effect on the economic risks. Meanwhile, financial openness has a negative impact on investment risks, which reflects that an increase in financial openness causes a decrease in investment risk in the short term. Related to the above, Yang (2011) analyzed the impact of political democracy and economic liberalization toward economic risks using the difference in difference method for 158 countries from 1970 -2005. The results have showed that economic liberalization %which are financial and trade openness% can reduce economic risks, but the same thing is not found in political democracy. However, the implementation of democracy after economic openness is a good choice because it gives a positive effect in reducing the economic risk. This result serves to provide additional support for policy recommendations that developing countries must liberalize their economies first and then consider to liberalize their political system by implementing democracy. Another interesting finding, that gave rise to a new theoretical perspective is the study done by Barrot et al. (2018). Barot et al. (2018) has identified four structural shocks -demand, supply, monetary and commodity shocks and linked their impact to state policies and structural frameworks. Shocks originating from external factors, such as trade, in the past few decades have had a greater impact on increasing the economic risk of a country compare to the shocks originating from domestic factors. Global monetary policy shocks are one of the main external sources of economic risk in developing countries. An increase in openness will increase external disruption to the economy. Another interesting fact is that in the case of trade openness, the contri-bution of variance to global disturbances is non-linear, following a Ushaped pattern. In this case, external disturbances will have a greater effect on low and high openness than intermediate levels. In addition to this, an important finding is that commodity intensive countries show are riskier for all types of external shocks. 1Panel Threshold Model The main purpose of this study is to identify whether there is a non-linear effect from financial liberalization policy toward economic risk in the Asia Pacific countries. Thus, to accommodate this purpose, this reseach adopt an approach called panel threshold model introduced by Hansen (1999). The main advantage of this approach is that the threshold value is not arbitrarily determined, so that it allows us to get the confidence interval for the chosen interval. Beside that, the other advantage of the endogenous threshold regression technique include (1) it does not require any specified functional form of non-linearity, and the number and location of thresholds are endogenously determined by the data, and (2) asymptotic theory applies, which can be used to construct appropriate confidence intervals. A bootsrap method to asses the statistical significance of the threshold effect is also available in order to test the null hyphothesis of a linear formulation against a threshold alternative (Chang et al. 2010). Hansen (1999) developed an econometric model that is suitable for threshold regression with panel data. The panel threshold model divides the observation into two or more regimes, depending on whether each observations is above or below the threshold level. The subscript i indexes the individual and t indexes time. The depen-dent variable, y t , and the threshold variable, q it , is scalar, and the regressor x it is a k vector. The structural equation of interest is Where I(.) is an indicator function. The observed data will be divided into two regimes, depending on whether the threshold variable q it is smaller or larger than the threshold ã. The regimes are distinguished by different regression slope ² 1 and ² 2 . In order to be able to identify ² 1 and ² 2 , regressor variable x it and threshold variable q it are not time-invariant; u it is the fixed individual effect, and error e it is assumed to be independently and identically distributed (iid) with zero mean and finite variance. Threshold level ã is estimated using least square method introduced by Hansen (2000). A bootstrap procedure is adapted to get approximate critical values of the test statistics which allows us to perform the hyphothesis test for the threshold effect. In this case, if the asymptotic p value is smaller than the desire critical value, we conclude that the null hyphothesis of no threshold is rejected. After a threshold value is found, the confidence intervals for the threshold value and slope coefficients are then estimated (Chang et al. 2010). The same procedure can also be applicated in the case of multiple thresholds. Related to this case, the potentiality of presence of more than one threshold represent another advantage of this method over the traditio-nal approach. Our purpose of this study is to analyze the effect of financial liberalization on economic risks with trade openness as the threshold variable. This research includes some variables that are important in effecting economic risk in a country as control variables, including (1) volatility inflation, (2) discretionary fiscal policy, (3) institu-tional quality, and (4) finlib is a financial liberalization policy which is proxied by two measures, namely (1) financial integration, and (2) financial deepening. These approaches have been adapted in several previous studies, including Kose et al. (2003), Kose et al. (2006), Bekaert, Harvey, & Lundblad (2006), Ahmed & Suardi (2009), and Nicolò & Juvenal (2012). Financial integration, which is used as a proxy for external financial liberalization in this study, was built using Lane & Milesi-Ferretti (2007) calculation methods. The measure of financial integration is the sum of international financial gross assets and international financial liabilities relative to GDP. Beside financial integration, this study also adapt financial deepening as a proxy for internal financial liberalization as measured by the value of M2 relative to GDP. Trade openness, which is treated as threshold variable, is a measure of the sum total exports and imports relative to total GDP. This measure has been widely used by various literatures as a proxy of economic openness such as Kose et al. (2003), Neaime (2005), Dupasquier & Osakwe (2006), Ahmed & Suardi (2009), Pancaro (2010) and Balavac & Pugh (2016). To address the country specific effects, we also consider some control variables to be included in empirical model which are terms of trade volatility, inflation rate volatility, institutional volatility and descretio-nary fiscal policy. Specifically, the discretionary fiscal policy is constructed by using the method proposed by Fatas & Mihov (2003), Where G is the logarithm of real government spending and Y is the logarithm of real GDP. Deterministic time trends are used to capture the observed trends in govern-ment spending at all times. Discretionary fiscal policy's data is represented by the value of å t . To calculate this data, we use annual data for 20 Asia-Pacific countries from the period 1975-2015 and estimate the following regression for each country. All data are collected at an annual frequency. The data used is collected through various sources, and it covers 20 countries in the Asia Pacific for period 1975 -2015. Operationally, as we mentioned before, GDP Growth Volatility is used as a proxy for economic risk; meanwhile, the financial integration and financial deepening are used as a proxy for financial liberalization. These proxies are used because it represents the impact of financial liberalization on financial sector both externally (financial integration %capital inflows and outflows) Table 1. Data and Data Source and internally (financial deep-ening). Trade openness, which is defined as the sum of import and export per GDP, is used as a proxy for trade openness condition in each country and is also treated as the threshold variable. Other variables, including terms of trade volatility, inflation rate volatility, discretionary fiscal policy, dan institutional quality are used as control variables. Tabel 2. Descriptive Statistics Before discussing the main result of this study about the effect of financial liberaliza-tion on economic risk, the descriptive statistics of each variable is represented in the study for financial liberalization, the score range between 13.43 and 344.21. For the trade openness, the score lies between 0.13 and 4.39. The somewhat high score range for volatility infla-tion indicate that there is a high variation of inflation in some of the sample countries. RESULTS As we mentioned earlier, this study suspects that there is a non-linear impact of financial liberalization policy on economic risk in the Asia Pacific, depending on the degree of trade openness in each country. In order to be able to detect this condi-tion, in terms of whether an economy with a high level of trade openness can show a different trend compare to the economy with a relatively low level of trade openness, this study accommodate the posibility of nonlinear effect (or threshold effect) of trade openness related to the impact of financial liberalization on economic risks. As we discussed earlier, it is still unclear based on empirical results whether financial liberalization can give a positive impact by reducing the economic risk in a country (Kose et al. 2003). Related to the above, to test our hyphothesis regarding the non-linear effect between financial liberalization and economic risk with trade openness as a threshold variable, we test for the existence of a threshold effect. This paper uses the bootstrap method to calculate the F statistics and the p value. The results are estimated with two different financial liberalization measures, which are financial integration (model 1) and financial deepening (model 2). To be noticed, the test statistic for a single thres-hold is significant for both financial integration and financial deepening, but the test for double and triple thresholds are insignificant. Based on this findings, we may conclude that there is a strong evidence that our hyphotesis stating that there is a non-linear effect of financial liberalization toward economic risks can be accepted. The threshold regression using Hansen (2000) method are shown in Table 3 below. Table 3 reports the results for models 1 and 2. Based on the threshold results, the bootstrap p value shows significant results for both models. These results confirm the existence of a threshold effect between Source: Data processed (2018) financial liberalization and economic risk. So that, rather linear, we can conclude that the effect is non-linear. The estimated threshold of trade openness is 143.8 percent for both financial integration (model 1) and given is calculated, the LR2 value at the estimated threshold value will always be equal to zero. The dashed line represents a critical value of 5%. Table 3. Threshold Effect Estimation After confirming the presence of nonlinear effects, we now try to divide the sample countries according to the level of In this study, financial liberalization is proxied by two variables which include (1) financial integration, and (2) financial deepening. Financial integration, which is measured by the sum of capital inflow and outflow divided by GDP, is used as a proxy its degree of trade openness threshold value which receive a score of 143.8 percent. Table 4. shows that the majority of countries, that are almost 80 percent from 1980 -2015, have a relatively lower trade openness value compare to the threshold level. In contrast, it is only 20 percent of the sample countries have a higher trade openness value compare to its threshold level. Based on this findings, we may say that only a small portion of our sample countries that are vulnerable to external shocks because of their trade openness condition higher than the threshold level. Table 4. Percentage of Countries in Each Regime by Year for financial liberalization in terms of external. Financial integration has been used as a proxy for financial liberalization in many researches including ), Neaime (2005), Bakaert et al. (2006, Ahmed et al. (2009) andMirdala et al. (2015). 1980 1985 1990 1995 2000 2005 2010 2015 Trade Moreover, the financial deepening is used as a proxy for financial liberalization in terms of internal according to Kose (2003), Neaime (2005) and Ito (2006). Table 4 reporting the regression results regarding the effect of financial integration toward economic risk of Asia Pacific Countries. The results show that financial liberalization as proxied by financial integration increases the economic risk when the countries liberalize their trade beyond the predetermined threshold (which receive a value of 1.438). This shows that the higher level of economic liberalization as indicated by the high financial liberalization and trade liberalization can simultaneously increase the risk of the economy by increasing the volatility of GDP growth. The similar result were also obtained when analyzing the effect of financial liberalization on economic risk using financial deepening as a proxy. Regarding to this concern, we found that financial deepening increase the GDP growth volatility when the countries liberalize their trade over the trade threshold. It means that an open domestic financiall market that is followed by high degree of trade opennes will tend to create an economic instability. It is because that the potential loss of foreign exchange reserves that are needed to finance the country's international trade are greater in the financial liberalization regime. According to Cardarelli, Elekdag and Kose (2009), capital inflows create important challenges for policymakers because excessive capital inflows, that can be triggered by financial liberalization policy, may give the potential excessive pressure, lose competitiveness due to the appreciation of the exchange rate and increase the vulnerability to the economic crisis. Related to this concern, Stiglitz (2002) revealed the negative effect of financial liberalization is that it is able to create an instability in the financial market if the economy is still not well developed. In the case of this research finding, the financial liberalization policy will tend to increase the country's economic risk that has a high degree of trade openness. Table 4 and 5 also explain other factors that may affect the economic risk of Asia Pacific countries. It is found in the estimation result that terms of trade volati-lity show a positive but not significant effect toward economic risk. The estimation results indicate that the inflation rate volatility significantly effect the economic risk in a positive manner. It means that inflation rate volatility increase the economic risk. This finding is in line with Ahmed andSuardi (2009) andNeaime (2005). Adverse effect of inflation volatility toward economic risk is because increasing uncertainty in inflation, which is represented by increasing inflation rate volatility, can distort price mechanism effectivity in allocating resources, thus it may increase the economic risk. Related to the above, it also found that the discretionary fiscal policy has a positive effect to the economic risk in both model. This is due to undisciplined fiscal policy can lead to output fluctuation giving pressure to the economic risk of a country through the decreasing governments credibility. In terms of the institutional quality, the result show that this variable can reduce economic risk of a country. It means that the existence of a good institutional quality can help a country to lower its economic risk. This risk reducing role played by institutional quality that is found in this study in line with Ahmad and Suadi (2009) which showed that good financial market insti-tution can help to reduce capital flight so that it will maintain the economic stability. CONCLUSION The purpose of this study is to analyze the effect of financial liberalization toward economic risks in a non-linear manner. In analyzing the non-linier effect of financial liberalization, we use trade openness as threshold variable. By using panel regression threshold introduced by Hansen (1999), we confirm that there is a non-linier effect of financial liberalization toward economic risk in Asia Pacific countries depending on the value of trade openness in each country. Regarding to this finding, when the trade openness is below the threshold value, financial liberalization policy can reduce the economic risk of Asia Pacific countries. However, when the trade openness exceeds the threshold value, the financial liberalization will increase the economic risk. So that, we conclude that an open domestic financial market that is followed by high degree of trade openness will tend to create an economic risk in the Asia Pacific countries.	2020-06-08T02:23:52.315Z	2020-05-13T00:00:00.000	{ "year": 2020, "sha1": "843fbf8d7eed6e4e436398196a7aee13e633b608", "oa_license": "CCBY", "oa_url": "https://journal.indef.or.id/index.php/BisnisEkonomiPolitik/article/download/26/10", "oa_status": "HYBRID", "pdf_src": "Anansi", "pdf_hash": "ce5c0835b243b7c7658d367826b8739d7d367223", "s2fieldsofstudy": [ "Economics" ], "extfieldsofstudy": [ "Economics" ] }
252533922	pes2o/s2orc	v3-fos-license	Systemic immune inflammatory index is an independent predictor for the requirement of decompressive craniectomy in large artery occlusion acute ischemic stroke patients after mechanical thrombectomy Background and purpose Following mechanical thrombectomy (MT), patients with large artery occlusive acute ischemic stroke (LAO-AIS) often have cerebral herniation due to its complications, resulting in poor prognosis. Decompressive craniectomy (DC) can markedly improve patient prognosis. This study aimed to verify the predictive value of clinical parameters such as the systemic immune-inflammatory index (SII) for DC in patients with LAO-AIS after MT. Methods Clinical data of a total of 173 patients with LAO-AIS treated with MT between January 2020 and January 2022 were retrospectively analyzed. Patients receiving DC were grouped into an experimental group or a control group (no DC). The patients were randomly divided into the training set (n = 126, 75%) and validation set (n = 43, 25%). Multivariate logistic regression was used to construct a nomogram predictive model. Results The SII value in the experimental group (median: 2851.1×109/L) was significantly higher than that in the control group (median: 1898.6 × 109/L) (P = 0.019). Receiver operating characteristic (ROC) analyses showed that the best cutoff value of the SII was 2505.7 × 109/L with a sensitivity of 55%, a specificity of 75.8%, and an area under the curve (AUC) of 0.649. Multivariate logistic regression indicated that the SII was an independent predictor for performing DC in patients with LAO-AIS after MT (OR = 3.579, 95% CI = 1.360–9.422, P = 0.01). The AUC was 0.728 in the training set and 0.583 in the validation set. The average error of the calibration curve was 0.032 in the training set and 0.023 in the validation set. The average error was relatively small and consistent in the training set and validation set. The C-index of the nomogram was 0.804 suggesting good accuracy. Conclusions The SII at admission is an independent predictor for the requirement of DC in patients with LAO-AIS after MT. The SII-based nomogram helps doctors make decisions on whether DC is needed timely and rationally, and thereby may improve the prognosis of these patients. Background and purpose: Following mechanical thrombectomy (MT), patients with large artery occlusive acute ischemic stroke (LAO-AIS) often have cerebral herniation due to its complications, resulting in poor prognosis. Decompressive craniectomy (DC) can markedly improve patient prognosis. This study aimed to verify the predictive value of clinical parameters such as the systemic immune-inflammatory index (SII) for DC in patients with LAO-AIS after MT. Methods: Clinical data of a total of patients with LAO-AIS treated with MT between January and January were retrospectively analyzed. Patients receiving DC were grouped into an experimental group or a control group (no DC). The patients were randomly divided into the training set (n = , %) and validation set (n = , %). Multivariate logistic regression was used to construct a nomogram predictive model. Results: The SII value in the experimental group (median: . × /L) was significantly higher than that in the control group (median: . × /L) (P = . ). Receiver operating characteristic (ROC) analyses showed that the best cuto value of the SII was . × /L with a sensitivity of %, a specificity of . %, and an area under the curve (AUC) of . . Multivariate logistic regression indicated that the SII was an independent predictor for performing DC in patients with LAO-AIS after MT (OR = . -. , P = . ). The AUC was . in the training set and . in the validation set. The average error of the calibration curve was . in the training set and . in the validation set. The average error was relatively small and consistent in the training set and validation set. The C-index of the nomogram was . suggesting good accuracy. Introduction Globally, and especially in low-and middle-income countries, stroke is the leading cause of death and disability, and most patients who survive the acute phase of stroke show varying degrees of neurological deficits (1,2). Large arterial occlusion (LAO) has become one of the most important causes of acute ischemic stroke (AIS) worldwide and is closely related to the poor prognosis of AIS patients (3). In the past 10 years, endovascular treatment of large-arterial occlusion acute ischemic stroke (LAO-AIS) has achieved good progress (4). For such patients, mechanical thrombectomy (MT) is one of the main treatment approaches. MT shows clear efficacy in recanalization if used within 6 h after the onset of LAO-AIS, and its complete recanalization rate is ∼50% (5-7). Brain reperfusion injury is the most serious complication after MT and mainly manifests as contrast extravasation (CE) and hemorrhagic transformation (HT) (8). CE and HT are closely associated with poor prognosis, with a mortality rate close to 50%. The increase and imbalance of intracranial pressure caused by CE or HT can induce brain herniation, which leads to severe neurological deterioration and even death (9). Decompressive craniectomy (DC), as a salvage operation, can treat patients with refractory intracranial hypertension and early brain herniation by relieving intracranial pressure (10). The current opinion is that once brain herniation is detected by imaging, DC can be performed as soon as possible without waiting for neurological deterioration (11). A Brazilian study showed that early DC could significantly improve the prognosis of AIS patients (12). Bruno Askiel et al. (13) proposed that patient age and midline shift after DC could predict neurological outcomes in post-DC patients. Oliveira BDD et al. (14) stated that ultrasound monitoring of the optic nerve sheath diameter can identify severe intracranial hypertension in patients with large vessel occlusion, and this parameter can be used as a predictive factor for DC. However, it is still necessary to find more convenient and practical predictors for the requirement of DC to allow doctors to make decisions more quickly and improve patient outcomes. In recent years, an increasing number of studies have focused on the inflammatory response after stroke. Inflammation is thought to be involved in recovery initiation and repair processes after stroke, yet some aspects of the inflammatory response may instead have detrimental effects in stroke patients (15). The systemic immune-inflammation index (SII), a new type of inflammatory index, is calculated by combining three immune inflammatory cells-lymphocytes, neutrophils, and platelets-and fully reflects the inflammation status (16). A high SII level increases the risk of stroke in patients with hypertension and affects patients' prognosis (17). Therefore, the purpose of our study was to calculate the SII according to the first routine blood test at admission, construct a model for the prediction of DC requirements, and verify the predictive value of the SII to provide a tool for the selection of DC in patients with LAO-AIS after MT. Patient selection We retrospectively analyzed the data of patients who were diagnosed with LAO-AIS and underwent MT in Huizhou Central People's Hospital from January 2020 to January 2022. We included patients who met the following criteria: (1) those who were admitted within 24 h after disease onset, with symptoms and signs of neurological dysfunction and confirmed with LAO-AIS by cranial CT angiography (CTA) or whole cerebral angiography; (2) those that met the MT criteria and received MT in our hospital. The indications of mechanical thrombectomy are as follows: AIS with National Institutes of Health Stroke Scale (NIHSS) ≥ 6; Cerebral infarction caused by large artery occlusion of anterior circulation; Alberta Stroke Program Early CT Score (ASPECTS) ≥ 6; CT or MRI ruling out ICH; and groin puncture for MT within 6 h of symptom onset (18). It is worth mentioning that the surgical indications for mechanical thrombectomy in wake-up stroke patients are defined as follows:(1) Patients with anterior circulation large vessel occlusion within 6-16 h from the last normal state of the patient and meeting the enrollment criteria of DAWN or DEFUSE-3 study. (2) Patients with anterior circulation large vessel occlusion within 6-24 h from the last normal . /fneur. . state of the patients and meeting the enrollment criteria of the DAWN study (19); and those that were (3) >18 years of age. The following patients were excluded: (1) those that did not complete routine blood tests at admission and lacked required laboratory data; (2) those with a history of cerebral infarction, and MRS ≥2 points; (3) those with a history of infection within 2 weeks; (4) those complicated by hematological diseases; (5) those undergoing administration of immunosuppressive drugs; (6) those with a history of malignant tumor or autoimmune disease; or (7) patients with posterior circulation occlusion. This study was approved by the Ethics Committee of Huizhou Central People's Hospital (kyll2022029) and was conducted in accordance with the Declaration of Helsinki. As this was a retrospective study, the Ethics Committee approved the application for a waiver of signed informed consent. To protect patient privacy, this study anonymized the identifiable information of the enrolled patients. Data collection We collected baseline clinical data of all patients, including age, gender, and medical history of coronary heart disease, hypertension, diabetes, smoking and drinking, atrial fibrillation, and hyperlipidemia. Hypertension was defined as a prior history of hypertension with systolic/diastolic blood pressure ≥140/90 mmHg at admission. Diabetes was defined as a history of diabetes, fasting blood glucose (FBG) ≥7 mmol/L, and random blood glucose ≥11.1 mmol/L at admission. Coronary heart disease was defined as a history of coronary heart disease. Atrial fibrillation was defined as a history of atrial fibrillation or an ECG diagnosis at admission. We collected the routine blood, blood biochemistry, hemostatic function, and other serological test results of the patients obtained in the emergency department or within 1 day of admission. White blood cell count in the range of (4-10) × 10 9 /L was defined as normal, otherwise abnormal. Neutrophil count in the range of 2-7 × 10 9 /L was defined as normal, otherwise abnormal. The absolute count of lymphocytes in the range of 0.8-3.5 × 10 9 /L was defined as normal, otherwise abnormal. The normal range of platelets was (100-300) × 10 9 /L. Abnormal test results also included: PT > 14 s, APTT > 37s, D-D dimer > 500 ng/ml, FIB > 4 g/L, LDL > 3.1 mmol/L, TG > 1.7 mmol/L, serum potassium < 3.5 mmol/L, serum calcium < 2.11 mmol/L and albumin < 40 g/L. The patients' inflammation was evaluated by collecting and calculating the platelets, neutrophils, and lymphocytes in the blood during routine examination of the patients when they were in the emergency department or just admitted to the hospital. The calculation formula is SII = platelets × neutrophils/lymphocytes. Outcome event In this study, whether patients underwent DC was regarded as the outcome event. DC was defined as a positive outcome, and no DC was defined as a negative outcome. Indications for DC included: (1) Patients with massive cerebral infarction had early symptoms of a cerebral hernia within 48 h of onset (Clinical manifestations such as disturbance of consciousness, Cushing reaction, unilateral mydriasis, Centerline offset ≥5 mm, Ipsilateral lateral ventricle compression, Sulci cistern compression); (2) After active medical treatment, there are still obvious manifestations of intracranial hypertension or progressive deterioration of neurological function (GCS score decreased ≥1 or new mydriasis or light reflex changes or new focal motor function defects); (3) For patients with large-area cerebral infarction in the middle cerebral artery region, EDEMA score ≥3 points or modified EDEMA score ≥6 points (19, 20). Statistical analysis Patients undergoing DC were classified as the experimental group, otherwise, the control group. The data were tested for normality using SPSS 23 (SPSS Inc., Chicago, IL, USA). The continuous data which conformed to the normal distribution were expressed as mean ± standard deviation, and the t-test was used for inter-group comparisons; otherwise, the continuous data were expressed as the median and quartile, and the rank sum test was used for inter-group comparisons. The count data were expressed as percentage n (%), and the chi-square test was used for inter-group comparison. If the expected frequency of more than 20% of cells in the contingency table was <5, Fisher's test was used. Baseline data of the two groups were compared to assess their comparability. The t-test or rank-sum test was used to compare the difference in SII between the experimental and control group. GraphPad Prism 8.3 was used to draw a bar plot of the SII values of the two groups and was also used to plot the receiver operating characteristic (ROC) curves to evaluate the predictive value of the SII for DC. The cutoff value of SII was determined according to the ROC curve, and the cutoff value was used to divide SII into the high and low groups. Univariate logistic regression was first conducted to identify factors associated with adverse outcomes, and odds ratios (ORs) and 95% CIs were calculated. Significant factors in the univariate analysis were included in the multivariate logistic stepwise regression analysis to identify independent predictors. All tests were two sided and a P < 0.05 was considered statistically significant. Significant factors in the multivariate regression and common clinical indices were used to construct a nomogram prediction model and plotted by R studio (version 4.1.0). The ROC curve and area under the curve . /fneur. . (AUC) were used to assess the predictive performance of the nomogram model. The goodness of fit of the nomogram was assessed by the calibration curve and average error. The C-index was calculated to evaluate the accuracy of the nomogram. Baseline characteristics We collected the clinical data of a total of 194 patients with LAO-AIS undergoing MT. Eleven patients did not undergo routine blood tests at admission and lacked the required laboratory data, 10 patients were excluded due to vertebrobasilar artery occlusion. These patients were excluded, and the remaining 173 patients were included in the analysis. Twenty and 153 patients were included in the experimental group and control group, respectively (Figure 1). The experimental group consisted of 13 men and 7 women with an average age of 56.1 ± 5.6 years. In the experimental group, the left side was involved in 11 patients and the middle cerebral artery was the responsible artery in 11 patients. The control group consisted of 105 males and 48 females with an average age of 56.6 ± 9.2 years. In the control group, 76 patients had lesions on the left side and the middle cerebral artery was the responsible artery in 99 patients. The average systolic blood pressure and diastolic blood pressure in the 173 patients were 143.8 ± 26.3 mmHg and 85.9 ± 16.3 mmHg at admission, respectively, and 52% of patients had a history of hypertension. In addition to hypertension, patients who had a history of atrial fibrillation, diabetes, hyperlipidemia, smoking, and alcoholism accounted for 22, 19.1, 26.2, 39.3, and 14.5%, respectively. There were no significant differences in the baseline characteristics between the two groups (Table 1), indicating comparability between these two groups. Factors associated with DC in patients with LAO-AIS after MT The SII value in the experimental group (median: 2851.1 × 10 9 /L) was significantly higher than that in the control group (median: 1898.6 × 10 9 /L) (P = 0.019, Figure 2). ROC analysis indicated that the best cutoff value of SII was 2505.7 × 10 9 /L, with a sensitivity of 55% and specificity of 75.8%. The AUC was.649 suggesting that the SII had a certain value in predicting DC ( Figure 3). In order to further understand the relationship between the SII and DC, we divided the 173 patients into the high SII group and low SII group according to the ROC cutoff value (SII > 2505.7 × 10 9 /L, n = 125; SII ≤ 2505.7 × 10 9 /L, n = 48). Univariate analysis showed that hypokalemia and high SII were associated with DC. Multivariate analysis demonstrated that high SII was an independent risk factor for DC in patients with LAO-AIS after MT (OR = 3.579, 95% CI = 1.36-9.422, P = 0.01) ( Table 2). Predictive nomogram development We then constructed a nomogram model for the prediction of DC. We incorporated the SII and common clinical indices into the nomogram. The patients were randomly divided into . /fneur. . the training set (n = 126) and validation set (n = 43). Age, sex, SII, serum potassium, responsible artery, and PT were included in the binary logistic regression and the nomogram model was developed based on the regression equation (Figure 4). With the exception of age, which was a continuous variable, other indices, including the SII, serum potassium, responsible artery and PT were binary variables. For SII, 0 represents low SII, and 1 represents high SII. For sex, 0 represents male and 1 Frontiers in Neurology frontiersin.org . /fneur. . represents female. For serum potassium, 0 represents normal and 1 represents hypokalemia. As for the responsible artery, 0 represents the internal carotid and vertebrobasilar artery, and 1 represents the middle cerebral artery. For PT, 0 represents normal PT and 1 represents prolonged PT. The nomogram, calibration curve, and ROC curve were plotted, and the C-index was calculated. Each item in the nomogram had a score and the sum of the scores indicated the risk of DC. The AUC of the training set was 0.728 ( Figure 5A), and the AUC of the validation set was 0.583 ( Figure 5B). The model showed good discrimination and predictive ability. The nomogram was calibrated using a calibration curve, and the training set calibration curve ( Figure 6A) suggested that the model had an average error of 0.032 in predicted vs. the actual risk of developing the adverse outcome. Similarly, the validation set calibration curve ( Figure 6B) showed that the average error was 0.023, proving that the predictions were in good agreement with the observations. The C-index was 0.804, confirming the accuracy of the nomogram. Discussion In China, the weighted prevalence of stroke is increasing year by year, from 2.28% in 2013 to 2.58% in 2019. People's understanding of stroke is also gradually changing, and more and more studies pay attention to the predictors of stroke (21). The role of inflammation in patients with LAO-AIS after MT remains unclear. Related inflammatory factors such as chemokines can be released in certain brain tissues in patients with AIS, which can aggravate neurological dysfunction (22). Infiltration of leukocytes and various inflammatory mediators can accelerate neuronal apoptosis, thereby aggravating brain injury (23). Neutrophils rapidly migrate to ischemic brain tissues and exacerbate stroke injury by releasing reactive oxygen species, proteases, and pro-inflammatory cytokines (24). Patients with AIS undergo a complex neurohormonal response that induces lymphopenia, leading to disruption of the blood-brain barrier (25). Activated platelets in patients with AIS can interact with T lymphocytes and produce inflammatory factors to slow down the recovery of cranial nerve function (26). As a new inflammatory marker, the SII has attracted considerable attention in recent years. Previous studies reported that the SII had a good predictive value for the prognosis of myocardial infarction, breast cancer, and small cell lung cancer (27)(28)(29). Luo et al. (30) included 76 patients with aneurysmal subarachnoid hemorrhage in their prognostic analysis, and the results showed that the SII at admission was closely related to the 6-month clinical outcome of patients. The study by Chen et al. (31) also showed that the SII was an independent predictor for delayed cerebral ischemia in patients with aneurysmal subarachnoid hemorrhage. Therefore, the purpose of our study was to use the routine blood test results at admission to calculate the SII and other inflammatory indicators, to evaluate the predictive value of the SII for DC in patients with LAO-AIS, and to construct a nomogram to allow doctors to make decisions timely, predictably and rationally. It is well-known that patients with LAO-AIS can develop cerebral herniation after MT due to increased cerebral edema, cerebral reperfusion injury, or other complications. DC can save these patients' lives; however, the decision to perform DC should be quick and reasonable (32). Our study suggests that the SII is . /fneur. . an independent predictor for the requirement of DC in patients with LAO-AIS after MT. However, the nomogram we developed based on patients' clinical data showed good predictive power. The results of routine blood and blood biochemistry tests can be obtained within a few hours after admission and the SII can be calculated immediately. Doctors are able to assess the necessity of DC according to the score of the nomogram and select the best treatment plan. Our nomogram model showed that older age and lower nomogram scores resulted in a lower probability of undergoing DC. Previous studies have reported that patients with cerebral infarction younger than 60 years old achieved a better outcome after DC (33). We suppose that age is negatively correlated with total brain volume, and elderly people usually have varying degrees of brain atrophy (34); therefore, brain edema after stroke is less likely to cause brain herniation in elderly patients, thereby reducing the probability of DC. Hypokalemia also contributed a certain weight to the nomogram. The study by Wang et al. (35) showed that serum potassium level at admission was associated with stroke recurrence in AIS patients, and some scholars have proposed that emergency stroke patients should be screened immediately for electrolyte disturbances. Early detection and correction of the imbalance of serum potassium improve the prognosis of stroke patients (36). The nomogram we developed showed the predictive power of common clinical factors for DC and can help doctors' decision-making. There are no relevant studies reporting predictors for DC in patients with LAO-AIS after MT, as well as the construction of . /fneur. . a similar predictive model. The SII may be a potential predictor although this needs to be validated by more studies. The limitations of this study include: (1) The study was a single-center study with a relatively small number of cases, and selection bias may have existed. (2) The study was retrospective, and some confounding factors may be inevitable. (3) There may be differences in the indications of DC among various medical centers, which may affect the statistical analysis, and further multi-center and prospective studies are needed to confirm our results. Conclusions The SII at admission is an independent predictor of the requirement for DC in patients with LAO-AIS after MT. This index can be obtained conveniently and quickly. The SII-based nomogram assists clinicians in assessing the necessity of DC and to make decisions faster and better, so as to improve patients' prognoses. Data availability statement The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Ethics statement The studies involving human participants were reviewed and approved by the Ethics Committee of Huizhou Central People's Hospital (kyll2022029). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. Author contributions W-CL, Y-XZ, and Z-ZQ conceived the review. GZ, K-LZ, H-YZ, and J-SC conducted data collection. Y-XZ, Y-FD, and Z-ZQ analyzed and interpreted the study data. All authors read and approved the final manuscript.	2022-09-27T13:47:47.317Z	2022-09-27T00:00:00.000	{ "year": 2022, "sha1": "c8a0dab0994bd3145253ee4bffa5032e31234976", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Frontier", "pdf_hash": "c8a0dab0994bd3145253ee4bffa5032e31234976", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
268998383	pes2o/s2orc	v3-fos-license	Biochemical dynamics during postharvest: Highlighting the interplay of stress during storage and maturation of fresh produce : The lifecycle of fresh produce involves a sequence of biochemical events during their ontology, and these events are particularly signi ﬁ cant for climacteric fruits. A high demand during ripening is observed in these plant products, which is re ﬂ ected in a high rate of respiration and ethylene production. Increased respiratory demand triggers the activation of secondary pathways such as alternate oxidase, which do not experience critical increases in energy consumption in non-climacteric fruit. In addition, biochemical events produced by external factors lead to compensatory responses in fresh produce to counteract the oxidative stress caused by the former. The dynamics of these responses are accompanied by signaling, where reactive oxygen species play a pivotal role in fresh product cell perception. This review aims to describe the protection mechanisms of fresh produce against environmental challenges and how controlled doses of abiotic stressors can be used to improve quality and pro-long their shelf-life through the interaction of stress and defense mechanisms. being tested in the primary production sector.Understanding emerging crops and identifying new nutritious fruits and vegetables better suited to current environmental conditions will be a key strategy.However, the process of characterizing, harvesting, and achieving ideal postharvest conditions is challenging [1], yet all these steps are essential to the preservation of the fresh produce marketing chain. During the postharvest period, fresh produce is exposed to numerous environmental factors that stimulate biochemical processes and accelerate ripening.Proper handling, transportation, and storage of fresh produce are therefore critical to maintaining superior quality, and temperature control is fundamental.Despite temperature still being a critical point, new technologies such as controlled and modified atmospheres (CA and MAP) and chemical treatments (e.g., sodium hydrosulfide, nitric oxide), or the use of growth promotors, and irradiation (e.g., UV-A, UV-B, and UV-C) have demonstrated efficiency applied in storage through the activation of antioxidant systems (enzymatic and non-enzymatic) mitigating oxidative stress caused by reactive oxygen species (ROS) [2,3]. Preserving specific fruits requires understanding their biochemistry and the responses induced by a given stress (e.g., the defense systems, both enzymatic and nonenzymatic).The enzymatic antioxidant system includes superoxide dismutase (SOD), catalase (CAT), ascorbate peroxidase (APX), and glutathione reductase (GR), which can alleviate oxidative stress by scavenging ROS [4].While the non-enzymatic systems encompass the ascorbate-glutathione (AsA-GSH) cycle, phenolics, flavonoids, and anthocyanins biosynthesis. Mitochondria are one of the key cellular structures to be monitored concerning oxidative stress.This cellular structure plays a crucial role in the production of ROS in plant cells.As a result, plants have developed alternative mechanisms to mitigate oxidative stress.One such mechanism is the presence of the alternative oxidase (AOX) in the electron transport chain (ETC) of the mitochondrial membrane, which helps to reduce ROS generation and overflow of electrons [5,6]. The application of postharvest treatments induces the accumulation of self-defense components involving stress responses and the interplay with fresh product perception.While high doses are harmful and can induce spoilage, the optimization of stress doses leads to the improvement of quality and prolongation of shelf life.Although cellular perception in plants and response mechanisms are reported, the understanding of fresh product perception needs to be studied deeper.Therefore, this review aims to provide new perspectives to readers unveiling the biochemical events during ripening and storage. Fresh products Plants have been a primary source of sustenance for humans since ancient times.Over the years, human development has advanced the processing, preservation, and consumption of these products.Currently, certified organizations, such as the World Health Organization (WHO), widely recommend consuming fresh products (daily intake of 400-600 g of fruit) [7]. Fruits and vegetables are rich in bioactive compounds, including phenols, carotenes, flavonoids, and isothiocyanates among others.These compounds contribute to their nutritional value and provide potential health benefits.It is important to be aware that these claims need further research before definitive conclusions can be drawn [8].The beneficial effects of these constituents have been subject to comprehensive scientific testing, affirming their antioxidant, anticarcinogenic, and antidiabetic qualities [9,10].Several researchers are currently investigating the importance of regularly consuming fresh produce within the context of a diet specific to certain diseases.For instance, men who consumed vegetables, fruits, and berries more than 27 times a month showed a 10% lower risk of all-cause mortality and a 20% lower risk of strokes compared to those with lower consumption [11].Specific bioactive components present in fruits and vegetables prompt their consumption as a solution to health problems.For example, the abundance of anthocyanins in blueberries has demonstrated biomedical benefits such as reducing the risk of cardiovascular diseases, age-induced oxidative stress, and inflammatory responses [12].Likewise, lycopene found in tomatoes has demonstrated a reduction in the risk of cardiovascular diseases [13].Furthermore, research has shown that sulforaphane, which is found in Cruciferae vegetables such as broccoli, kale, cabbages, and cauliflower, can help reduce the risk of colon and prostate cancer [14]. Failure to follow proper postharvest practices will result in numerous reactions, such as an increase in respiration rate and ethylene production, acceleration of enzymatic activity (inducing softening by degradation of cell wall components), chlorophyll degradation, and wounding, which may increase bacterial and fungal populations [15,16].Therefore, proper handling, transportation, and storage during harvest can delay spoilage and prevent postharvest losses. Quality of fresh products and storage fundamentals Preserving the quality of fresh produce has posed a challenge to both the horticultural industry and the scientific community.In the early stages of human history, individuals utilized temperature control and stored their products in naturally cooler locations to prolong fresh produce shelf life.However, storage is a complex undertaking that requires careful consideration, as excessively low temperatures can lead to chilling injury [17]. Although temperature control is critical in storage, the use of technologies such as controlled atmospheres (CA), MAP, and coatings is complements used in fresh product industries.Controlled and MAP involve exposing fresh produce to an environment with artificially regulated gas composition, which can be implemented in packaging or storage.This application often involves the reduction of O 2 and CO 2 increment [18].The use of CA has demonstrated successful results in fresh product preservation through the reduction in respiration rate, microbial growth, chlorophyll breakdown, and enzymatic browning [19].Likewise, MAP refers to the fresh produce self-creation of an atmosphere utilizing carbon dioxide (CO 2 ), and ethylene (C 2 H 4 ) produced by biochemical events (e.g., respiration and ethylene production). Meanwhile, while CA is commonly used for long-term storage, MAP has been proven effective for fresh-cut products, including climacteric ones such as avocados, figs, and tomatoes, as well as, for non-climacteric ones like strawberries and berries [20][21][22]. The combination of these treatments is usually utilized for seeking the prolongation of fresh products shelf-life.For instance, a combination of CA short-term storage (20% CO 2 and 5% O 2 ) and MAP packaging was applied in haskap, improving the antioxidant activity, and retaining the firmness after 14 days of storage at 2°C/90-92% RH [23].Additionally, CA (10% CO 2 ) was effective in delaying the fungus count population in strawberries stored at 5°C for 10 days [24].Moreover, the use of CO 2 (10 and 20 kPa) and O 2 (10 kPa) CA in five cultivars of highbush blueberry helped in the preservation of fruit quality stored at 1°C for 6 weeks.Although CO 2 has been demonstrated effective results in CA, other shreds of evidence suggest a decrement in quality traits such as firmness and sugar, acid ratio, and the increment in undesirable conditions (e.g., ethanol, acetaldehyde, and ethyl acetate accumulation) [25]. Firmness Firmness is one of the most relevant factors in human perception of the quality of fresh produce.Accordingly, defining and enhancing this attribute has captivated postharvest researchers.Berries, for example, naturally lose firmness during fruit maturation, such as swift softening in cultivated strawberries that diminishes their shelf-life.[26].Storage temperature is crucial for firmness preservation.For instance, chili peppers experience less loss of firmness stored at 20°C than when stored at 30°C, likewise, the firmness of blueberries stored at 5°C, maintained greater firmness than those stored at 10°C during 7 days [27,28].On the other hand, zucchini fruit stored at 4°C shows an increase in the enzymatic activities of pectin methylesterase, polygalacturonase, and cellulase; these enzymes have been related to cell wall deterioration.Likewise, chilling injury and cell wall degradation are related [29]. Postharvest treatments, including the exogenous application of methyl jasmonate (MeJA), salicylic acid (SA), calcium chloride (CaCl 2 ), and polyamines, have been used to preserve the firmness characteristics of different fruits [30][31][32].For instance, 10 μM MeJA for 24 h at 20°C treatment has been applied in loquat fruit and then stored at 1°C for 35 days.Firmness preservation in comparison to the control was reached through the reduction of chilling injury [30].SA is an exogenous treatment related to firmness preservation and the ripening delay through the inhibition of ethylene biosynthesis [33].Dipping peaches in 1% CaCl 2 solution demonstrated an increment of firmness of the fruit storage (0, 4, and 10°C/95% RH); nevertheless, temperature demonstrated a significant relation to firmness.Higher temperatures lead to a lower firmness in treated peaches [34].In addition, CaCl 2 treatment has shown the strengthening of the cell wall in litchi fruit (5 g L −1 ) solution vacuum infiltration (5 min).This property is attributed to an increment of Ca 2+ , water-soluble pectin, and upregulates the activities of pectinesterase and β-galactosidase while decreasing the activities of polygalacturonase and cellulase [35]. Color Color is a critical factor by which consumers select fresh produce, but it holds more significance than being an organoleptic characteristic.Chemically, pigments like carotenoids can stimulate defense mechanisms in fresh produce and possess powerful antioxidant properties against 1 O 2 , primarily due to the conjugated double bond [36].Likewise, coloration encompasses several physiological and biochemical events mainly related to fresh product ripening.Thus, color is an indicator of the maturity stage of fresh produce.Therefore, the evaluation of color during storage has been reported and is a critical step in determining the effectiveness of postharvest treatment and gaining knowledge of ripening delay.Postharvest treatments are effective in retarding color development, for example, in tomatoes 1-methylcyclopropene (1-MCP) treatment of 250 nL L −1 for 16 h and stored for 18 days at 22°C reached a delay in the color change rate by 4-6 days [37].Likewise, changes in the color of fresh products involve enzymatic activities mainly by polyphenol oxidase (PPO) and peroxidase generating an enzymatic browning which occurs by exposure to air after cutting, slicing, mechanical damage, and cold storage [38].Evidence to prevent enzymatic browning with chemical methods (e.g., sodium bisulfite, ascorbic acid, SA, melatonin, and nitric oxide) has been demonstrated [39,40].For example, sweet cherries treated with melatonin at 100 μM demonstrated a reduction of browning index of 38.5%, in comparison to the control after 45 days of storage at 0°C [41].Moreover, ascorbic acid at 20 mM treatment has confirmed the alleviation of browning in avocados stored at 4°C/95% RH for 14 days through the decrement of POD and oxidative stress [42]. Flavor and aroma Flavor and aroma are essential qualities for fresh produce.Although the visual appeal of fruits and vegetables is often what initially attracts consumers, producers must prioritize preserving and improving flavor and aroma during storage [43]. In fruits, sugars including sucrose, glucose, and fructose, and acids such as malic in apples, tartaric in grapes, quinic in cranberries, and citric in citrus and tomatoes, contribute significantly to flavor.It is important to note that additional elements, specifically polyphenols and flavan-3-ols, can also impact taste by introducing bitterness [44].Conversely, aroma is primarily composed of aldehydes, terpenoids, esters, and alcohols [45].The macro-and micronutrient composition in fresh produce changes as it ripens and is stored, which results in an altered flavor.During this process, crucial roles in the formation of flavor are attributed to events such as the production of ethylene, intensified respiration, and the development of color.Additionally, the response of fruits and vegetables to abiotic and biotic stress during storage leads to the accumulation of taste and aroma-related metabolites, such as anthocyanins, phenolic acids, and terpenes.It should be noted however that oxidative damage to lipids has the potential to generate off-flavors and aromas in fresh products [46].For instance, storing mandarins at 2°C can result in the accumulation of terpenes, which may decrease overall flavor [47].Furthermore, chemical treatments like 1-MCP have been reported to influence the flavor of stored peaches.They can lead to increased sweetness due to a higher accumulation of sucrose, while simultaneously reducing the presence of negative flavor compounds like benzaldehyde and histidine.These outcomes might be attributed to the induction of sucrose synthesis and resynthesis, potentially involving the conversion of reducing sugars to sucrose via sucrose phosphate synthase.Additionally, the transport of cytosolic sucrose to vacuoles is facilitated by the sugar transporter tonoplastic monosaccharide transporter.Genetically, this process is regulated by the expression of genes such as PpSUS4, PpNINVI8, PpSPS3, and PpTMT2 [48]. Weight loss Typically, most fruits become unsellable after experiencing a decrease in weight of less than 5-10% of their original weight [49].The main cause of weight loss is water evaporation through transpiration.During this process, water vapor moves from the surface of the fruit to the surrounding air.The gradient between the water vapor pressure and the environment in contact with the fruit is what triggers this event [50].Due to this, reducing water evaporation in postharvest has been a challenge for postharvest technologists.Reducing weight loss in postharvest treatments has been reported.For instance, in haskap, the use of a CA (20% CO 2 and 5% O 2 ) reduced the percentage of weight loss after 7 days of storage at 2°C, 1.1% in comparison to the control [23].Likewise, chemical treatments such as nitric oxide (NO) at 15 μmol L −1 , showed reductions in weight loss of 5% after 4 weeks of storage at 0°C and 90% RH in peach fruit [51].Moreover, the use of H 2 S (3 mmol L −1 ) demonstrated an enhancement in weight loss in passion fruit.The results showed that the control was 6 times higher than the treatment [52]. Fresh products ripening Fresh produce is distributed in nature and is essential to the human diet.However, consumption limitations arise during the ripening stage due to natural physiological, biochemical, and organoleptic changes [54]. Ripening involves several processes that occur in the latter stages of fresh product growth and the early stages of senescence.The main reactions associated with the ripening process involve a change in membrane permeability, which damages cellular integrity [54].Management of postharvest systems for understanding ripening in fruits involves the comprehension of physiological, biochemical, and molecular events mainly related to ethylene synthesis [55]. Fresh products are subclassified according to the ripening process; thus, climacteric products are related to dramatic increments in respiration and ethylene production during the ripening process, normally climacterics reach significant ethylene levels which induce biological responses [56].On the other hand, non-climacterics do not exhibit significant ethylene production, and recent research suggests the central role of abscisic acid (ABA) in non-climacteric ripening [57]. Physiology of fresh products Respiration The description of cellular respiration is the fundamental energy-conserving process generating adenosine triphosphate (ATP), which can be anaerobic and aerobic.Throughout this process, sugars (mainly sucrose and starch) are reduced into simpler molecules (CO 2 and H 2 O) (Figure 1). In the process of glycolysis, glucose is subjected to a sequence of enzyme-driven reactions in the cytosol, leading to the formation of pyruvate, ATP, and NADH.Pyruvate enters the tricarboxylic acid (TCA) cycle, where it is converted to acetyl-CoA and further metabolized to generate NADH and FADH 2 .In the TCA cycle, acetyl-CoA combines with oxaloacetate to form citrate, which undergoes a series of transformations, including isomerization and decarboxylation, to produce alphaketoglutarate, succinyl-CoA, succinate, fumarate, and malate.These reactions result in the generation of ATP, NADH, and FADH 2 , providing energy for cellular processes [58,59].This process generates reducing equivalents which are used by the mitochondrial ETC to perform the synthesis of ATP [60].During ETC, 0.2-2% of electrons do not follow the process.Instead, a direct leak out of ETC interacts with O 2 to produce − ˙O2 and H 2 O 2 .Thus, plants can modulate the production of ROS through the activation of AOX.Via the activation of AOX, ROS production is minimized through the reduction of − ˙O2 , and the dissipation of the electrons flow across the mitochondrial membrane which prevents electron leakage [61]. During the growth of fresh produce, respiration plays a key role in metabolic activities.For example, the development of pigments (e.g., anthocyanins and carotenoids) results in an increase in fruit respiration due to the energy required for biogenesis. Fresh produce texture can be influenced by high respiration rates which can cause softening due to pectin degradation through metabolic acceleration and induction of cell wall degrading enzymes.In terms of flavor, the changes in total sugar content are affected due to the sugar breakdown to generate CO 2 .Additionally, weight loss occurs as energy is consumed through respiration, which utilizes O 2 to produce CO 2 and H 2 O.A high respiration rate induces the overproduction of ROS which participates directly in programmed cell death inducing spoilage in fresh produce. Ethylene as a key phytohormone during ripening Ethylene is a gaseous plant hormone which is related to fruit ripening, growth, and biological events (e.g., responsiveness to stress and pathogen attack).Moreover, ethylene defines the classification of fresh produce as climacteric.Hence, understanding the ethylene process in fresh products is essential in post-harvesting for determining the optimization of handling, storage, and transportation [62]. Ethylene biosynthesis is catalyzed through two systems, system 1 is an auto-inhibitory phase with basal levels of the phytohormones and reduces C 2 H 4 responses.After, a catalytic process, which increases C 2 H 4 production, is carried by system 2 [63].These pathways start with the initial methionine precursor which is converted into S-adenosyl methionine (SAM).Then, a reaction with pyridoxal phosphate to produce a Schiff base forming cyclopropane ring and 5′-methylthioadenosine (MTAN), subsequently a cyclopropane ring is converted to 1-aminocyclopropane-1-carboxylate (ACC) by ACS enzyme which is an essential event Biochemical dynamics of produce during storage  5 for C 2 H 4 synthesis.ACC is oxidized by ACC oxidase to form ethylene [64].The perception of C 2 H 4 in the endoplasmic reticulum is carried by transmembrane-receptor proteins [65].Ethylene perception is classified into two subfamilies, ethylene receptor 1 (ETR1) and ethylene response sensor 1 (ESR1) belong to subfamily-I, whereas subfamily-II contains receptor isoforms ETR2, ERS2, and ethylene insensitive 4 (EIN4).Moreover, a protein kinase constitutive triple response (CTR 1) and ethylene intensive 2 release the transcription factors ethylene intensive 3 (EIN3), EIL1, EIL2, and ERFs, leading to ethylene responses [66]. Various quality parameters, such as fruit firmness, sugar content, acidity, starch, pectin, enzymes, and aroma volatiles, influence the production of ethylene in fresh produce.Furthermore, environmental factors, such as temperature, humidity, and gas composition, significantly impact ethylene biosynthesis.High temperatures raise ethylene production, while high humidity mitigates it [67]. Ethylene triggers the activation of enzymes involved in cell wall degradation, such as polygalacturonase, which induces softening by breaking down cell wall polysaccharides.Ethylene also coordinates ripening, inducing carotenoid biosynthesis, chlorophyll degradation, and starch hydrolysis into simple sugars, resulting in color changes in fresh produce [68].Moreover, ethylene in conjunction with respiration rate may accelerate metabolic activities through the acceleration of ripening, thus, the higher CO 2 production induces weight loss leading to deterioration in fresh produce quality.Therefore, ethylene content is monitored, and postharvest treatments are focused on its control.For example, the use of edible coatings, such as 1% and 1.5% Tragacanth Gum (TCG), can improve the quality of persimmon fruit by reducing ethylene synthesis.Additionally, abiotic stressors, such as UV-C and blue LED light, can induce ethylene synthesis and alter the quality of fresh produce [69,70]. Abiotic stress Abiotic stress refers to unfavorable situations (high temperature, salinity, high metals, drought, air pollution, and radiation) that plants must face throughout life [71].These conditions cause extensive losses in agricultural production.Nevertheless, in fresh products, controlled abiotic stressors are utilized for enhancing and prolonging quality and shelf life; once the stress is applied, fresh produce launches specific defense responses.For example, heat stress induces metabolic changes accumulating heat shock proteins, and the induction of antioxidant systems in tomatoes and broccoli [72,73].Likewise, the use of chemicals such as exogenous plant growth regulators [e.g., auxins, ABA, gibberellins, jasmonic acid (JA), ethylene, nitric oxide, and SA] can be effective in fresh product stress mitigation [74].The perception of stress triggers the disruption of functional sensors such as calcium (Ca 2+ ), ROS, NO, and phospholipids [75]. The use of abiotic stressors on fresh produce can also trigger genetic responses.In cucumbers, the application of cold temperature (4°C) and MeJA induces CsLOX4 and CsLOX8 gene expression, which are involved in aromatic composition [76].Abiotic stress perception of fresh products leads to a defense mechanism activation; thus, the enhancement of fresh produce quality can be induced.UV-C and the use of NaHS treatment for H 2 S endogenous production will be described below. Ultraviolet radiation Radiation is a physical phenomenon which has been utilized in postharvest for enhancing fresh product quality and prolonging shelf-life.In plant physiology, light is a major factor that has a major influence on driving carbon metabolism [77].Fluorescent lamps produce artificial light and are adjustable for set treatments under different light levels and photoperiods.Ultraviolet light wavelength corresponds to 200-400 nm.In this wavelength range, UV is divided into three subgroups: UV-A (320-400 nm), UV-B (280-320 nm), and UV-C (200-280 nm).These wavelengths have different effects in plants; UV-B and UV-C have been suggested as non-enzymatic and enzymatic antioxidant system promoters (Table 1) [78].Likewise, light exposure represents a critical factor for fresh products and their perception of stress.Thus, changes in light intensity can lead to acclamatory responses and cellular damage triggering ROS production [88,89]. Chemical postharvest treatments enhance quality and prolong shelf-life New techniques and the use of chemical solutions to induce defense mechanisms can improve the quality and extend the shelf-life of fresh produce.In addition to physical treatments, the use of 1-MCP demonstrated the inhibition of ethylene delaying ripening in climacteric fruit.Similarly, the use of NO and hydrogen sulfide (H 2 S) in fresh produce inhibits postharvest diseases and chilling injury, and the use of growth promoters, such as SA, MeJA, and ABA, as exogenous chemical agents are associated with the accumulation of antioxidants [90]. Although the early stages of chemical treatments were focused on decontamination, current research suggests their application in defense mechanisms (enzymatic and non-enzymatic antioxidants, and genetical induction).For instance, the use of 2 mM SA solution in peaches incremented the inhibition percentage of DPPH radical by 10% after 42 days of storage at 0°C; likewise, phenolic content increment 29 mg in comparison with the control.Moreover, firmness displayed higher values and weight loss percentage was reduced [91].Chemical treatments have been utilized as inducers of enzymatic and non-enzymatic antioxidant systems.For example, SOD, CAT, APX, and phenolic components have been stimulated by MeJA (50 μmol L −1 ) in blueberries stored at 22°C for 7 days [92].Likewise, an emergent chemical treatment utilized for enhancing fresh products is H 2 S. In eggplants, the use of 0.1-0.3mM obtained higher firmness levels, anthocyanins, TSS, and vitamin C accompanied by lower weight loss and chilling injury indices after 21 days of storage at 7°C [93]. H 2 S H 2 S is an emerging signaling molecule which induces defense systems against adverse environmental situations such as heavy metals, salinity, postharvest senescence, and biotic stress resistance [94].H 2 S is an endogenous molecule produced in plants; however, the use of sodium hydrosulfite (NaHS) as a donor is used to apply exogenous treatments in the postharvest process [95].The endogenous production of H 2 S in plants involves the participation of sulfite reductase, which catalyzes the reduction of sulfite to sulfide; nonetheless, this process involves two cysteine-dependent reactions and residuals (e.g., β-cynoalanine, synthase + hydrogen cyanide), which produce H 2 S through the detoxification of cyanide at the expense of cysteine [96]. In plants, H 2 S functions have been described in several processes (e.g., alteration of genes and enzymatic activities, regulation of secondary metabolites) [97].Recent studies suggest the importance of H 2 S during fresh product storage due to the enhancement of the antioxidant system.Meanwhile, the enzymatic antioxidant system enhancement has been reported in broccoli, eggplant, and kiwifruit [98].The induction of non-enzymatic antioxidant system has been SOD and CAT increases [87] reported through the induction of the glutathione-ascorbate cycle and phenolic content increment [99].Likewise, the use of NaHS (1.5 mM) treatments resulted in the increment of H 2 S in strawberries from 0.71 to 1.81 mmol kg −1 .On the other hand, heat treatments decreased the levels to 0.85 mmol kg −1 .Moreover, the increment of H 2 S levels stimulated AOX accumulation through the induction of FaAOX3 expression level [100].Other authors' hypotheses suggest the induction of AOX through the inhibition of cytochrome oxidase; in the presence of an aerobic system, cytochrome c, and ascorbate, cytochrome oxidase is rapidly inhibited by sulfide.Thus, when cytochrome oxidase is inhibited, the detection of mitochondrial dysfunction releases the AOX pathway [101]. Oxidative stress Fresh fruits and vegetables are the main sources of antioxidants (e.g., polyphenols, flavonoids, and vitamins) for the human being, nevertheless, a decrease in antioxidants shows up at the harvest and storage stages.Likewise, the generation of ROS throughout these conditions depletes antioxidants through reduction-oxidation reactions [102].Cellular energy metabolism is based on ATP.During this process, 1 O 2 accepts electrons and H + then is reduced to H 2 O [103].ROS are highly reactive molecules with organic components, and the main ROS transformation involves oxide reduction processes from 1 O 2 until OH˙(Figure 2). ROS Singlet oxygen Oxygen is a uniquely structured molecule that undergoes serial reactions because of its electron configuration.The continuous spinning of oxygen electrons alters the nature of the molecule and results in transformations to ROS, The complete mechanism of singlet formation was defined by Apel and Hirt [107] and is illustrated in Figure 3. Singlet oxygen acts as a signaling molecule and allows interaction with membrane carotenoids and thylakoid components.Therefore, 1 O 2 is essential for chloroplast nuclear signaling pathways involved in physiological processes such as growth and activation of defense mechanisms. On the other hand, excessive levels of 1 O 2 induce the oxidation of lipids, proteins, and nucleic acids, resulting in cell damage [108].The double bonds are the primary targets in proteins, specifically amino acids with aromatic ring structures, polyunsaturated fatty acids in lipids, and guanine bases and thiol groups in DNA [109]. Superoxide anion Overall, − ˙O2 is a reduced form of molecular oxygen which is considered a charged anion with a strong oxidant potential.Biologically, this molecule is generated in the mitochondria as a by-product of cell respiration [110].Superoxide anion is produced by the electron reduction of oxygen and can react with reducing substances or macromolecules in the cell.In addition, − ˙O2 can react with SOD to produce H 2 O 2 and react with NO to produce peroxynitrite.Superoxide anion is formed by NADPH oxidase and is generated continuously in the photosynthesis system one moreover is produced in the membrane during photosynthesis, and its production triggers the generation of more active ROS such as OH˙.This conversion is described in Figure 4 and involves a system called the Haber and Weiss reaction [111,112]. Hydrogen peroxide (H 2 O 2 ) Hydrogen peroxide is the two-electron reduction product of O 2 and is vastly generated in biological processes acting as a signaling molecule which responds to various stimuli in plants through serial activities, for instance, calcium mobilization, protein phosphorylation, gene expression, and downstream signaling [114,115].Hydrogen peroxide can diffuse across membranes acting as a signal for fresh product stress perception [116].However, high concentrations of H 2 O 2 have repercussions in plant cells due to its toxicity and induction to ROS generation leading to cell death [117].In cells, H 2 O 2 is produced in peroxisomes, being the photorespiratory glycolate oxidase (GOX) reaction the main event for its production (2-to 50-fold higher than [113].The reaction associates iron with the production of ROS where ˙O2 is a key player participating in two main activities, ferric ion reduces to ferrous ion via the reaction with ˙O2 to turn into OH˙radicals.chloroplasts and mitochondria) [118].The photorespiratory H 2 O 2 system has been suggested by Foyer et al. [119] who described the contribution of the H 2 O 2 pool to signaling pathways that mediate plant growth and stress responses.Hydrogen peroxide participation in photorespiratory metabolism is described in Figure 5. In fresh products, H 2 O 2 synthesis is related to the ripening process.In tomatoes, the high light stress (700 μmol m −2 s −1 ) promoted the accumulation of H 2 O 2 and the induction of ripening due to ethylene interaction [120].In cucumber stored at 5°C for 12 days, the exploration of H 2 O 2 excess has been associated with the induction of chilling injuries and the use of postharvest treatments with 1 μm MeJa and 1 mM NO with diphenyl iodonium 5 mM (DPI) reached promising results reducing about 50% of the chilling injury in comparison with the control [121]. Defense mechanisms Non-enzymatic antioxidants Antioxidant defense systems in fresh products do not only involve enzymes, which have been described before.Fresh products avoid cell death generated by oxidative stress excess using non-enzymatic antioxidants in berries phytochemicals (e.g., phenolic compounds, and ascorbic acid), and glutathione [123].Waśkiewicz et al. [124] described the role of non-enzymatic antioxidants in plants and the mechanism against ROS.Further sections provide detailed explanations of these mechanisms. Phenolic compounds Phenolic compounds are the widest metabolites found in fruits.Phenolic acids, flavonoids, and anthocyanins are studied due to their influence on fruit quality parameters, principally flavor, color, and aroma [125].A phenolic in terms of chemistry involves components with at least one aromatic ring (C 6 ) bearing one or more hydroxyl groups, and their solubility is according to their structure complexity [126].These secondary metabolites are biosynthesized using essential carbon sources and frameworks, principally the glycolytic pathways of phosphoenolpyruvate, and pentose phosphate in the process of the decomposition of sugar.In addition, the shikimate pathway involves the production of aromatic amino acids such as L-phenylalanine (L-Phe), L-tyrosine, (L-Tyr), and L-tryptophan (L-Trp) which are involved with secondary metabolite generation mainly L-Phe and L-Tyr are present in phenolic pathways such as flavonoids, coumarins, chalcones, and lignans generation [127].Phenol component's behavior throughout storage has been deeply studied by postharvest researchers due to their capacity to mitigate ROS.The essential function against oxidative stress is attributed to the reduction of H 2 O 2 to form singlet O 2 through the constitution of a phenylpropanoid group with aromatic rings of one or more OH − groups generating a chemical bonding with H 2 O 2 [128].External chemical treatments and stressors for the stimulation of phenolic compounds in postharvest have been studied.Postharvest accumulation of phenolics by applying light stressors has been reported.For instance, tomatoes exposed to UV-C (4 and 8 kJ m −2 ) and then stored at 14°C, 95% RH for 35 days increased gallic, p-coumaric, syringic, and chlorogenic acids.Consequently, the antioxidant capacity was enhanced [129,130].[119].In the process of photorespiration, the peroxisomal pathway and its relationship with H 2 O 2 as a signaling molecule play a crucial role.During the initial stages of photorespiration, ROS such as H 2 O 2 engages in the formation of glycolate oxidase (GO), there is a correlation between the expression of serine: glyoxylate aminotransferase (SGAT) against pathogens and an increase in H 2 O 2 yield through the enhancement of GO.Furthermore, glycolate and glyoxylate cycling also contribute to the formation of NADP + and NADPH in conjunction with H 2 O 2 . AsA-GSH cycle Glutathione is defined as a tripeptide synthesized from cysteine, glutamate, and glycine.In fresh product defense, GSH is a major low-molecular-weight thiol tripeptide in plant tissues [131].GSH acts against ROS reducing their generation by a disulfuric bond between two glutathione molecules to form its oxidized form [132]. Early research suggested GSH as the major cellular thiol with active cellular redox participation [133].Further research suggested the key role of GSH in the AsA-GSH cycle as a center reaction in the antioxidant system of defense in plants; subsequently, this concept was transferred to postharvest [134].This concept will be defined further.Overall, the GSH concentration is found in cells cytosol, mitochondria, and endoplasmic reticulum, and its biosynthesis involves a downstream of biochemical reactions based on two enzymatic steps using ATP and the constituent amino acids [135].Due to this feature, GSH has been applied as an exogenous treatment triggering an enhancement of postharvest traits in fresh products.For example, a study on bell peppers showed that quality improved when treated with GSH (0.05% w/v) through a sprayed application.The peppers were then stored at 4°C with 80-85% relative humidity for 25 days.The treatment showed a reduction in chilling injury postponing a sharp increment for 10 days and reaching a decrement of 40% after 25 days of storage.These results suggested an enhancement in the bell peppers' antioxidant defense system [136]. Ascorbic acid (ASC) has wide functions in biological processes (e.g., stress resistance, cell expansion division, and light protection).ASC is synthesized in the D-man- nose/L-galactose pathway and its accumulation involves a balanced result of biosynthesis, oxidation, and recycling [137].Moreover, the conversion of ascorbate to mono-dehydroascorbate by APX has a huge contribution to maintaining cellular redox homeostasis under stress conditions [138].Under stress, ASC plays a key role in fruits and its effects have been demonstrated (e.g., cherry tomatoes under cold stress, peaches under heat air exposure and hypobaric condition, likewise in the application of UV-B radiation in lettuce) [139,140].Ascorbate (AsA) which is the reduced form of ACS, is an essential factor in the activation of the AsA-GSH system and its participation in plant cells involves the following steps: APX uses two ascorbate molecules to convert H 2 O 2 to H 2 O and produce MDHA.MDHA can be directly reduced to ascorbate with the help of MDHA reductase and DAHR, using GSH as a reducing agent.GSH is also regenerated by NADPHdependent GR from oxidized GSSG.This process is critical for scavenging H 2 O 2 and maintaining AsA and GSH in different cellular compartments [141].Although AsA and GSH participate in the cycle, the influence of environmental factors and molecular repercussions are different.While AsA synthesis and accumulation are related to light reception, GSH responds with higher irradiances which are not responsive to ascorbate contents (Figure 6) [142]. Enzymatic antioxidants SOD The role of SOD is to function as a defense mechanism by catalyzing the disproportionation of superoxide to molecular oxygen and peroxide.Since SODs play a crucial role in safeguarding cells against toxic molecules produced during aerobic respiration, they have a prominent protective response [143].SODs were recognized as a group of metalloproteins that act in all oxygen-metabolizing cells and all sub-cellular compartments, their localization is accorded to the metal co-factor [144].Moreover, SOD is responsible for the conversion of − ˙O2 into less harmful pro- ducts for plants.In fresh products, SOD activity demonstrated an increase after an atmospheric cold plasma for 60 s in blueberries, likewise, during apple senescence SOD increased in contrast to the healthy apples [145].The use of exogenous treatments seeking the reduction of oxidative stress through increased SOD activity has been used by postharvest researchers to extend the fresh products' shelf life.Ding et al. [146] utilized exogenous oxalic acid and SA to prevent chilling temperature stress.In comparison with the control group, SOD activity was higher in treated mangos.The oxalic acid represented the highest activity, these results enhanced fruit tolerance to low-temperature stress.Likewise, MeJA (0.1 mmol L −1 ) treatment on kiwifruit achieved an enhancement of SOD activity through the AcSOD genetic expression.Generally, SOD is associated with enhanced resistance leading to improved defenses and postharvest features in kiwifruit [147].Melatonin treatment has been reported to enhance SOD activity and contributed to the decay browning and senescence in litchi fruit, 0.4 mM of melatonin treatment was applied, and then, litchis were stored at 25°C for 8 days [148].Gibberellic acid has been utilized as an exogenous source for enhancing fresh quality.In Toon Sprout, this treatment enhances the quality through SOD activity increments during postharvest in short-term cold storage for 5 days [149].SOD activity enhancement in fruits by chemical treatments is described in Table 1. CAT CAT is present in all aerobic organisms and is a key player in catalyzing H 2 O 2 into water and oxygen.Due to this feature, it is a key player in the regulation of stress in plants [150].Also, CAT plays a role in reducing the overall level of − ˙O2 and maintaining cellular homeostasis under normal growth conditions; thus, CAT reduction increases the sensitivity of plants to oxidative stress [151].In fresh products such as strawberries, blueberries, litchi, apples, and oranges, CAT activity increase is a signal of stress response enhancement [152,153].Given this significance, postharvest has led to insights into the regulation of treatments to increase CAT activity.A chitosan treatment in sweet cherry fruit was reported by Pasquariello et al. [154].The cherries were immersed in a 0.5% chitosan solution for 60 s and stored at 2°C for 14 days.The cherries showed an increase in CAT activity.Additional treatments are outlined in Table 2. APX The role of APX is crucial for scavenging ROS, particularly under abiotic stress conditions. MDA accumulation showed a reduction in its accumulation [176] APX is involved in the AsA-GSH and utilizes ascorbate for donating electrons to reduce H 2 O 2 to water [155].APX is the first enzyme involved in the AsA-GSH.APX thus prevents the accumulation of H 2 O 2 at toxic levels for cells [156].APX activity has been reported against stressors such as drought, temperatures, water deficit, salinity, and UV radiation [157].Due to this, the increment of APX during postharvest is an indicator of the oxide reduction state of cells during storage.In mume fruit (Prunus mume), Imahori et al. [158] reported an increment of APX activity during storages at 1 and 6°C for 15 days.Fruit stored at 6°C showed lower activity of APX compared to fruit stored at 1°C.Although chilling injuries occur frequently according to the decrement of temperature, the combined action of antioxidant enzymes resulted in minor chilling injuries stored at 1°C.Other examples previously described exogenous postharvest treatments for improving antioxidant enzymes.APX enhancement has been reported utilizing chitosan coating in loquat fruit (1% w/v) at 7°C for 21 days.APX enhancement was reached.APX activity enhancement in fruits by chemical treatments is described in Table 2. GR GR is a flavoprotein that belongs to the family of NADPHdependent oxidoreductase and plays a huge role in focusing on plants' defense against oxidative stress through the scavenging of ROS [159].GR in collaboration with APX has a key role in AsA-GSH cycle.GR acts in the conversion of oxidized glutathione (GSSG) to reduced glutathione (GSH) using NADPH [160].Harshavardhan et al. [161] described the catalytic activity of GR.In the first step, NADPH acts as a reducing agent of the flavin moiety, which is then oxidized.This leads to a redox activation of the disulfide bridge, resulting in a thiolate anion and a cysteine.In the second step, GSSG is reduced through a thiol-disulfide interchange reaction.The reduction of enzymes is re-oxidized by GSSG.GR is predominantly found in the chloroplast.Nevertheless, isoforms have been found in the cytosol, mitochondria, and peroxisomes [162].In postharvest, ozone (O 3 ) has been involved with the activation of the AsA-GSH cycle.Doses of 6.432, 10.720, and 15.008 mg m −3 , 4°C were evaluated for 0, 14, and 42 days in cantaloupes.The increment of GR activity in treated samples has been related to the enhancement of AsA-CSG as a defense system [163].The addition of ascorbic acid and oxalic acid (40 and 2 nmol L −1 ) to a CA (5% CO 2 and 1% O 2 ) resulted in an increase in GR during 28 days of storage in litchi fruit.This also led to a decrease in H 2 O 2 , which prevented cell leakage and malondialdehyde (MDA) accumulation [164].Likewise, NO treatment has delayed winter jujube ripening, 20 μl L −1 for 3 h and stored at 0°C for 75 days.GR activity reached a sharpening increment over 45 days (11-15 U kg −1 ) as well as, other antioxidant enzymes such as SOD, CAT, and APX demonstrated an enhancement in their activities.In contrast, ROS (H 2 O 2 , − ˙O2 ) decreased significantly [165].GR activity enhancement in fruits by chemical treatments is described in Table 2. Enzymatic participation in the respiration process The role of AOX in fresh products In plants, AOX is oriented for the oxidization of ubiquinol to the four-electron reduction of O 2 to H 2 O; this process is conducted by AOX bypassing proton-pumping through the complexes III and IV in mitochondrial cellular respiration.Meanwhile complexes I, III, and IV of the respiratory chain transport electrons through translocation of protons across mitochondrial membranes inducing ATP synthesis, AOX is not coupled with ATP synthesis and energy accumulation for electron transporting (Figure 7) [101].Furthermore, the suggestion of AOX induction by altering the cytochrome pathway has been a trend in understanding the stimulation of AOX and plant signaling coordination for altering the respiration pathway.Likewise, AOX activity has been related to maintaining cellular homeostasis through detoxification of ROS overproduction. The alteration in the ETC in the mitochondrial membrane induces the regulation of AOX relating to enhanced cyanide-resistant respiration; under adverse environmental conditions total state of respiration is altered through the inhibition of electron transfer, thus, excess ROS are produced.The activation of AOX allows the consumption of excess molecular oxygen and reduces the excessive ROS accumulation.For instance, the enhancement of AOX induction has reached the decrement of − ˙O2 and H 2 O 2 in papaya stored at 1°C for 60 days through the enhancement of cyanide-resistant pathway.In addition, the cytochrome pathway was suppressed [177]. It has been proposed by a number of researchers that AOX is activated when mitochondria are under antagonism stimulus, such as biotic/abiotic stress (e.g., salinity, drought, high light, and chilling).Likewise, the high production of ROS generated by stressors can lead to the AOX activation for alleviating oxidative damage in mitochondria [178].Nevertheless, the general role of AOX is to provide alternatives in energy production. AOX is encoded by a small group of genes divided into two subfamilies, Aox1 and Aox2, meanwhile, Aox1 induction is related to stress responses (e.g., Aox1a, Aox1b, and Aox1b), the Aox2 genes are mainly related to constitutive or developmentally regulated pathway [179].For instance, in sweet potatoes stored at 4°C 90% RH, the upregulation of IbAox1 and IbAox2 mediates the redox homeostasis and proline accumulation being induced by progesterone treatment (100 mmol L −1 ).The first 7 days of storage demonstrate the maximum point of IbAOX1 expression and AOX (0.30 U g −1 ) activity.On the other hand, IbAOX2 had a sharp increase after 14 days of storage [180]. Recapitulation Proper postharvest handling of fresh produce is crucial to prevent losses.Developing effective strategies for harvest, transportation, and storage is particularly challenging for crops that are not well-studied.While temperature and relative humidity are the key factors that delay fresh product senescence, advances in technology have introduced novel techniques that involve controlled and MAP, edible coatings and films, the use of postharvest treatments such as chemicals (e.g., NO, SA, MeJA, H 2 S), and physical treatments such as heat, high light, UV-A, UV-B, and UV-C, which focus primarily on mitigating oxidative stress in fruits and vegetables.These treatments have been shown to improve quality and prolong shelf-life by activating antioxidant enzymes (i.e., SOD, APX, GR, and CAT) and a non-enzymatic system (i.e., phenolics, flavonoids, anthocyanins, glutathione, and ascorbate).This opens new possibilities for emerging crops that often experience inadequate postharvest handling. An emerging area of postharvest research is the use of controlled stress doses on fruit.Fresh produce undergoes biochemical changes when subjected to stressors.For example, treatments with 1-MCP and NO inhibit ethylene biosynthesis, which delays ripening.In the same vein, NaHS can function as a donor for producing endogenous H 2 S, thereby mitigating ROS production in the ETC across the mitochondrial membrane and activating the AOX pathway.Furthermore, UV radiation results in an overproduction of ROS, leading to the activation of AOX in ETC. Conclusion and perspectives The reduction of postharvest losses and the enhancement of product quality to satisfy customer demand have provided new insights into the biochemical dynamics of fresh produce during postharvest.Although oxidative stress has been thoroughly explored, understanding the perception and adaptation systems of fresh produce (e.g., activation of enzymatic and non-enzymatic antioxidant systems) remains a challenging task due to the variety of crops.Despite our theoretical understanding of fresh products and the interplay with stress, further scientific research is needed to determine beneficial doses of stressors.Additionally, the participation of AOX in mitigating oxidative stress and serving as an indicator for the climacteric nature of fresh products can enhance postharvest practices, such as storage, as well as aid in choosing the appropriate type of stress to apply to specific fruits. Nonetheless, understanding biochemical dynamics is challenging and requires particular analysis based on the specific species and cultivar of the fresh produce.Moreover, the introduction of new crops necessitates faster classification of the climacteric nature of fresh products to reduce the time required to create storage strategies.The implementation of these events requires the use of modern tools that ensure precise results.For example, determining climacteric nature can be achieved by utilizing biomolecular tools such as quantifying AOX gene expression and monitoring ethylene and respiration rate.Additionally, in vivo measurement of ROS production and perception across fruits can be conducted to determine the time of response of antioxidant mechanisms and how a specific fresh product perceives specific stress.This review explains the biochemical events that reveal the interplay between stress, storage, quality, and ripening and how they can be optimized to improve postharvest practices, reduce losses, and achieve optimal quality in the future. Figure 2 : Figure 2: ROS from singlet oxygen to hydroxyl radical.Inspired by Kamata and Hirata [103].The mechanism of ROS generation in mitochondria, chloroplasts, and peroxisomes is explained.ROS generation begins with O 2 and ends with the formation of OH in the presence of Fe 2+ via the Fenton reaction.During this process, both enzymatic (e.g., SOD) and non-enzymatic (e.g., glutathione, ascorbate, and carotenoids) antioxidant defenses are activated to defend the cell.Lipid oxidation, protein oxidation, and DNA damage can result from overproduction of ROS. including1 O 2[104].The 1 O 2 production occurs mainly in the chloroplast, induced by the triplet excitation state of chlorophyll in photosynthesis system two (PSII) at an excitation energy of 94 kJ mol −1[105].Thus, 1 O 2 production is mainly related to response under excess light stress when a 3 O 2 molecule becomes excited by photosensitizers, two unpaired electrons become paired, and the ground state of 3 O 2 turns into 1 O 2[106]. Figure 3 : Figure 3: Reaction center of PSII in singlet oxygen production.Inspired by Apel and Hirt [107]. 1 O 2 is generated using light as input energy, the light absorption by chlorophyll P680 and precursors leads to its generation, likewise, the antenna participates in the formation of a triplet state of chlorophyll (Chl 3 ) through the reaction with triple oxygen state ( 3 O 2 ) forms a reduction in Chl 3 to Chl and a molecule of 1 O 2. Figure 4 : Figure 4: Haber and Weiss reaction.Inspired by Koppenol[113].The reaction associates iron with the production of ROS where ˙O2 is a key player participating in two main activities, ferric ion reduces to ferrous ion via the reaction with ˙O2 to turn into OH˙radicals. Figure 5 : Figure 5: Peroxisomal photorespiratory pathway.Model defined by Foyer et al.[119].In the process of photorespiration, the peroxisomal pathway and its relationship with H 2 O 2 as a signaling molecule play a crucial role.During the initial stages of photorespiration, ROS such as H 2 O 2 engages in the formation of glycolate oxidase (GO), there is a correlation between the expression of serine: glyoxylate aminotransferase (SGAT) against pathogens and an increase in H 2 O 2 yield through the enhancement of GO.Furthermore, glycolate and glyoxylate cycling also contribute to the formation of NADP + and NADPH in conjunction with H 2 O 2 . Figure 6 : Figure 6: AsA-GSH.Inspired by Foyer and Noctor [142].Schematic illustration of AsA-GSH cycle, enzyme participation GR glutathione reductase, DHAR dehydroascorbate reductase, MDHAR, monodehydroascorbate, APX ascorbate peroxidase.The final product of the cycle is the reduction of H 2 O 2 to H 2 O through APX. Figure 7 : Figure 7: AOX activity in the ETC.Model proposed by Vanlerberghe [101].Membrane respiration change and AOX participation catalyzing cyanideresistant reduction of oxygen to water without the translocation of protons across the mitochondria membrane, this process acts as a non-energy conserving respiration EFC, complexes are I: NADH-ubiquinone oxidoreductase, II: succinate-ubiquinone oxidoreductase, III: ubiquinol-cytochrome c oxidoreductase, IV: cytochrome c oxidase, and V: ATP synthase. Author contributions: A.D.-S.revised and supervised the manuscript and designed the figures.C.G. oversaw and corrected the manuscript.E.A.L-C performed the literature search, prepared the figures, and drafted the final version of the manuscript.All authors reviewed and approved the final version of the manuscript. Table 1 : UV treatments and antioxidant defense system enhancement Fresh products matrix UV treatment conditions Table 2 : Impact of postharvest chemical treatments on enzymatic activity and ROS	2024-04-09T06:17:25.868Z	2024-01-01T00:00:00.000	{ "year": 2024, "sha1": "13726555f61e88f18c37901fe60e2072ebb57a4b", "oa_license": "CCBY", "oa_url": "https://www.degruyter.com/document/doi/10.1515/bmc-2022-0048/pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "7bc10d948ea0178dcf189e12e203bdee82754c25", "s2fieldsofstudy": [ "Environmental Science", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
53770936	pes2o/s2orc	v3-fos-license	Prognostic factors for recovery and non-recovery in patients with non-specific neck pain: a protocol for a systematic literature review Introduction Neck pain is a common musculoskeletal disorder worldwide. It can result in significant disability and impaired quality of life. More than 50% of patients with neck pain still report symptoms 1 year later despite receiving different forms of non-pharmacological and pharmacological treatment. Identifying patient characteristics that are modifiable or predict recovery and non-recovery for an individual patient might identify ways of improving outcomes. This systematic review aims to comprehensively summarise the existing evidence regarding baseline patient characteristics associated with recovery and non-recovery, as defined by measures of pain intensity, disability and global perceived improvement. Methods and analysis Six electronic databases, PubMed, CINAHL, PEDro Database, EMBASE, Cochrane Library and Web of Science, will be searched, with terms related to the review question such as neck pain, prognostic or predictive research, from inception to 28 September of 2018. Studies will be included if they have investigated an association between patient characteristics and outcomes, with at least one follow-up time point. Two independent reviewers will screen the titles and abstracts followed by a full-text review to assess papers regarding their eligibility. Data from included papers will be extracted using standardised forms, including study and participants’ characteristics, outcomes, prognostic factors and effect size of the association. The risk of bias of each study will be assessed using the Quality in Prognostic Studies tool. A narrative synthesis will be conducted considering the strength, consistency of results and the methodological quality. Ethics and dissemination This systematic review does not require ethical approval. The results will be disseminated through publication in a peer-review journal, as a chapter of a doctoral thesis and through presentations at national and international conferences. PROSPERO registration number CRD42018091183. Introduction Neck pain is a common musculoskeletal disorder worldwide. It can result in significant disability and impaired quality of life. More than 50% of patients with neck pain still report symptoms 1 year later despite receiving different forms of non-pharmacological and pharmacological treatment. Identifying patient characteristics that are modifiable or predict recovery and non-recovery for an individual patient might identify ways of improving outcomes. This systematic review aims to comprehensively summarise the existing evidence regarding baseline patient characteristics associated with recovery and non-recovery, as defined by measures of pain intensity, disability and global perceived improvement. Methods and analysis Six electronic databases, PubMed, CINAHL, PEDro Database, EMBASE, Cochrane Library and Web of Science, will be searched, with terms related to the review question such as neck pain, prognostic or predictive research, from inception to 28 September of 2018. Studies will be included if they have investigated an association between patient characteristics and outcomes, with at least one follow-up time point. Two independent reviewers will screen the titles and abstracts followed by a full-text review to assess papers regarding their eligibility. Data from included papers will be extracted using standardised forms, including study and participants' characteristics, outcomes, prognostic factors and effect size of the association. The risk of bias of each study will be assessed using the Quality in Prognostic Studies tool. A narrative synthesis will be conducted considering the strength, consistency of results and the methodological quality. Ethics and dissemination This systematic review does not require ethical approval. The results will be disseminated through publication in a peer-review journal, as a chapter of a doctoral thesis and through presentations at national and international conferences. PrOsPErO registration number CRD42018091183. IntrOduCtIOn rationale Neck pain is one of the most common musculoskeletal disorders worldwide and may have a significant impact on function and quality of life. [1][2][3][4] In the general population, 30%-50% of adults will experience an episode of neck pain at least once in their lifetime, 5 and the prevalence peaks in adults between 40 and 45 years of age. 2 According to the Global Burden of Disease Study 2016, neck pain is the second most common musculoskeletal condition after low back pain. 4 Disability-adjusted lifeyears increased from 17 million (95% CI 11.4 million to 23.7 million) in 1990 to 24 million (95% CI 16.2 million to 33.4 million) in 2006, and increased again to 29 million (95% CI 19.5 million to 40.5 million) in 2016. 4 Of those who experience acute neck pain, up to 50%-85% will report pain 1-5 years later. 6 Neck pain is also often associated with other complaints such as low back pain and headache, and with poorer self-rated health. 7 Neck pain presents an economic burden for society since it may result in extended periods of sick-leave from work and high use of health services. 8 Most often a specific cause of neck pain symptoms cannot be identified and the label strengths and limitations of this study ► The planned systematic review aims to comprehensively synthesise the available evidence about prognostic factors for recovery and non-recovery in patients with non-specific neck pain. ► This protocol has been developed following the guidance of the Preferred Reporting Items for Systematic Review and Meta-analyses Protocols and has been registered with the Prospective Register of Systematic Reviews. ► Heterogeneity of the definition of recovery or non-recovery in terms of pain intensity, disability or global perceived improvement, may hamper the consistent conclusions. Open access of non-specific neck pain is given, defined as pain in the cervical region without an identifiable specific anatomopathological diagnosis. [9][10][11] Patients with neck pain are commonly referred for non-pharmacological or pharmacological treatment, with considerable heterogeneity of outcomes and recovery rates. [12][13][14][15][16] The literature on recovery and non-recovery from neck pain has shown associations with clinical, sociodemographic and psychosocial patient characteristics, and some authors suggest that these prognostic factors are the key to explaining the outcomes achieved in these patients. 17 18 Factors such as older age, female gender, the presence of low back pain, past history of neck symptoms and previous trauma have been associated with less favourable outcomes in terms of disability, pain intensity or global perceived improvement. 11 19-22 The knowledge of these prognostic factors may help clinicians to better distinguish between patients with a good versus less favourable prognosis, leading to provision of better advice about likely outcomes and better management decisions. 17 23 Despite the increasing number of primary studies published in recent years, the last systematic review on prognostic factors in patients with neck pain was published in 2009. 11 19-21 24 This and other systematic reviews have examined risk factors for developing neck pain and prognostic factors for course and recurrence, but they have combined populations with traumatic and non-traumatic causes of neck pain and none of them has focused on prognosis for recovery or non-recovery. 11 19-21 24 Given its high rate of prevalence, the identification of prognostic factors for recovery and non-recovery in patients with non-specific neck pain is relevant for effective patient management. However, to our knowledge, no published systematic review has yet identified and synthesised the available evidence concerning prognostic factors for recovery and non-recovery, regarding pain intensity, disability and global perceived improvement, in adult patients with non-specific neck pain. Objectives The aim of this review is to identify, assess and synthesise the available evidence about prognostic factors for shortterm and long-term recovery and non-recovery, in terms of pain intensity, disability and global perceived improvement in patients with non-specific neck pain. MEthOds And AnAlysIs This review protocol is registered with the International Prospective Register of Systematic Reviews by the Centre for Reviews and Dissemination at the University of York, with the number CRD42018091183. 25 The protocol and completed review will be reported following the guidance of the Preferred Reporting Items for Systematic Review and Meta-analyses Protocols, 26 and the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA), 25 respectively. Eligibility criteria Studies will be selected according to the criteria outlined below. Participants and settings Studies will be included in this review if they investigate adults (aged over 18 years of age) with non-specific neck pain, defined as pain in the cervical region without a specific anatomopathological diagnosis. [9][10][11] The studies that include people with specific causes of neck pain (eg, nerve root compression, trauma, malignancy, infection), inflammatory arthritis (eg, rheumatoid arthritis, spondyloarthritis) or neurological diseases (eg, multiple sclerosis) will be excluded. No restrictions will be applied to the setting. Exposure and outcome measures This systematic review seeks to identify demographic, clinical and psychosocial factors assessed at baseline that examine the association with recovery or non-recovery in patients with non-specific neck pain. The definition of demographic, clinical and psychosocial factors was established according to a theoretical framework, based on the following constructs: Demographic factors: 'Socioeconomic characteristics of a population expressed statistically, such as age, sex, education level, income level, marital status, occupation, religion, birth rate, death rate, average size of a family, average age at marriage'. 27 Clinical condition: signs and symptoms of the disease or clinical condition, presented or described by the patient or found in clinical evaluation, defined in Medical Subject Headings (MeSH) terms as 'disease attribute-clinical characteristics of disease or illness'. Psychosocial factors: 'Social factors include general factors at the level of human society concerned with social structure and social processes that impinge on the individual. Psychological factors include individual-level processes and meanings that influence mental states'. 28 Study participants might or might not be treated. If treated, the interventions may include any management and care of any duration, with an explicit description of treatment (eg, medication, exercise, manual therapy, acupuncture, surgery, etc). The outcome of this systematic review is recovery and non-recovery regarding disability, pain intensity and global perceived improvement. The definition of recovery/ non-recovery presented by the authors of original articles (eg, based on definition of 'recovery' or 'complete remission' or 'functional recovery' or 'non-recovery') will be accepted, if they are determined by at least one of the following outcome domains: disability, pain intensity and global perceived improvement. Any recovery/non-recovery criterion presented by the authors of the original studies will be accepted. It can be based on a cut-off point (eg, recovery in disability: decrease at least 27% in Neck Disability Index; score higher than 5 in a Global Recovery Scale), established Open access by authors, or based on a recovery continuum. In the absence of a clear recovery/non-recovery criterion, the studies will be excluded. All outcome measures of disability, pain intensity and global perceived improvement will be accepted if they correspond to validated instruments (appropriate for culture and language), specific for people with neck pain, and with at least one follow-up time point. Where several measures were used, all the measures will be extracted and classified according to the domain. Outcomes will be considered at two time points from baseline defined as ≤6 months and >6 months. If multiple data of disability, pain intensity or global perceived improvement are provided, data close to these time points will be prioritised. Study designs Eligible studies will be prospective cohort studies or randomised controlled trials (RCTs) with prognostic factor analysis, published in peer-reviewed journals. RCTs will only be included if information regarding prognostic factors for recovery and non-recovery are available in the published report. The studies will have to identify prognostic factors at baseline and report a statistical association (or lack of association) with an outcome (disability, pain intensity and/or global perceived improvement). Studies including people with pain elsewhere will be included provided if the data for the participants with non-specific neck pain are reported separately. Cross-sectional studies, case series, case reports, case-control studies, systematic reviews, conference proceedings and Masters or PhD theses will be excluded. Language and time frame No language or geographical restriction will be applied. The search in each database will be performed from inception to 28 September 2018. Information sources and search strategy Searches will be conducted in six electronic databases: PubMed, CINAHL (via EBSCO), PEDro Database; EMBASE (via Elsevier); Cochrane Library (via Wiley Online Library) and Web of Science (via Clarivate Analytics). Additionally, hand searches of the reference lists of all included studies and previously published systematic reviews of prognostic factors for non-specific neck pain will be conducted to ensure completeness of the search. 11 19 21 The search strategy will be developed in consultation with a medical librarian with expertise in systematic review searching. A variety of terms related to key subject areas of the review question such as neck pain, prognostic or predictive research will be used. Keywords or database-specific subject headings (eg, MeSH) and the Boolean operators 'OR' and 'AND' will be used to combine the search terms. The search terms will be adjusted to the specificities of the different databases. A draft of the PubMed search strategy is included in table 1. study selection All potentially eligible articles will be retrieved and organised in the Mendeley reference manager software and duplicate publications will be deleted. Two independent reviewers will screen the titles and abstracts identified in the search to identify potentially eligible studies and will perform full-text review of all those identified to determine inclusion. Reasons for exclusion will be documented in tabular format. Any disagreements between reviewers will be resolved by consensus or by consulting a third independent reviewer if consensus cannot be reached. The PRISMA flow diagram will document included and excluded studies with the reasons for exclusion. 26 29 Reviewers will not be blind to the study authors, institutions or journals. data extraction Two independent reviewers will extract relevant data from each selected study, using a piloted standardised data extraction form. To ensure consistency of data extraction, the form will be tested on a sample of five studies, prior to the main data extraction. Any discrepancies in data extraction will be resolved by consensus or discussion with a third reviewer if needed. Reviewers will not be blind to the study authors, institutions or journals. The extracted data will consist of: 1. Study identification (authors, year of publication, volume, issue and pages); 2. Study characteristics (study setting, study design, sample size, type of intervention, follow-up length, dropout rate, source of funding and country of origin); Open access 3. Participants' characteristics (age, gender, duration of neck pain, pain intensity, disability); 4. Prognostic factors (any demographic, clinical and psychosocial factors defined according to a theoretical framework described in 'Exposure and outcome measures' section); 5. Outcomes (any measure of recovery or non-recovery in terms of pain intensity, disability and global perceived improvement, the cut-off point or the statistical analysis of clinical improvement to define recovery/non-recovery and rate of recovery, when available); 6. Effect size (measures of unadjusted and adjusted associations (with description of variables adjusted) reported between prognostic factors and outcomes). If the studies have assessed multiple outcomes, only the information that is relevant to this systematic review research question will be extracted. If there are missing data about study characteristics, methods or measures of association, the study authors will be contacted and asked to provide these data with a maximum of three email attempts in a 2-month period. If the same data have been reported in multiple study publications, the duplicated data will be presented only one time, to minimise the overrating of any prognostic factors investigated in the same sample. Assessment of risk of bias The risk of bias of the included studies will be assessed through the Quality in Prognostic Studies (QUIPS) tool. 30 This tool addresses six potential domains of bias: study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding and statistical analysis and reporting. Two independent reviewers will assess the risk of bias of each included study, following training in risk of bias assessment. The training will consist of application of the QUIPS tool to three articles, with either low, moderate or high risk of bias, investigating prognostic factors for recovery or non-recovery from low back pain. Interrater agreement for each domain will be evaluated by percentage of agreement (calculated as the number of agreement domains divided by the total number of domains). If <90% agreement is observed, a second training programme will be conducted. The results of the risk of bias assessment for each included study will be presented in tabular format by each domain and overall. Reviewers will first assess the relevant risk of bias items in each domain and then produce an overall judgement based on these ratings. Each domain and overall risk of bias will be categorised as low, moderate or high risk of bias. An overall judgement of low risk of bias will require all six bias domains to be rated as low risk of bias. Any disagreements between reviewers will be resolved through discussion or consultation with a third reviewer if required. data synthesis A formal meta-analysis is not planned for this systematic review as the populations, type of treatment received, definitions or prognostic factors, outcomes and methods/ tools are anticipated to be too heterogeneous. 31 32 Consequently, a narrative synthesis will be conducted taking into account risk of bias and the strength and consistency of significant associations. We will extract and report all unadjusted and adjusted measures of association from included studies. Associations with outcome will be defined as a significant (p<0.2) univariable association, a significant (p<0.05) adjusted association (multivariable) or a significant (p<0.05) association in other predictive analysis (linear or multiple regression). Effect sizes will be represented as an OR or relative risk (RR) and considered as significant when the 95% CIs do not include 1, or as a coefficient of determination (R 2 ) and beta-coefficient (b) when the 95% CIs do not include 0. Results will be analysed using the levels of evidence proposed by Furlan et al 33 : A. strong evidence, defined as consistent (>75%) findings among multiple (≥2) high-quality studies; B. moderate evidence, defined as findings in one high-quality study and consistent (>75%) findings in ≥2 low-quality studies; C. limited evidence, that is, findings in one high-quality study or consistent findings in ≥3 low-quality studies; D. conflicting or inconclusive evidence, that is, <75% of the studies reported consistent findings or the results were only based on one study. To simplify the data presentation, the prognostic factors will be categorised into three main domains: demographic, clinical and psychosocial factors, according to the theoretical framework defined a priori. A tabular form for each category will be made with the discrimination of all the factors, with data of outcomes, effect size as well as the level of evidence. Additionally, homogeneous subgroups will be defined according to pain duration (acute/subacute (0-12 weeks) and chronic (>12 weeks)), type of treatment (medication, exercise, manual therapy, acupuncture, surgery, etc) and outcome measurement (pain, functional limitations, global perceived improvement/recovery). Should we decide on any other categorisation, we will note this and provide a rationale. To assess the robustness of our evidence synthesis, a sensitivity analysis will be carried out to examine the influence of the risk of bias whenever possible. Patient and public involvement This article reports a protocol of the systematic review. Therefore, patients and or public were not involved and individual patient data will not be collected. EthICs And dIssEMInAtIOn A comprehensive systematic review to synthesise the available evidence about prognostic factors for recovery Open access and non-recovery in patients with non-specific neck pain is needed considering the multiplicity of the clinical, sociodemographic and psychosocial patient characteristics identified in the literature. Moreover, and given the expected clinical and/or methodological heterogeneity in the individual studies that will be included in this review, the review aims to provide possible directions for standardisation of participants, outcome measures, cut-off points and follow-up time points in future observational studies. Formal ethical approval is not required, as individual patient data will not be collected. The results will be disseminated through a peer-reviewed publication and conference presentations and included in a chapter of a Doctoral thesis. AMEndMEnts If there are any amendments, we will register the date, describe the change and the rationale. Changes will not be incorporated into the protocol but in the PROSPERO register and will be identified in the systematic review.	2018-12-02T16:55:09.733Z	2018-11-01T00:00:00.000	{ "year": 2018, "sha1": "92ce9564039b5fa59400512133ab23138bac1735", "oa_license": "CCBYNC", "oa_url": "https://bmjopen.bmj.com/content/bmjopen/8/11/e023356.full.pdf", "oa_status": "GOLD", "pdf_src": "Highwire", "pdf_hash": "db0ef45ddeb86dd429f007a436fc4a2ac5d1d18d", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
201646597	pes2o/s2orc	v3-fos-license	Coaxial GaAs/(In,Ga)As dot-in-a-well nanowire heterostructures for electrically driven infrared light generation on Si in the telecommunication O band Core-shell GaAs-based nanowires monolithically integrated on Si constitute a promising class of nanostructures that could enable light emitters for fast inter- and intrachip optical connections. We introduce and fabricate a novel coaxial GaAs/(In,Ga)As dot-in-a-well nanowire heterostructure to reach spontaneous emission in the Si transparent region, which is crucial for applications in Si photonics. Specifically, we achieve room temperature emission at 1.27 $\mu$m in the telecommunication O band. The presence of quantum dots in the heterostructure is evidenced by a structural analysis based on scanning transmission electron microscopy. The spontaneous emission of these nanowire structures is investigated by cathodoluminescence and photoluminescence spectroscopy. Thermal redistribution of charge carriers to larger quantum dots explains the long wavelength emission achieved at room temperature. Finally, in order to demonstrate the feasibility of the presented nanowire heterostructures as electrically driven light emitters monolithically integrated on Si, a light emitting diode is fabricated exhibiting room-temperature electroluminescence at 1.26 $\mu$m. Introduction InAs quantum dots (QDs) embedded in strain-reducing (In,Ga)As layers, so-called dot-in-a-well (DWELL) structures, have been established as suitable heterostructures to efficiently extend the emission range of the InAs/GaAs material system to the 1.3 µm spectral region. 1,2 In particular, these structures have found application as the active region of optolectronics devices monolithically integrated on Si. 3,4 However, the heteroepitaxy of III-As semiconductors on Si substrates for such applications is challenging and requires thick buffer layers between the Si substrate and the active region. Furthermore, advanced designs, such as the addition of layers acting as dislocation filter to reduce the density of threading dislocations and minimize their impact on the device performance, have to be employed. In the context of III-V integration on Si substrates, nanowires (NWs) have been recognized as a promising alternative approach. 5 The ability of NWs to accommodate large lattice-mismatches without introducing extended defects, thanks to the high surface-to-volume-ratio and small footprint on the substrate, 6 and the possibility to realize heterostructures in a core-shell geometry, 7 which provides a large active area in relation to the footprint of the NWs on the substrate, make these structures excellent candidates for the development of optolectronic devices such as laser or light emitting diodes (LEDs). [8][9][10] In this work, we demonstrate the growth of a coaxial GaAs/(In,Ga)As DWELL NW heterostructure, combining the NW geometry for integration on Si substrates with the DWELL heterostructure that enables to shift the emission to longer wavelengths. In fact, this novel heterostructure allows us to reach the transparent window of Si and, in particular, achieve room temperature operation in the telecommunication O band (1.26-1.36 µm), demonstrating a new functionality of a heterostructure integrating QDs with NWs. Hierarchical structures combining QDs and NW have so far been considered for specific applications such as single photon emitters 11 or opto-mechanical systems for sensing, 12 but not to extend the spectral range of light emission. To realize this novel structure, we had to overcome the fact that InAs QDs do not form on the {110} sidewall facets of GaAs NWs. 13 Thus, we exploited the recently demonstrated capability of Bi surfactants to promote QD formation on GaAs NW sidewalls. 14,15 Normally, if InAs is deposited on GaAs{110} surfaces, including the {110} sidewall facets of self-seeded GaAs NWs, strain relaxes plastically before the Stranski-Krastanov transition from two-to threedimensional growth takes place. 16 The use of a Bi surfactant modifies surface energies, enabling the formation of InAs QDs to develop a coaxial DWELL structure. In addition to a structural analysis based on scanning transmission electron microscopy (STEM), we present characterization of the optical emission properties of the DWELL NWs by cathodoluminescence (CL) and photoluminescence (PL) spectroscopy. Furthermore, in order to establish the feasibility of the presented heterostructures as electrically driven light emitters monolithically integrated on Si, we demonstrate room temperature operation of a DWELL NW LED with emission wavelength of 1.26 µm. Results and discussion All samples studied in this work were grown by molecular beam epitaxy (MBE) on Si(111) substrates. Scanning electron micrographs of a DWELL NW sample synthesized on a patterned Si(111) substrate are shown in Figure 1. A cross-sectional schematic of the DWELL NW heterostructure is included. The coaxial DWELL heterostructure is grown around GaAs NWs with a diameter of about 100 nm and consists of an InAs QD shell embedded in an about 9 nm-thick In 0.15 Ga 0.85 As quantum well (QW) shell. A Bi flux was supplied during the deposition of InAs to promote QD formation. 14,15 A 50 nm thick outer GaAs shell completes the confinement in the DWELL heterostructure. The total diameter of the DWELL NWs is approximately 220 nm. Microstructural characterization by STEM of a DWELL NW is presented in Figure 2. A cross-sectional micrograph acquired using the scanning high-angle annular dark field (HAADF) mode 17 is shown in Figure 2a. In this imaging mode the contrast is mainly determined by the atomic number Z (Z-contrast). The NW exhibits a hexagonal cross-section defined by {110} side facets and the coaxial active region is clearly visible in the HAADF micrograph. An enlarged false-color micrograph of the facet indicated by the dashed box in Figure 2a is presented in Figure 2b to illustrate the presence of QDs in the DWELL heterostructure. The yellow areas indicate higher HAADF intensity, corresponding to a locally enhanced In composition within the projected volume, which can be attributed to the QDs. A quantitative analysis is presented in the lower part of Figure 2. HAADF intensity (I HAADF ) profiles taken for two different facets highlighted in Figure 2a are depicted in Figures 2c and 2d, respectively. In Figure 2c (blue profile), the I HAADF level for the DWELL active region is almost constant, which we attribute to an (In,Ga)As QW segment where QDs are not present/detected. In contrast, in Figure 2d (magenta profile) the I HAADF profile shows higher intensity spots (indicated by black arrows) within the DWELL active region, that correspond to the extended yellow areas in Figure 2b. As introduced above, we attribute these local variations in I HAADF to InAs QDs present in the DWELL heterostructure. The slope in the I HAADF profile in Figure 2d arises from a small gradient in the thickness/tilt of the TEM NWspecimen. The In content profile obtained for this facet of interest is presented in Figure 2e. The estimated In distribution is obtained from the analysis of the noise-corrected HAADF intensity profile (corresponding to the area highlighted in Figure 2c) following the procedure reported in Ref. [18]. In brief, the method relies on the dependence of the HAADF contrast R = I InGaAs(x) /I GaAs on the In content x 18,19 after correcting for specimen thickness variations and assuming γ = 1.7 in I HAADF ∝ Z γ . The suitability of γ = 1.7 is determined from measurements on planar reference samples of known composition. The estimated In content in the QDs lies in the range x= 17-27%. We should note that the In content that we estimate is averaged over the specimen thickness and that the reported values should be thus taken as lower bounds. Hence, the true In content associated to the QDs is probably higher. The determination of the actual In content and distribution within the QDs, as well as their precise shape, would require additional advanced microstructural characterization beyond the scope of this work. 13,20 The luminescence of the DWELL NWs is first studied by CL spectroscopy as presented in the upper part of Figure 3 (Figure 3a-d). A low temperature CL spectrum is displayed in Figure 3a. The spectrum corresponds to the averaged emission of the area presented in Figures 3b and 3c, containing about 30 NWs, and is dominated by a broad band centered at 1.1 eV (1.13 µm). At higher energy, there is a second band of lower intensity peaking at 1.35 eV (0.92 µm). We attribute this high energy band to transitions in the (In,Ga)As QW and the low energy band to transitions in (In,Ga)As QDs embedded in the QW, as discussed in more detail further below. The broad energy distribution of the low energy band reflects the variations of dimensions and composition of the QDs. Monochromatic CL maps collected at the peak energies 1.345 eV and 1.088 eV are presented in Figure 3b and Figure 3c, respectively. The CL map in Figure 3b shows that the high energy emission originates from an approximately 350 nm long segment close to the top of the NWs, that we assign to a wurtzite (WZ) (In,Ga)As coaxial QW segment, as we have recently reported for similar structures without QDs. 21 The formation of a WZ segment on top of zincblende (ZB) GaAs NWs during core growth is related to the consumption of the Ga droplets prior to shell growth. The CL map in Figure 3c shows that the low energy emission originates from spatially localized states distributed along the NWs, consistent with its attribution to QD states. A more detailed analysis of the CL data reveals that the QW emission from the ZB region is very weak, pointing to an efficient carrier transfer to the QDs. Interestingly, we observe no overlap between the WZ QW segment and the areas with localized QD emission. This suggests that QDs do not form in the WZ segments of the NWs. In Figures 3b and 3c, no CL signal is detected from the parasitic layers grown on the substrate, as expected for the polycrystalline material deposited on the amorphous oxide layer. An additional hyperspectral CL linescan acquired for a single DWELL NW is presented in Figure 3d, which further corroborates the attribution of the low energy emission band to QDs. Spatially localized sharp spectral features assigned to the recombination in the QDs are present along the NW. These emission lines are elongated along the NW axis, which can be explained by diffusion of carriers excited by the electron beam to the QDs. It is important to note that the limited spectral range of the Si detector used to record the linescan (cut-off at ≈ 1060 nm) does not allow us to investigate the broad low energy emission band of the DWELL NWs; and the QDs revealed in this CL linescan only correspond to the smaller ones in the NW. The bottom part of Figure 3 presents low temperature PL spectra of as-grown DWELL NWs. In these experiments, only 2-3 NWs are excited simultaneously. Sharp spectral features are revealed at low excitation power density (Figure 3e), similar to the transitions observed for InAs QDs on GaAs NW facets. 22 Figure 3f shows in linear scale the spectral range marked in Figure 3e. The spectral linewidth of the most prominent transitions is approximately 200 µeV, limited by the spectral resolution of the experimental setup. The sub-meV spectral linewidth shows that these transitions originate from localized excitons, consistent with optically active QDs in the DWELL heterostructure. In addition, we investigate the spontaneous emission of the DWELL NWs by PL spectroscopy as a function of temperature from 10 to 300 K in Figure 4. In these experiments, around 60-70 NWs are excited simultaneously. The temperature-dependent PL spectra are presented in Figure 4a. Again, we observe the two different emission bands seen in the CL spectrum (Figure 3a), with a high energy component (1.35 eV at 10 K) attributed to QW transitions and a low energy component (1.10 eV at 10 K) attributed to QD transitions. At room temperature, the emission is dominated by the low energy QD component. The evolution with temperature of the peak emission energy and full width at half maximum (FWHM) for the QD band are presented in Figures 4b and 4c, respectively. With increasing temperature, the emission energy shows a strong redshift to 0.97 eV at 300 K. We should note that the observed redshift is about 20 meV larger than that expected for a random In 0.385 Ga 0.615 As bulk alloy with the same peak emission at low temperature, 23 as presented by the solid line in Figure 4b. Simultaneously, the FWHM is reduced from 162 meV at 10 K to 104 meV at 300 K. At room temperature, the peak emission wavelength of these DWELL NWs is 1.27 µm, thus reaching the O band (1.26-1.36 µm) used for data communication. Figure 4d shows that the PL intensity I PL integrated over the full spectral range decreases with increasing temperature by about two orders of magnitude. We analyze this variation with a simple phenomenological model with two non-radiative channels for charge carriers: . T is the temperature, k is the Boltzmann constant, and I 0 is a constant. The non-radiative channels are described by the prefactors a 1 and a 2 and activation energies E 1 and E 2 , respectively. A fit of the experimental data results in the activation energies E 1 = 44 ± 5 meV and E 2 = 188 ± 37 meV. These activation energies should be considered as an average description over all NWs in the ensemble. Interestingly, these values are similar to the activation energies (E 1 = 34 meV and E 2 = 262 meV) reported for a planar InAs/(In,Ga)As DWELL structure with (001) orientation. 24 Furthermore, the activation energy of the first channel E 1 = 44 ± 5 meV is similar to one of the decay channels reported for (In,Ga)As/GaAs NWs with shell QW. 25 The activation energy of the second channel E 2 = 188 ± 37 meV is comparable to the energy difference between the QW and QD bands observed in the low temperature PL spectra (≈ 250 meV) and, therefore, points to the thermal escape of carriers from the QDs to the QW. This finding along with the narrowing and the strong redshift of the emissionas well as the fact that the thermal quenching is moderate compared to four orders of magnitude intensity reduction observed for standard (In,Ga)As/GaAs shell QW NWs without any QDs, 25 suggest a thermal redistribution of charge carriers in the QDs of the DWELL heterostructure. This mechanism is sketched in Figures 4e and 4f and it is well established for planar DWELL heterostructures. 24 At low temperature (Figure 4e), the charge carriers populate QD levels, as well as the QW levels (predominantly in the WZ segment in our case, due to the efficient carrier transfer in the ZB NW) and recombine mostly radiatively, thus contributing to the broad PL spectrum. With increasing thermal energy (Figure 4f), charge carriers can escape shallow energy levels of smaller QDs and undergo non-radiative recombination, as observed by the quenching of the PL intensity above 160 K, or are recaptured in larger QDs with deeper energy levels, which manifests itself in the narrowing and redshift of the emission. The deeper confinement in larger QDs provides protection from non-radiative recombination centres and explains the enhanced room temperature emission as well as the longer emission wavelength. The presented optical characterization evidences that the emission range of the coaxial DWELL NWs lies in the Si transparent window. In order to demonstrate the feasibility of the presented DWELL NWs as electrically driven light emitters on a Si platform, we fabricated LEDs from self-assembled ensembles of DWELL NWs with a radial p-i-n junction. GaAs NW based LEDs on Si substrates operating in the near infrared region have already been demonstrated by different research groups. 8,9,26,27 An emission wavelength of 985 nm has been demonstrated for a coaxial multishell (In,Ga)As/GaAs NW LED, 27 and a nanoneedle LED with a similar (In,Ga)As/GaAs QW active region reached 1.13 µm, 28 which constitutes the longest wavelength of electroluminescence (EL) demonstrated for GaAs-based NW-like structures. For the present study, the LED multishell structure grown around an approximately 100 nm thick Be doped p-type GaAs NW core consists of a 10 nm thick undoped GaAs shell, a coaxial DWELL heterostructure, a 10 nm thick undoped GaAs shell, and an outer 50 nm thick n-type GaAs shell, which was doped with Si in an approach established previously. 29 The schematic cross section of the LED structure is shown as an inset in Figure 5b. The total diameter of the LED NWs is approximately 260 nm. The current-voltage (I-V) characteristics of a 300 µm diameter LED, contacting approximately 10 5 NWs in parallel, are presented in Figure 5a, revealing rectifying behavior of the fabricated device. Nevertheless, we observe an exponential increase of the current at negative bias beyond −2 V in the reverse region of the I-V curve, which is probably caused by defect-supported intraband tunneling. The analysis of the forward region of the I-V curve using the standard diode equation gives an ideality factor n = 15.2, a series resistance R S = 16.9 Ω and a saturation current I S = 8.40×10 -7 A. High ideality factors have been previously reported for core-shell NW LEDs 9 and, in addition to the turn-on voltage of about 2.5 V (larger than the GaAs bandgap), suggest that the actual device consists of more than one diode in series, as a result of additional barriers in the device. One additional barrier could be present at the interface between the Si substrate and the NWs. A limited incorporation of dopants in the initial stages of NW growth, as well as a possible indiffusion of Si atoms into the NW from the substrate, could result in a doping level too low to obtain a good contact to the p-type substrate. The thin native oxide layer and possible structural defects at the interface may also contribute. An additional barrier could be located in the top contact between the top transparent layer made of indium tin oxide (ITO) and the n-type GaAs outer shell of the NWs. 30 The study of the interfaces described above and optimization of the dimensions and doping profiles of the doped sections of the radial p-i-n junction in our devices are necessary to improve the electrical performance of the LEDs. In addition, the use of patterned substrates should lead to more homogeneous NW arrays and better isolation of the p-i-n junction from the substrate. The EL spectrum of one DWELL NW LED measured at room temperature at an injected current of 19 mA is presented in Figure 5b. The emission band peaks at 1. 26 28 Optically excited, monolithically on Si integrated light sources based on GaAs NWs or related structures have been demonstrated up to 1.1 µm using GaAs-[(In,Ga)As/(Al,Ga)As] core−shell NWs with multiple (In,Ga)As QWs 31 and up to 1.44 µm using (In,Ga)As NWs. 32 More recently, the tunability of the GaAs bandgap up to 1.42 µm has been demonstrated exploiting strain partitioning in NWs. 33 LEDs operating in the telecommunication S band (1.46 -1.53 µm) have been demonstrated for InP based nanopillars on Si with radial (In,Ga)As shell QW. 34 Nanoneedle/nanopillar structures resemble NWs, but they typically exhibit base diameters in the micrometer range and are characterized by a tapered geometry, which has been demonstrated to sustain helicoidal guided photonic modes. 35 The DWELL NW heterostructure introduced here is grown around a GaAs NW core with diameter of about 100 nm and presents a smaller footprint on the substrate. This characteristic could be beneficial for the integration on photonic elements such as Si waveguides [36][37][38] and offers greater design flexibility to tune the total diameter by simply modifying the thicknesses of the GaAs shells, which would enable the tailoring of leaky or guided modes for specific applications. 39 Conclusion In conclusion, we have presented novel GaAs/(In,Ga)As DWELL NW heterostructures grown on Si substrates as sources for infrared light emission in the Si transparent spectral window. Microstructural characterization by STEM illustrates the presence of QDs in the DWELL heterostructure; and spectral widths below 1 meV, as observed in low temperature PL measurements, reveal the QD nature of the emission. Carrier redistribution to larger QDs in the structure by thermal escape and recapture explains the strong redshift of the emission with temperature and allows us to reach long wavelength emission at room temperature, similarly to planar DWELL structures. We have demonstrated room temperature operation of DWELL NW LEDs on Si at 1.26 µm; thus establishing the application of GaAs-based NWs for monolithically integrated electrically driven optoelectronics on a Si platform. These results illustrate the potential of heterostructures combining QDs and the NW geometry and push the electroluminescence from GaAs-based NWs into the spectral region relevant for integrated optical interconnectors and data communication, opening up new perspectives for Si photonics integration. Methods Growth -The samples were grown by solid-source MBE on Si(111) substrates. The MBE system is equipped with In, Al, Bi and two Ga effusion cells, Be and Si effusion cells for dopants and two valved cracker sources for supply of As 2 . Fluxes are calibrated and expressed in the following in terms of equivalent growth rate on a GaAs (001) surface in monolayers per second (ML/s). An optical pyrometer was used for measuring the substrate temperature. DWELL NWs were grown on p-type (B doped) Si(111) substrates covered with a 20 nm thick thermal oxide mask patterned by electron beam lithography (EBL). Before loading into the MBE system, patterned substrates were etched with a 1% HF solution and treated in boiling water for 10 minutes. 40 Ga was deposited on the substrate for 70 s at 0.3 ML/s GaAs equivalent growth rate to form Ga droplets on the surface and GaAs NWs were then grown by the self-catalyzed vapor-liquid-solid (VLS) method at 630 ᵒC with a Ga flux of 0.3 ML/s and V/III ratio of 3.9 for 30 minutes. Then, the Ga flux was closed and the Ga droplets atop of the NWs were consumed by crystallization to GaAs under a 4 ML/s As 2 flux at the growth temperature of 630 ᵒC. Subsequently, the DWELL structure, which consists of a 2 nm thick In 0.15 Ga 0.85 As shell, a 0.3 nm (2 ML) thick InAs quantum dot (QD) shell and a 7 nm thick In 0.15 Ga 0.85 As shell, was grown as active region around the GaAs NW core at 420 ᵒC, a growth rate of 0.2 ML/s and a V/III flux ratio of 20. Growth parameters were chosen to limit adatom diffusion in order to obtain homogeneous shells. 27 Samples were rotated at 6 revolutions per minute during growth and, due to the inclination of the MBE Transmission electron microscopy -STEM images were obtained with a JEM-2100F system operating at 200 kV equipped with a HAADF detector for Z-contrast imaging conditions. The plan-view TEM specimen preparation was performed by a two-step procedure adapted from the procedure previously reported for NWs longer than 1 µm. 18,42 First, the NWs were mechanically stabilized on the original substrate using a solution of epoxy Gatan G1 (Gatan) in acetone over the sample. Gatan G1 is very effective in filling out the interwire space and is stable under electron-beam irradiation. 42 In the second step, the NW samples were thinned, from the back-side only, using standard methods of grinding, dimpling, and Argon ion milling until electron transparency was reached. This method provides large thin plan-view sections with a large number of NWs (>200), although it is important to note that the observed features are strongly dependent on the QD area probed by the electron beam, i.e. on how the QDs are "cut" during the thinning process, like in any investigation of QD structures using TEM (cf. planar layers). Estimations of the In distribution profile are based on the analysis of the HAADF STEM intensity following the procedure in Ref. [18]. The noise of the annular dark-field detector was determined by averaging the intensity I vac of the vacuum region around the NW, imaged under identical conditions as the NW. The noise I vac was then substracted from all measured intensities including at the NW position. HAADF intensity profiles from the noisecorrected micrographs were determined by integrating across areas with about 10 nm width perpendicular to the QW. The determination of the composition using HAADF is based on the dependence of the contrast R, defined as the HAADF intensity of (In,Ga)As, I InGaAs , divided by that of GaAs, I GaAs on the In content, x. Thickness variations of the TEM specimen led to strongly varying HAADF intensity distributions in the GaAs reference region which hinders the unambiguous determination of I GaAs and do not arise from a change in composition. 18 In order to compensate for variations in specimen thickness, the GaAs intensity profile was fitted by a polynomial function of 5 th order, to estimate GaAs intensity at every single position in the area scan. The measured, noise-corrected intensities were then divided by thickness-corrected I GaAs yielding the measured contrast R, defined as where Z i (i = Ga, As, In) is the atomic number and the factor γ is determined from the analysis of reference planar layers of known composition. In the present case, the In content is estimated assuming γ = 1.7. Indium contents assuming γ = 1.8 are about 5% lower than those obtained using γ = 1.7. These values are within the experimental error of the technique, which is about 15%. Indium contents are obtained with an accuracy of ±2.5%. Cathodoluminescence -CL measurements were performed using a Gatan MonoCL4 system mounted to a Zeiss Ultra55 field-emission scanning electron microscope equipped with a He-cooled sample stage operated at 10 K. The luminescence is collected using a parabolic mirror, passed through a grating monochromator and detected using a liquid N 2 -cooled, infrared-optimized photomultiplier (InP/(In,Ga)As photocathode. Hyperspectral CL linescans were acquired using a Si CCD with a cut-off at about 1060 nm. Photoluminescence -PL measurements were performed in a continuous-flow He cryostat using either a Ti:Sapphire or He:Ne laser for excitation focused onto the sample by either a 10× or 50× microscope objective. The PL signal was collected by the same microscope objective, dispersed with a monochromator equipped with a 750 lines/mm grating and detected with a liquid N 2 -cooled (In,Ga)As detector array with a cut-off at ≈1.6 µm. LED processing -LEDs were processed from a DWELL NW sample grown on a p-type Si wafer following our previous work. 27 The p-type GaAs:Be core of our LED structure provides a direct electrical contact to the substrate. NWs were embedded in benzocyclobutene (BCB) for electrical isolation. The BCB layer was etched down to approximately 1.5 µm by reactive ion etching (RIE) (CF 4 /O 2 ) and the NW tips were contacted with sputtered indium tin oxide (ITO). Devices of different size were patterned in the ITO layer by optical lithography and inductively coupled plasma reactive ion etching (ICP-RIE) (BCl 3 /Cl 2 ). Back-side Ni/Au and top cross-hair Ti/Au contacts were deposited to improve the electrical contact. Finally, processed samples were mounted and wire bonded in chip carriers. Electroluminescence -LEDs were driven under continuous bias and the EL signal was collected by a lens, dispersed with a monochromator equipped with a 150 lines/mm grating, and detected by a liquid N 2 -cooled (In,Ga)As array detector with cutoff at about 1.6 µm. The current-voltage characteristics of the investigated devices was recorded using a semiconductor analyzer.	2019-08-27T11:21:41.000Z	2019-08-27T00:00:00.000	{ "year": 2019, "sha1": "1412308dd897ffc26636a7b79dfc8389b4f51fec", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1908.10134", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "1412308dd897ffc26636a7b79dfc8389b4f51fec", "s2fieldsofstudy": [ "Physics", "Materials Science" ], "extfieldsofstudy": [ "Materials Science", "Physics" ] }
259528894	pes2o/s2orc	v3-fos-license	Effect of mobile based educational program on mother’s knowledge and practice regarding care of children with helicobacter pylori infection Background: One of the most common human pathogens is the spiral and microaerophilic bacteria H. pylori , which also causes stomach adenocarcinoma, peptic ulcers, chronic gastritis, and mucosa-associated lymphoid tissue lymphoma in children. Mobile education based on smartphones is applied in many different areas. Due to its great portability, learning space is increased and learners can work at their own pace. Simulation learning is also possible. Information could be replayed. Objective: This study sought to assess the effect of a mobile-based teaching program on mothers' knowledge and reported practice regarding care of children with helicobacter pylori infection. Methods: The current study utilized a quasi-experimental design. The current investigation has been carried out in the outpatient pediatric clinics of Benha University Hospital as well as Benha Teaching Hospital. From the abovementioned setting, a non-probability purposeful sample of 100 women and their children was selected. Three tools have been utilized: (1) a structured interview questionnaire sheet; (2) a self-reported practice questionnaire sheet; and (3) a medication adherence scale. Results: Prior to the implementation of a mobile-based educational program, the majority of the examined mothers (89%) had an unsatisfactory level of knowledge; however, after the program, the vast majority (98%) had a satisfactory level of knowledge. The majority of the studied mothers (96%) had an inadequate level of practice in pre-program implementation. In contrast, 93% of them had an adequate level of practice post-program. Conclusion: Improvements in mothers' knowledge and practices regarding the care of their children with Helicobacter pylori infections were more effectively with mobile-based education. Again, there had been a highly statistically significant positive link between mothers' overall knowledge level, overall practice level, and overall medication adherence level before and after the mobile-based program's implementation. For mothers of children with Helicobacter pylori infection, healthcare practitioners could employ mobile-based educational methods to promote their children's health. Introduction One of the most common human pathogens is the spiral, microaerophilic bacteria H. pylori, which also causes stomach adenocarcinoma, peptic ulcers, chronic gastritis, and MALT lymphoma in both children as well as in adults. The key factors influencing the frequency of H. pylori infections are socioeconomic position and living circumstances. This infection is typically acquired throughout childhood and can last a lifetime if left untreated (Crowded and poor hygiene). The continuation of bacterial exposure causes gastritis and other serious consequences [1] . The cause of chronic gastritis has been linked to Helicobacter pylori. Nevertheless, extra-digestive impacts on growth indices in young children have been documented. Childhood growth impairment and H. pylori infection have been linked in a variety of studies. One mechanism is that Helicobacter pylori suppresses stomach acid secretion. That may lead to infection with enteropathogens and diarrhea, nutrient malabsorption, decreased food intake due to dyspepsia, and iron-deficiency anemia (IDA) [2] . Roughly 50% of the population is afflicted with Helicobacter pylori. Children around the world are known to be affected, and the incidence varies by country. It can vary geographically within a country and is less in high-income nations (34.7%) than in low-income and middle-income nations (50.8%). The infection is most commonly acquired during childhood and is more frequent in developing countries, where around 70% of children remain afflicted until the age of 15, but it is disappearing in industrialized countries [3] . Person-to-person transmission of H. pylori infection between family members is common, with mother-to-child transmission becoming the most common and oral-oral and oral-faecal transmission routes becoming the most significant. Because the bacterium can persist for a long time in a viable condition in contaminated water, it could also be a cause of infection. Several link diet and socioeconomic factors to H. pylori infection, raising the possibility that these factors play a significant role in the transmission of the illness through foods like milk, meat, and vegetables. Another potential cause of infection is saliva [4] . www.nursingjournal.net Eradication treatment should be administered to children who have a proven H. pylori infection and peptic ulcer. Antibiotics and a proton pump inhibitor (PPI) are used in eradication treatment to improve the antimicrobial efficacy. Given that longer courses are more effective, the recommended length of therapy is 14 days. It is crucial to select the appropriate regimen. Over time, eradication success rates have decreased and frequently fall short of the required 90% threshold. Therapy failure may happen as a result of poor compliance, but rising antibiotic resistance (especially to clarithromycin) is a key motivator. Patients and relatives must receive counselling on the significance of strict adherence to enhance the eradication chances and prevent resistance development [5] . There are several factors can lead to poor medication adherence among H. pylori-infected pediatric patients such as; poor socioeconomic status, age, educational and awareness level of parent, complexity and duration of therapeutic regimen, and improper counseling by health care providers. Nursing staff considered the most responsible personnel for providing health education for infected child and his parent regarding good personal hygiene, proper sanitation, a decrease in the number of close contacts, and importance of medication adherence to improve patients' outcomes and reduce H. pylori intrafamilial transmission [6] . A smartphone can be used anywhere, at any time. It is a mobile device that is highly portable and available. People may now more quickly and easily look for and give information on cultural activities, educational resources, economic activities, and social communication due to smartphones. Mobile learning via smartphones is applied in many different areas. Educational space is enlarged due to its high portability; simulation education is also possible, utilizing a user's location-based data; and self-directed learning could be accomplished, in which learners could practice specific skills and knowledge repeatedly without spatial restraints. Data could be replayed, and learners could work through it at their own pace [7] . Significance of the study One of the most prevalent chronic infections, Helicobacter pylori, affects roughly 50% of people globally. Such an infection is most commonly acquired throughout childhood, particularly in developing nations. The frequency of H. pylori differs greatly between nations; in developing countries, almost half of children are afflicted by the age of ten. Children with symptoms attending the gastroenterology unit's outpatient clinics in Egypt had a high frequency of H. pylori infections (64.6%) [2] . Roughly half of the globe's population is infected with H. pylori, a major gastric pathogen that can lead to a number of disorders of the stomach, such as chronic gastritis, gastric atrophy, gastric MALT lymphoma, peptic and duodenal ulcerations, as well as gastric adenocarcinoma. H. pylori infections are more common in early childhood and are transmitted intrafamilially via both the fecal-oral and oral-oral pathways [8] . There are several factors can lead to poor medication adherence among H. pylori-infected pediatric patients such as; poor socioeconomic status, age, educational and awareness level of parent, complexity and duration of therapeutic regimen, and improper counseling by health care providers [6] . Therefore providing health education for infected child and his mother regarding good personal hygiene, proper sanitation, a decrease in the number of close contacts, and importance of medication adherence is very important to improve pediatric patients' outcomes. Smartphone-based learning may offer a self-directed educational environment where people may access information and practice skills frequently without being limited by time or space. Mobile education based on smartphones is applied in many different areas. Learners may work at their own pace and may replay information. Because of its high portability, the learning space is increased. By employing a user's location-based information, simulation learning is also likely to occur [9] . So, this study aims to improve mother's performance regarding care of children with helicobacter pylori infection through design and implement a mobile based educational program for mothers based on their actual needs. Aim of the study The goal of this research was to enhance mothers' knowledge and practice of caring for children with helicobacter pylori infection. through the following objectives: 1-Assess mothers' knowledge level regarding care of children affected with helicobacter pylori infection. 2-Assess mothers' reported practice level regarding care of children affected with helicobacter pylori infection. 3-Assess mothers' medication adherence level of children affected with helicobacter pylori infection. 4-Design and implement an educational program for mothers using mobile phones that is based on their real requirements. 5-Assess the impact of a mobile-based educational program on mothers' knowledge and practice regarding care of children with helicobacter pylori infection. Research hypothesis 1-Mothers'knowledge regarding care of children with helicobacter pylori infection will be increased after implanting mobile based educational program. 2-Mothers' practice regarding care of children affected with helicobacter pylori infection will be improved after implanting mobile based educational program. 3-Mothers' medication adherence level regarding helicobacter pylori infection will be improved after implanting mobile based educational program. 4-There has been a positive association among mothers' total knowledge, practice, and medication adherence level. Subject and Methods I-Technical design: The study's technical design included the following elements: study design, setting, participants, and data collection tools. Research design: A quasi-experimental design has been used in this study. Research Settings: The present research has been carried out at the pediatric www.nursingjournal.net outpatient clinics at the Benha Teaching Hospital and Benha University Hospital, both of which are connected to the Egyptian Ministry of Health and Population. Both settings' outpatient pediatric clinics were on the ground floor and had two rooms. Both outpatient pediatric clinics provide health care for children. Subjects Two types of samples were used as subjects. Type (1): Following fulfilment of the following inclusion criteria, a nonprobability purposive sampling of 100 moms has been selected from the aforementioned settings who were willing to take part in the research. Tools of data collection The following tools were used to collect data: Tool (I): A sheet for a structured interview questionnaire: It has been created by researchers based on [2,10] to assess mothers' knowledge regarding care of children with H. pylori infection. It has three main parts and has been written in Arabic. Total scoring system The study's mothers' answers have been compared to a model key answer, which was rated as follows: complete correct answers received a score of 2, incomplete correct answers received a score of 1, and wrong or unknown answers received a score of 0. The range of total knowledge scores was 0 to 30. In this regard, mothers' knowledge levels were classified as satisfactory (60-100%; ranged from 18-30 points) or unsatisfactory (< 60%; ranged from 0 < 18 points). Tool II: Self-reported practice questionnaire sheet It was adapted from Lynn [11] and Nguyen et al. [12] to assess mothers' reported practice level regarding care of children with H. pylori infection, modified by the researchers to suit the nature of the study. It included 20 steps grouped under five main parts as practice related to hand washing (3 steps), practice related child hygiene (4 steps), house hygiene practices (2 steps), food and water supply related practices (5 steps) and medication administration related practices (6 steps). Total scoring system The study's moms' answers have been compared to a model key answer that was graded as done had a (1) score and not done had (0) score. The total practice scores varied from 0 to 20 points. In this regard, the level of reported practice by mothers was classified as adequate (70% to 100%) and ranged from 14-20 points, while inadequate (< 70%), ranged from 0 < 14 points. Tool III: Medication adherence scale It was adopted from Culig and Leppée [13] to assess the level of mothers' adherence to helicobacter pylori eradication therapy, it consisted of 10 questions answered by yes or no. Total scoring system Total medication adherence scale scores ranged from 0-10. As a result, the level of medication adherence has been classified as poor (< 50%) with a range of 0 < 5 points, average (50% to <70%) with a range of 5 < 7 points, and good (70-100%) with a range of 7-10 points. II-Operational design The preparatory phase, content validity of the tools used, their reliability, pilot research, and field research were all included in the operational design Preparatory phase This phase involved studying relevant literature and theoretical knowledge of different study aspects; employing books, papers, the internet, periodicals, and journals at the local and international levels to construct the study instruments and become familiar with the different study elements of the study difficulties. Designing mobile based educational program: The mobile based educational program (WhatsApp group) were developed by the researchers and aimed to improve mothers' performance level regarding care of children affected with helicobacter pylori infection. Designing the program had taken period extended from the www.nursingjournal.net beginning to the end of December 2020. The researcher designed program based on the actual need assessment of the study sample. Content validity Tools for data gathering have been translated into Arabic and tested for content validity by three experts (two in pediatric nursing and the other in the area of medicalsurgical nursing specialization from Benha University's faculty of nursing) to assess the tools' clarity, comprehensiveness, significance, accuracy, and simplicity. All of their suggestions have been taken into account, and some have been revised to create the final version of the tools. From their perspective as experts, the tools have been thought to be reliable. Reliability The Cronbach's alpha coefficient test has been used to determine the tools' reliability. This turned to be (α = 0.92) for a structured interview questionnaire sheet, (α = 0.80) for mothers self-reported practice and (α = 0.89) for medication adherence scale. This indicates the favorable internal consistency and high reliability for the study tools. Ethical considerations The Ethics Committee of Benha University's Faculty of Nursing approved the study. The researchers clarified the aim of the study and the expected outcomes to all the studied mothers during the initial interview. Verbal approval was requested to participate in the current study. The studied mothers were assured that all information would be confidential. Additionally, mothers were allowed to leave the study at any moment and without explanation. Pilot study A pilot study has been carried out to assess the clarity and application of the study tools, as well as to determine the time required for every instrument. It has been conducted on 10% of the total individuals, (10) children who had H pyloric infections, and (10) moms who were omitted from the current research to prevent sampling bias and contamination. The last form has been established after modifications have been made in light of the results of the pilot study analysis. From the start to the end of January 2021, this phase lasted one month. Field of Work The mobile based educational program was implemented to achieve the aim of the current study by these phases, assessment, planning, implementation and evaluation phase. The actual work was carried out over period 6 months from the earliest starting point of February 2021 to the end of July 2021. The actual field work was divided into four phases. Assessment phase Mothers were interviewed during the evaluation stage to get baseline data. It took roughly eight weeks, and the researchers were present four days a week (Saturday, Monday, Tuesday, and Thursday) alternately in every study setting from 9.00 AM to extended till 12.30 AM. The average number of interviewed mothers was 3-4 mothers per day. The researchers welcomed every mother, discussed the goal, length, and activities of the research, and obtained written consent at the start of the interview. The data of children affected by H pyloric infection was collected by the researcher and each child required about 15 minutes. The researchers offered the mothers studied a questionnaire to complete in order to evaluate their knowledge, stated practice, and medication compliance, which took 30 minutes. Planning phase The mobile-based educational programme has been built by the researchers relying on baseline data from the evaluation stage and a relevant literature review, as evidenced by mothers' level of comprehension in simple Arabic. Various teaching techniques have been employed, including modified lectures, brainstorming sessions, demonstrations, and group discussions. The appropriate teaching media comprised audio-visual aids, case studies and manikins to help proper understanding of the content by mothers. And it took about 4 weeks. Statement of objectives General objective The main goal of this program is to improve studied mothers' knowledge, practice and medication adherence level regarding care of children affected with helicobacter pylori infection. Specific objectives At the conclusion of this educational program, the mothers studied ought to be able to: Practice related child hygiene  Washing the child's hands before and after eating. www.nursingjournal.net  Washing the child's hands before and after using the toilet.  Keeping the child clean after playing with animals  Avoid sharing childs' personal objects with others. 3. House hygiene practices  Using chlorine regularly to clean the house.  Ventilate the house and allow sunlight to enter daily. Food and water supply related practices  Proper boiling the milk before serving it to the child.  Wash fruits and vegetables properly.  Preventing the child from eating food from street vendors.  Utilize filtered source of water for child uses.  Good cooking of the meats and eggs. Implementation phase This phase took two months beginning from the beginning of March to the end of April 2021 The implementation phase was achieved through using WhatsApp to communicate with the mothers. The researchers had designed a group on the WhatsApp called: improve mother's knowledge and reported practices level regarding care of children with helicobacter pylori infection and available at: https://chat.whatsapp.com/HGu9R78IYnCHkLtvx4Ctr8. The researchers added all mothers in (n= 100), viewed all theoretical and practical sessions supported with pictures, videos, and more discussion to facilitate the process of education and to be applicable for all mothers at any time. The researchers had communicated with the mothers, answered all the questions, and clarify any misunderstanding at any time online. The implementation phase was achieved through sessions; every session began with a summary of the prior session and the goals of the new one. However, the researchers took into consideration the use of the Arabic language that suited all mothers' educational levels. Moreover, the researchers utilized motivation and reinforcement during sessions to improve sharing in the study. The total number of sessions was six, distributed as follows: There were three sessions for the theoretical part; every session lasted 60 mins, and there were three sessions for the practical part; each session lasted 60 mins. The practical part included the following The first session of the practical part included: Hand washing related practices and practice related child hygiene. The second session of the practical part included: House hygiene practices and food preparation practices The third session of the practical part included: Medication administration practices. Evaluation phase This phase took four weeks starting from the end of December 2021 to the end of January 2022. After implementation of the mobile based program, the researchers designed an electronic questioner contains the same pretest tools to assess the impact of mobile based program on the mother's performance level regarding care of children affected with helicobacter pylori infection. And send link to the mothers on the WhatsApp group. Available at: https://forms. office.com/Pages/ResponsePage.aspx?id=IHdfGoEwLkCcy pctfhXITlr83YTjJeVJprbsUp1_iaFUMzZJMkVLSDRTTzZ RNEkzQUVBTkVHRDk0RC4u. Administrative design Following a brief description of the study's goal and anticipated results, official letters from the dean of the nursing faculty at Benha University were sent to the directors of the previously listed settings to request their approval to gather the required data for the present research. Statistical design Statistical Package for Social Science (SPSS) version 21 for Windows was used to organize, tabulate, and statistically analyze the collected data on an IBM compatible computer. For qualitative variables, data has been represented as numbers and percentages, while quantitative variables have been represented as mean and standard deviation. Qualitative variables have been compared employing Chisquare test. A Fisher exact test has been employed instead when the predicted values are <5. Pearson correlation analysis has been employed to evaluate the interrelation of quantitative variables. P-value <0.05 has been considered statistically significant, and P≤0.001 has been considered highly statistically significant. www.nursingjournal.net Table (1) revealed that, (38%) of mothers were in the age group 30<35 years with mean age 33.41±5.66 years, more than half (64%) of them had secondary school and more than two fifth 42% of them were housewife. Regarding their residence; more than three quarters (76%) of them were from rural area, more than two thirds (68%) didn't had pure water supply and (72%) of mothers had high crowding index. Also showed that more than three quarters (79%) of mothers eating from street vendors. Table (3) clarified that, there was highly statistically significant improvement between pre-program and post program regarding mothers' knowledge about H. pylori infection (P ≤ 0.000). There was high mean scores of knowledge at post program 27.07±2.02 compared to preprogram phase 11.31±3.33. Table 4: Correlation between mothers' total knowledge level, total practice and total medication adherence level pre and post -program implementation (n=100) Pearson correlation coefficient Pre-program implementation Post-program implementation R P-value R P-value Knowledge -practice 0.933 0.000 0.978 0.000 Knowledge -medication adherence level 0.883 0.000 0.971 0.000 Practice -medication adherence level 0.862 0.000 0.945 0.000 Correlation is significant at the 0.01 level (2-tailed) ** Table (4) illustrated that, there was a highly statistical significant positive correlation between mothers' total knowledge level, total practice level and total medication adherence level pre and post-program implementation (p<0.000). Table ( 5) Reflected that, there is no statistical significant relation between total score of mothers' knowledge and their age, residence, pure water supply, eating from street vendors, crowding index and sources of information about H. pylori infection pre educational program implementation (p>0.05). However, there is a highly statistically significant relation between total score of mothers' knowledges and their educational level pre-educational program implementation (p<0.000). Moreover, there is a highly statistically significant relation between total score of mothers' knowledges and their occupation, residence, crowding index and sources of mother's information about H. pylori infection post-educational program implementation (p<0.000). Table ( 6) Revealed that, there is no statistical significant relation between total score of mothers' practice and their age, residence, pure water supply, eating from street vendors, crowding index and sources of information about H. pylori infection pre and post educational program implementation (p>0.05). However, there is a highly statistically significant relation between total score of mothers' practice and their previous experience about H. pylori infection pre-educational program implementation (p<0.000). Discussion One of the most prevalent chronic infections is Helicobacter pylori (H. pylori), which affects roughly 50% of people globally. Such an infection is most commonly acquired throughout early childhood, particularly in developing nations. Helicobacter pylori was identified as the cause of chronic gastritis. Nevertheless, extra-digestive influences on growth parameters in young children were documented. Numerous studies have linked H. pylori infection to an impairment in children's growth [2] . www.nursingjournal.net In terms of mothers' characteristics, the present study found that more than a third of the studied mothers belonged in the age group of 30<35 years, with an average age of 33.41±5.66 years. This conclusion was in the same context as Agossou et al. 2020 [14] who performed their research on H. pylori Infection (Hp) among children in the Northern Benin in 2018 and stated that the mean ages of the parents of children with H. pylori infection were 32.80±6.62 years for mothers and 40.82±8.48 years for fathers. Regarding the educational level of the mothers examined, the current research found that less than two-thirds of them possessed a secondary school certificate. This result is supported by Salih et al. 2017 [15] who conducted a study about the incidence of H. pylori among Sudanese children hospitalized at a specialist children's hospital and reported that 49% of the children's mothers had a high school certificate. On the other hand, this finding is contraindicated with Galal 2019 [2] , who conducted a study about H. pylori infection in symptomatic Egyptian children: incidence, risk factors, and growth effects and found that more than half of the mothers (61%) were not educated and showed that illiteracy of mothers was significantly associated with H. pylori. Regarding theoccupation of studied mothers, the present research clarified that less than half of mothers are housewives. This study's results were parallel with those of Spotts et al. 2020 [16] , who investigated concurrent intestinal parasite and H pylori infections in school-aged children in Central Ethiopia and found that (45%) of the mothers of the children studied were housewives. The current research findings about the mothers' residences revealed that more than two-thirds of them came from rural areas. These findings are supported by several prior research findings by Galal et al. 2019 [2] and Gravina et al. 2016 [17] who noted that the H. pylori prevalence rate is high in rural areas. This may be due to increase household crowding index, unsafe water supply and lack of nutrition education in rural area. The finding of the current study was reported that more two thirds didn't had pure water supply and had high crowding index. Researchers viewed that this fact more precipitates to H. pylori infection though enhancing intrafamilial transmission by using and sharing the same personal objects This conclusion was in the same context as the findings of Aitila et al. 2019 [18] , who conducted research on the incidence and risk factors of H. pylori among children aged 1 to 15 years in Mbarara, South Western Uganda, at Holy Innocents Children's Hospital, and found that the rate of infection had been higher in children who used to have an unsafe supply of water at home, no sanitary facility at home, overcrowded families, or an unsafe origin of drinkable water (p<0.001). This result is consistent with the findings of Mohammed et al. [10] , who did research on educational interventions to improve drug adherence, awareness, and practise concerning intrafamilial transmission in patients with H. pylori and found that more than half of both groups had a high crowding index (≥3 members per room). Also, it was supported by several other studies who reported that household crowding index was found to be significantly related to positive infection of H. pylori [2,19] . According to the current research, more than 75% of mothers were found to be eating from street vendors. This study's results were supported by Monno et al. 2019 [20] , who did research on H. pylori infection: connection with food habits and socioeconomic circumstances and found that eating food from street sellers and having meals away from home were strongly related to infection with H. pylori. This result was in harmony with research by Ahmed et al. 2014 [21] on the transmission of H. pylori from food cooked and consumed under sanitary and unsanitary conditions: a preliminary study employing biopsy samples found that individuals who bought and consumed food from street vendors on a daily basis had a prevalence of 90.9%, whereas those who consumed the food products once a week or once a month had a lower incidence. Those who never purchased food from street sellers had a lower incidence of 42.7%. The current study reported that majority of mothers sources of information regarding H. pylori infection from internet. From the point of view of researchers, given the widespread usage of the internet, it can play a significant role in spreading health knowledge and helping patients and their families get the assistance they need to manage their illness. This finding is supported by Kubb & Foran 2020 [22] , who did research titled "Online Health Information Seeking by Parents for their Children: Systematic Review and Agenda for Further Research." Parents all over the world are frequent users of the internet to find health-related data for their children under a wide range of conditions. This finding was in parallel with the findings of a study by Nicholl et al. [23] titled "Internet Use by Parents of Children with Rare Conditions: Findings from a Study on Parents' Web Information Needs," which reported that parents routinely and habitually utilized social media and the internet to learn about their child's condition. In terms of the characteristics of the children tested, the present research's results showed that half (50%) of children were in the age group of 7-<10 years, with a mean age of 5.32±6.43 years. This result agrees with Aitila et al. 2019 [18] who reported that infection with Helicobacter pylori was most prevalent in children aged 6 to 10. This result is supported by Gala et al. 2019 [2] who found that the median age of the participants was 7 years, with a range of 1 to 15 years. With no significant differences, children over the age of ten had the highest frequency of H. pylori infection (32.9%), whereas those under the age of three had the lowest (13.8%), excluding sex and age. The current study also showed that 55% of the studied children were male. The results of this research agreed with those of Deng et al. [24] , who investigated the antibiotic resistance of H. pylori strains obtained from Southwest China pediatric patients and found that patients who were found to be H. pylori positive (53%) were boys. This current study reported that majority of children had no previous hospitalization related to H. pylori infection. This finding supported by Ibrahiem & Saad 2021 [25] who reported regarding previous infectious diseases among family members that only about 14.7% of family member had a history of H. pylori infection. Regarding mothers' medication adherence, the present research found that more than three-quarters of the studied mothers exhibited poor levels of medication adherence in the pre-program. Meanwhile the majority of them had good level of medication adherence in the post program. In addition, mothers' medication adherence levels for H. pylori infection showed a highly statistically significant improvement between pre-program and post-program (P≤0.000). The view point of researchers that, these findings reflect the effectiveness of the mobile based educational program. These findings were parallel with those of Mohammed et al. [10] , who found that 74% of the study group and 42% of the control group adhered to H pylori eradication therapy, showing a statistically significant difference in P-value (0.001). On assessing children complain, the present study reported that 95%, 88%, 96%, 3% and 10% of children suffer from abdominal pain, anorexia or vomiting, pallor, hematemesis, and melena respectively. This result is supported by a study conducted by Xiaohong et al. [26] , who found that symptoms of the digestive tract in children with H. pylori infection comprise reflux, pain in the abdomen, epigastric pain, vomiting, nausea, hematochezia, and peptic ulcers. Abdominal pain was the most frequent among them, while reflux, nausea, vomiting, as well as hematochezia were the least frequent. This result was in accordance with Galal (2019) [2] , who found that pain in the abdomen, vomiting, hematemesis, as well as pallor were clinical signs and symptoms that were significantly related to a positive H. pylori infection. The current study showed a highly statistically significant difference in total mothers' knowledge level between preand post-program implementation, whereas the majority of mothers had an unsatisfactory level of knowledge preprogram implementation, which indicated that they didn't have knowledge about Helicobacter-pylori. In contrast, the majority of them possessed satisfactory post-program knowledge levels (p<0.000), which indicated the effectiveness of the mobile-based educational program for improving mothers' knowledge levels. The result of this study was in harmony with Ibrahiem & Saad 2021 [25] , who reported that in the pre-test the vast majority of the family members' means±SD in all knowledge items were very low, which indicated that they didn't have any knowledge about Helicobacter-pylori. After implementing the health awareness package, there were highly statistically significant improvements were observed in the family members' mean scores in all tested items of knowledge (P>0.000). On investigating mothers' practices level regarding the care of their children with Helicobacter-pylori the present study revealed that, majority of mothers had inadequate level of practice pre-mobile based program implementation. In contrast, majority of them had adequate level of practice post-program. Furthermore, there was a highly statistically significant difference among studied mothers before and after program implementation in relation to reported practice regarding H. pylori infection (p<0.001). This indicated an improvement in mothers' practice postimplementation of the mobile-based program, which may be due to the efficient application of the mobile-based program. This result was in line with that of El Husseniy et al. [27] , who conducted research on the impact of mobile-based learning compared to booklet-based learning on mothers' knowledge and practice regarding their children with bronchial asthma. They reported that one month after receiving mobile-based learning, the mobile-based learning groups displayed improvements in their overall level of knowledge and practice The current study reported a highly statistically significant positive link among mothers' total knowledge level, total practice level, and total medication adherence level before and after program implementation (p<0.000). The result of this study was in harmony with Ibrahiem & Saad 2021 [25] , who found a statistically significant, highly positive association between the total of knowledge scores and reported practices pre and post receiving a health awareness package (p>0.001). This study agreed with Aravindalochanan et al. [28] , who conduct study about the effect of the health awareness package prevention of school childhood obesity and confirmed in this study that the radical positive significant effect on both knowledge and practice reported levels. It means that family members should be acquainted with adequate knowledge related to the prevention of H. pylori infection to improve their practices It is reflected that there is no statistically significant relation between moms' total knowledge score and their residence, pure water supply, eating from street vendors, crowding index, and sources of information about H. pylori infection pre-educational program implementation (P>0.05). This finding contradicted that of Deeb et al. [29] , who discovered a highly significant difference among seropositive and seronegative patients in terms of family crowding, number of family members, number of rooms and bed sharing, and residence area (P = 0.001), denoting those cases with higher family crowding and those residing in slums had a higher prevalence of infection. The findings of the present research were reported to show that there is a highly statistically significant relation between the total score of mothers' knowledge and their educational level in pre-educational program implementation (p<0.000). The finding of this study was supported by Ibrahiem & Saad 2021 [25] , who found highly statistically significant relationships between family members' socio-demographic characteristics' such as educational level, regarding total scores of knowledge pre and post health awareness package (P= 0.000). The present investigation demonstrated there's no statistically significant relation between the total score of mothers' practice and residence, crowding index, and sources of information about H. pylori infection pre and post educational program implementation (P>0.05). This finding contradicted with Ibrahiem & Saad 2021 [25] who reported that there was statistically significant relationships between family members' educational level and no. of family members with their total scores of reported practices pre and post health awareness package where (p>0.05). As well as a highly statistically significant relationship between occupation and crowding index with total scores of reported practices with (p = 0.000). The present research showed that there was no statistically significant relation between mothers' medication adherence score and their occupation, residence, pure water supply, eating from street vendors, and sources of information about H. pylori infection pre-educational program implementation (P>0.05). However, there is statistically significant relation between mothers' medication adherence score and their age, educational level, and crowding index pre-educational program implementation (p<0.003,0.000,) respectively. www.nursingjournal.net Moreover, there no statistically significant relation between mothers' medication adherence score and their personal data post educational program implementation (P>0.05). This result was in harmony with the findings of O'Connor et al. [30] , who did research on enhancing compliance using H. pylori eradication treatment: how and when. Advances in gastroenterology therapy reported that adherence to the drug regimen among infected patients with H. pylori was improved by education and found that patients who received more knowledge about their disease and the significance of medication compliance achieved significantly higher levels of eradication and compliance. This is also, supported by the study of Shakya et al. 2016 [31] , who found that educational interventions to improve patients' attitudes towards treatment and enhance their knowledge of medication achieved the goal of successful treatment. Conclusion According to the present study's findings, it could be concluded that mobile-based education was effective in improving mothers' performance regarding care of their children with Helicobacter pylori infection. Again, there has been a very statistically significant positive link between mothers' overall knowledge level, overall practice level, and overall medication adherence level before and after the implementation of the mobile-based program (p<0.000). Recommendations These points are recommended based on the current study's findings  For mothers of children with Helicobacter pylori infections, healthcare practitioners can employ mobilebased educational techniques to promote their children's health.  Applying mobile based health education about H. pylori infection on a large sample in different settings to confirm the current results.  Further studies are recommended to investigate the factors influencing the mothers' medication adherence. Declaration of Conflicting Interest The authors declare no conflict of interest. Funding This research received no specific grant from any funding agency in public, commercial or non-profit sectors	2023-07-11T16:32:01.624Z	2022-07-01T00:00:00.000	{ "year": 2022, "sha1": "98a47de9ee28633de6a0a56059d30fe5a35a46f4", "oa_license": null, "oa_url": "https://www.nursingjournal.net/article/view/273/5-1-44", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "72ffbdd6911cdce6e928bfb06c0ffaf813b0066d", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [] }
264815058	pes2o/s2orc	v3-fos-license	Finishing the task of eliminating neglected tropical diseases (NTDs) in WHO South-East Asia Region: promises kept, challenges, and the way forward Summary Globally 20 Neglected tropical diseases (NTDs) are prioritized by World Health Organization (WHO), of which 15 are present in the South-East Asia Region (SEAR) with all 11 countries being affected. As the region bears 54% of the global burden, “Finishing the task of eliminating neglected tropical diseases and other diseases on the verge of elimination” was identified as a regional flagship priority in 2014 with focus on lymphatic filariasis (LF), kala-azar, yaws, trachoma, and leprosy. Intensified efforts have been made to raise and sustain political commitment and momentum among partners innovate tools, interventions and strategies to accelerate elimination, and establish the process and support countries to accelerate and validate achievement of elimination targets. Seven countries have verified or validated for having eliminated at least one NTD since 2016, including yaws, LF and trachoma. Between 2010 and 2020, the number of people requiring interventions against NTDs in the South-East Asia Region reduced by 20%. The priorities in the next decade are to strengthen last-mile efforts to eliminate identified NTDs, sustain it and to use the lessons learnt to eliminate other NTDs. Funding None. Introduction Neglected tropical diseases (NTDs) are a diverse group of disease conditions that usually affect people living without access to adequate sanitation, basic infrastructure, and health services in tropical and sub-tropical regions.NTDs are both a cause and consequence of poverty.NTDs cause physical and psychosocial morbidity and disability, preventing children from attending schools and reducing economic productivity.Globally 20 disease conditions are prioritized as NTDs by WHO, affecting more than one billion people and cost developing economies billions of dollars every year. 1 In the WHO South-East Asia Region (SEAR), 15 of these NTDs are present, with at least one NTD endemic in each of the 11 countries of the Region (Table 1).Home to a quarter of the world's population, the WHO South-East Asia Region bears a substantial burden of NTDs.In 2013, over 700 million people in the Region required interventions against lymphatic filariasis (LF) accounting for 59% of the global estimate. 2Over 72% of the new cases of leprosy reported globally was from the Region. 3In 2014, over 354 million children in the Region required regular deworming against soiltransmitted helminthiases, accounting for 42% of the global burden. 4iven this major contribution of the Region to the global burden of NTDs, the Regional Committee of WHO South-East Asia Region declared "Finishing the task of eliminating neglected tropical diseases (NTDs) and other diseases on the verge of elimination" focusing particularly on lymphatic filariasis, kala-azar, yaws, trachoma, and leprosy as one of the Regional Flagship Priorities in 2014.For successful eradication or elimination of communicable diseases, three principal conditions need to be met: availability of effective interventions and strategies to interrupt transmission of the agent, availability of practical diagnostic tools with sufficient sensitivity and specificity to detect levels of infection that can lead to transmission, and an absence of animal reservoir and amplification in the environment. 5These five diseases were chosen because these more or less satisfied these conditions. Since then, intensified efforts have been made to (i) raise and sustain political commitment and momentum on elimination of these diseases among Member States and donor agencies, (ii) innovate tools, interventions, and strategies to accelerate elimination, and (iii) establish the process and supporting countries to accelerate and validate achievement of elimination targets through global and regional partnerships of Member States, implementing and R&D partners and WHO.The present paper summarizes the major progress and achievements in these priority areas in the last decade, emerging challenges, priorities, and the way forward to finish the task of eliminating NTDs, based on the synthesis of data from annual reports submitted by Member States to WHO on the progress, and proceedings of relevant WHO technical advisory group meetings and peerreviewed publications. Progress on elimination of major NTDs Gaining political commitments at the highest level The elimination of any communicable disease requires sustained political commitment and adequate funding for disease surveillance, prevention and control measures, healthcare infrastructure, and research and development of new treatments and vaccines.However, NTDs, as its name indicates, continue to be neglected in the global and health agenda and therefore continued advocacy was critical to sustain momentum and availability of sufficient resources for finishing the task of eliminating NTDs.To this end, a series of high-level advocacy efforts were made in the Region in the last decade. In September 2014 through the Dhaka declaration, Ministers of Health of countries of the WHO South-East Asia committed to institutionalize a cohesive, comprehensive and integrated approach and build partnership and capacity for controlling and eliminating vector-borne diseases in the Region. 6At its side meeting, a memorandum of understanding (MoU) to cooperate and jointly achieve regional kala-azar elimination, originally signed by Bangladesh, India and Nepal in 2005, was renewed and joined by Bhutan and Thailand, to expand and sustain the high-level commitment for Regional collaboration towards kalaazar elimination. 7In 2017, WHO-SEARO further convened a high-level regional meeting of ministers and high-level delegates from the Region as well as partners on "Keeping the Promise: ending NTDs on time in the South-East Asia Region" which concluded with the "Jakarta Call for Action" on accelerating progress towards eliminating NTDs endemic in the South-East Asia Region. 8uch opportunities encouraged Member States, WHO and partner agencies to intensify advocacy efforts for prioritizing NTDs in the national health agendas and also strengthened regional partnership to coordinate Health Policy and collaborate to eliminate NTDs in the Region.This also led to continuous or increased financial support from international donor agencies, such as Bill and Melinda Gates Foundation (BMGF) for LF and kala-azar elimination, USAID for LF elimination, Sasakawa Health Foundation for leprosy elimination and the Korea International Cooperation Agency (KOICA) for LF elimination in Timor Leste. Innovating tools, interventions, and strategies to accelerate elimination The public health interventions recommended by WHO for elimination of five priority NTDs are shown in Table 2. Continuous efforts to develop better tools and strategies and optimize implementation of such interventions on the ground in the last decade have resulted in introduction of a few game changers that have significantly contributed to the progress of elimination of priority NTDs in the region. The first game changer was Triple Drug Therapy of IDA (ivermectin, diethylcarbamazine (DEC) and albendazole) as preferred treatment regimen for accelerating LF elimination.LF is endemic in 9 out of 11 countries in the Region.The primary strategy for interrupting transmission of LF is annual mass drug administration (MDA) targeting the entire population in endemic areas until the prevalence of infection is reduced to a threshold below which transmission is considered no longer sustainable even without interventions, at which point the continuation of MDA is no longer warranted.Conventionally, two drug combination (DEC/albendazole or ivermectin/albendazole) have been used for MDA which does not always kill adult worms while clearing microfilaria circulating in the blood.Therefore, MDA must be given till the adult worms retain fecundity, which is an average of 5-7 years. 9,10In 2016, a study showed superior parasite killing effects of the triple drug therapy, suggesting permanent sterilization or destruction of adult worms followed in 2017 by WHO recommendation on use of triple drug combination for MDA. 11High coverage with one or two rounds of annual MDA of three drugs should be able to reduce the prevalence of infection below the proposed threshold. 10n June 2018, India, one of the first countries in the world to adopt the IDA strategy, initiated its implementation in five districts, with support from WHO, BMGF and several partners.12 Technical committee was established to help identify the target districts, develop implementation guidelines, develop micro-plans, improve social mobilization, and ensure pharmacovigilance.As a result, all five districts achieved effective treatment coverage of over 80%.12 Since then India continues to scale up IDA.Indonesia adopted IDA in 2021, Nepal in 2022 followed by Myanmar in 2023. The second example is introduction of the rK39 rapid immunochromatographic test (ICT), a field-based rapid diagnostic test, and improved treatment alternatives, including miltefosine, the first effective oral agent for kala-azar, and liposomal amphotericin B, which can be administered as a single dose and boasts an efficacy of over 95%, for accelerating elimination of kala-azar. 13ala-azar is endemic in Bangladesh, India and Nepal, with sporadic cases also reported in Bhutan, Sri Lanka and Thailand. 14In the last decades, the Special Programme for Research and Training in Tropical Diseases (TDR) and WHO have coordinated and financed research for development of new innovative tools to support the kala-azar elimination initiative in the Indian subcontinent. The performance of rK39 ICT for antigen detection of kala-azar demonstrated it to be highly sensitive and specific. 13Pentavalent antimony has been the mainstay for kala-azar treatment for more than 6 decades, despite its toxicity, need for parenteral administration in a healthcare setting, and a long course of therapy. 14evelopment and evaluation of various treatments for kala-azar was supported by TDR and other partners such as DNDi and local research institutes.These led to better treatment alternatives in the last decade, including miltefosine, the first effective oral agent for treatment of kala-azar, and liposomal amphotericin B, which can be given as a single dose and has an efficacy of more than 95%. 14,15In 2012, WHO secured a donation of liposomal amphotericin B from Gilead Sciences.In the same year, Bangladesh adopted single dose liposomal amphotericin B as the first-line treatment for kala-azar, which was soon followed by India and Nepal. 16vailability of these tools were game changers in kalaazar elimination efforts in India sub-continent, by enabling early detection and treatment of cases even in Disease Preventive chemotherapy Intensified disease management Vector control WASH Surveillance and M&E NTDs: neglected tropical diseases; WASH: water, sanitation and hygiene; M&E: monitoring and evaluation.remote communities.This encouraged implementors on the ground to strengthen and innovate active case detection in the last decade.The next gamechanger is single dose rifampicin for post-exposure prophylaxis (SDR-PEP) of leprosy.Leprosy is endemic in 10 Member States of the Region.From the 1980s, a steady decline was observed in the prevalence of leprosy globally, due to the introduction of multidrug therapy (MDT) and shortening of the treatment period.However, the decline has been slow in the past 15 years, indicating the limitation of the current strategy to further reduce transmission of M. leprae (Fig. 1). 17In 2018, a study in Bangladesh showed the effectiveness of SDR-PEP in reducing the risk of leprosy among the treated contacts. 18In the same year, WHO published the Global Leprosy Strategy 2021-2030 "Towards Zero Leprosy", which identified SDR-PEP as a new approach for achieving the goal of interruption of leprosy transmission, and in 2020 WHO published a technical guidance on contact tracing and PEP of leprosy. 19,20ince then, India, Indonesia and Maldives have been rolling out SDR-PEP as part of the national leprosy programs and other countries in the Region are preparing to do so. Achievement of elimination targets and their validation In 2015, the WHO Strategic and Technical Advisory Group for Neglected Tropical Diseases endorsed standardized processes for confirming and acknowledging success for NTDs targeted for eradication, elimination of transmission, or elimination as a public health problem (Table 3).Subsequently, WHO published the standard operating procedures (SOPs) for validation of elimination of LF and trachoma as a public health problem. 24,25WHO-SEARO also developed the process of validation of elimination of kala-azar as a public health problem in South-East Asia in 2016. 26ince the launch of the Global Program to Eliminate Lymphatic Filariasis in 2000, 68% of LF-endemic districts across nine endemic countries in the Region have achieved the target thresholds as seen in transmission assessment surveys (TAS).As a result, the population requiring MDA has come down from 883 million in 2013 to 519 million in 2021 (Fig. 2).The number of people requiring MDA has been stable since 2017 because the districts with persistent transmission have rolled out IDA in a phased manner and also India has changed its implementation units from districts to blocks (sub-districts) and intensified LF elimination efforts.As per the SOP, Maldives and Sri Lanka compiled the dossier to claim achievement of elimination of LF as a public health problem and submitted to WHO, which has been validated by SEARO's Regional Dossier Review Group in 2016. 27,28In 2017, Thailand too was validated for having eliminated LF as a public health problem. 29,30angladesh stopped MDA nationwide in 2016, passed the final TAS as post-MDA surveillance in 2021 and submitted a national dossier to WHO for validation of elimination of LF as a public health problem in December 2022.After a refinement of the dossier as per the guidance from the Regional Dossier Review Group, WHO validated Bangladesh's claim for having eliminated LF as a public health problem in April 2023.While India, Indonesia, Myanmar and Nepal are progressively expanding IDA, Timor Leste met the criteria to stop MDA through TAS nationwide and moved into the post-MDA surveillance phase in 2021. Trachoma was known to be a public health problem in three countries in the South-East Asia Region, namely India, Myanmar, and Nepal.In 2018, Nepal became the first country in the Region to eliminate trachoma as a public health problem. 31In 2020, Myanmar joined Nepal Elimination (interruption of transmission) Leprosy Global There is no more local transmission of M. leprae, evidenced by zero new autochthonous cases among children (less than 15 years old) for at least five years. 22imination as a public health problem Kala-azar Regional i) Annual incidence of kala-azar at the district or sub-district level in all endemic countries in the SE Asia Region: less than one per 10,000 population; and ii) Case-fatality rate due to primary visceral leishmaniasis (VL): less than 1%. 14mphatic filariasis Global Prevalence of infection with Wuchereria bancrofti, Brugia malayi or Brugia timori less than target thresholds in all endemic areas. 23achoma Global Reduction in the prevalence of trachomatous trichiasis "unknown to the health system" to less than 0.2% in adults aged 15 years and older; a reduction in the prevalence of the active trachoma sign "trachomatous inflammation-follicular" (TF) in children aged 1-9 years to less than 5% (sustained for at least two years in the absence of intervening antibiotic mass drug administration); and the presence of a system to detect and manage incident cases of trachomatous trichiasis, with evidence of appropriate support for that system. 24ble 3: Operational criteria/definition for global or regional disease-specific eradication, elimination and control of NTDs under the Regional Flagship Priority in the South-East Asia Region. in eliminating trachoma, despite challenges posed by the onset of the COVID-19 pandemic.India remains the only country in the Region yet to eliminate trachoma.The country, however, is carrying out a nationwide prevalidation survey and, despite the interruption of the prevalidation survey during the COVID-19 pandemic, aims to submit the dossier to WHO for validation of elimination of trachoma as a public health problem in 2023. The new cases of kala-azar reported in the Region have reduced by 96.8% between 2011 and 2022 (Fig. 3). By the end of 2022, 772 (99%) implementation units (IU) in the Indian subcontinent (i.e.sub-districts or upazilas in Bangladesh, blocks in India and districts in Nepal) achieved the elimination target and only 3 IUs remained above the elimination threshold. 13Specifically, the elimination target for kala-azar was reportedly achieved in all endemic upazilas of Bangladesh, in 99% of the blocks in India and in 95% of the endemic districts in Nepal. 13Bangladesh has sustained the target of less than one kala-azar case per 10,000 population in all the IUs since 2017 and initiated development of a dossier to stake claim for elimination of kala-azar as a public health problem in 2023, the first country in the world to reach this stage.India was the first country to be verified for yawsfree status in the world.In 2006, India declared yaws elimination three years after the last case was detected in 2003. 32,33While India sustained post-zero yaws surveillance, they also conducted serological surveys in the formerly endemic districts in 2009-2011, followed by compilation and submission of a dossier for WHO certification.In 2015, WHO convened an International Verification Team (IVT) to verify yaws-free status in India. 32In 2016, WHO certified India as "yaws-free". In 2021, Timor Leste conducted an innovative integrated NTD survey, composed of LF transmission assessment survey, screening and serological confirmation of yaws, screening of scabies and stool examination for assessing the prevalence of soil-transmitted helminthiases and taeniasis among over 11,000 school children nationwide.As a result, Timor Leste passed the critical threshold and stopped MDA for LF nationwide.Timor Leste is planning to conduct the next transmission assessment survey to assess whether the country is eligible for validation of elimination of LF as a public health problem in 2023.The integrated NTD survey in 2021 also found no serologically positive yaws cases among over 12,000 children, and another serological survey to verify yaws-free status is due in Timor Leste. For leprosy too, WHO developed a framework that defines criteria including epidemiological cut-offs for verification of interruption of transmission and elimination of leprosy and also developed the Leprosy Elimination Monitoring Tool to help national programme visualize and assess the progress in leprosy elimination at any sub-national level. 22In 2020, Maldives became the first country to consolidate a Framework for Zero Leprosy in the Maldives with the aim of "100 Leprosy Free Islands by 2023".There is a need to demonstrate success of leprosy elimination in low-burden countries such as Maldives and Bhutan to motivate other countries in the Region and beyond. Emerging challenges, opportunities and priorities in the South-East Asia Region With unprecedented achievements in the last decade, the NTD landscape in the South-East Asia Region is changing fast and new challenges and opportunities are being seen.A new global roadmap for eradication, elimination and control of NTDs-Ending the neglect to attain the Sustainable Development Goals: A roadmap for neglected tropical diseases 2021-2030-was endorsed by the Seventy-third World Health Assembly in 2020. 34here is a need for a new vision and direction to accelerate the control and elimination of NTDs and to sustain gains in the South-East Asia Region in the next decade. Optimizing ongoing interventions and addressing social and environmental determinants in the last mile of elimination of NTDs through innovation After a decade of efforts on kala-azar elimination in the South-East Asia Region, the last one percent of the implementation units in the Indian subcontinent is left to achieve the target threshold for elimination as a public health problem.However, despite substantial progress, there remain a series of gaps and challenges.This includes post kala-azar dermal leishmaniasis (PKDL) and kala-azar-HIV co-infection cases that are recognized as important disease reservoir of the parasite but in which the current rK39 RDT has limited use; PKDL cases require long treatment duration, which leads to poor treatment completion and diagnosis and treatment of kala-azar-HIV co-infection require specialized capacity at tertiary health-care facilities. 14Despite a substantial reduction in the overall incidence in the Region, kala-azar outbreak continues to be reported.In Nepal, the disease endemicity is geographically expanding to the hilly and mountainous districts that were considered formerly non-endemic. 35Intensified efforts are needed to strengthen surveillance, enhance active case detection, complete case management, outbreak investigation and vector control.Regional intensification of efforts for LF elimination has led to substantial achievements but also to programmatic fatigue in many parts of the endemic countries.The presence of a substantial number of individuals who have never been treated during the decade of annual MDA is proof of this. 36These individuals act as a reservoir of transmission in areas where LF transmission continues despite many rounds of annual MDA with reportedly high coverage in India, Indonesia and Nepal. 37,38Identifying and addressing such "never treated" population is a priority to accelerate LF elimination in the coming years.Additionally, a potential contribution of zoonotic transmission to persistent transmission of Brugia species in some countries is an emerging concern, which might need to be addressed through One Health approach. Despite gradual overall decline in leprosy burden in the Region, proportions of new leprosy cases and those with Grade 2 disability in many countries remain relatively high indicating late case detection and ongoing transmission.There is a critical need to strengthen regional partnership to enhance cross-learning and to revitalize the national leprosy elimination efforts across the Region, and also to facilitate adoption of newer strategies to enhance early case detection and response actions rather than continuing business-as-usual.SDR-PEP strategy is considered as a game changer but so far, adoption of this new strategy in the Region is slow mainly because of the lack of awareness and local evidence about the benefits of SDR-PEP as a preventive measure and resource constraints for procuring single dose rifampicin and rolling out this new strategy. In intensification of ongoing efforts, it is imperative to secure unwavering political dedication and astute leadership by continuous engagement of Heads of States through national and international forums for advocacy and joint monitoring of elimination progress.This commitment should extend its reach to the grassroots level of governance, facilitated by a well-structured subnational framework designed for the elimination of NTDs.To achieve this, active involvement of communities and civil society is also paramount.Moreover, a comprehensive approach is necessary, one that delves into the environmental and social factors influencing NTD transmission.Inclusion of relevant stakeholders both within and beyond the health sector is essential to this endeavor.The persistence of NTDs can be attributed to enduring behavioral, environmental, and social risk elements, such as inadequate health-seeking behavior, substandard housing, and insufficient environmental hygiene.The effectiveness of public health interventions hinges on a populace that comprehends the urgency, clamors for services, and personally drives the changes required to interrupt disease transmission.A holistic societal approach is vital, underpinned by impactful social and behavioral change communication strategies and the empowerment of communities.This comprehensive approach stands as a pragmatic, sustainable, and cost-effective solution to tackle the social determinants of health that expedite and sustain the battle against NTDs, fostering their eradication, elimination, and control. Strengthening and sustaining essential NTD interventions and services in the elimination phase through integration and cross-cutting approaches The last mile of disease elimination can be particularly resource-intensive, demanding intensified efforts and increased allocation of funding to overcome the remaining hurdles.Without a continuous influx of resources, the progress made thus far can be jeopardized, and hard-won gains may be lost.Therefore, prioritizing the enhancement and sustainability of resources in the face of rising costs during the concluding phase of disease elimination is of paramount importance to ensure the success of these multifaceted endeavors.Furthermore, in the pursuit of efficient resource allocation, it is imperative to explore integrated elimination strategies that encompass multiple diseases. 39By harmonizing efforts and leveraging shared resources, such integrated approaches can lead to substantial cost savings, optimize impact, and enhance the overall effectiveness of disease control and elimination initiatives.This approach not only maximizes the utilization of available resources but also fosters synergies among various health programs, ultimately accelerating progress towards achieving comprehensive public health goals. In counties that achieved elimination of a NTD as a public health problem, efforts are also needed to sustain the elimination status in the post-elimination phase, with a focus on integration of surveillance and response with other disease programme and health system, while accelerating R&D of new tools and strategies to make further progress. 34ala-azar, LF and trachoma are presently targeted for elimination as a public health problem, as there are no appropriate tools to achieve and/or verify interruption of transmission.Therefore, a system for continued surveillance, case finding, outbreak response and targeted response after validation of such status needs to be established.However, there is a high chance that elimination as a public health problem gets mistaken for elimination of transmission and both donor fatigue and program complacency may shift attention and investment to the next unfinished agenda. 13Therefore, there is a critical need to continue the investment in optimizing, integrating and strengthening post-validation interventions within the primary health care and health system for sustainability, and in accelerating research and development of new tools and strategies to further progress towards interruption of transmission. For any NTDs, at present, there is limited evidence for WHO to recommend any specific post-validation surveillance strategy.It is, however, clear that postvalidation strategy will need to be country-specific and feasible to maintain government and donor commitment, and thus integrated with other existing platforms and mainstreamed in the health system functions.There is an urgent need for countries, with support of research and implementation partners, to generate the evidence needed to determine feasible, cost-effective and sustainable post-validation surveillance options for NTD and build necessary diagnostic, entomological and analytical capacity to sustain implementation of such options. Expanding focus on controlling all NTDs of public health importance in the region While Member States in the South-East Asian Region rapidly progresses in elimination of kala-azar, LF, trachoma and yaws as per the Regional Flagship priority, WHO-SEARO and its Member States are progressively expanding its focus to the next unfinished agenda in control and elimination of other NTDs such as schistosomiasis, rabies, snakebite envenoming, dengue and other arboviral diseases and neglected parasitic zoonoses. The Region is one among the most affected by snakebite envenoming, with nearly 70% of annual global snakebite deaths occurring in South Asia alone.In 2022, WHO SEARO launched the Regional Action Plan for prevention and control of snake-bite envenoming in the South-East Asia Region 2022-2030. 40It aims to reduce the number of deaths and cases of disability associated with snakebite envenoming by 50% in the South-East Asia Region by 2030 through community engagement and empowerment to prevent snakebites and provide effective first-aid, health system strengthening to ensure access to life-saving treatment and care, and coordinated technical support to improve availability of quality, effective, safe and affordable antivenoms. The South-East Asia Region also bears the highest burden of dog-mediated human rabies in the world.More than 1.4 billion people in the Region are at risk of rabies infection, and approximately 45% of worldwide rabies deaths occur in Asia.In 2019, World Health Organization, Food and Agriculture Organization of the United Nations (FAO), World Organization for Animal Health (OIE) and Global Alliance for Rabies Control (GARC) jointly launched Zero by 30: the global strategic plan to end human deaths from dog-mediated rabies by 2030. 41This target is specifically included in WHO's new global NTD roadmap 2021-2030.To accelerate elimination of rabies in the Region, WHO-SEARO established the Regional Technical Advisory Group on dog-mediated human rabies in 2023 to guide WHO and Member States in providing evidence-based recommendations on strategies to accelerate the progress in elimination of dog-mediated human rabies in the Region towards the 2030 global elimination targets. Dengue fever has emerged as the world's most common and rapidly spreading vector-borne disease.Except for the Democratic People's Republic of Korea, around 1.3 billion people from the WHO South-East Asia Region live in dengue-endemic areas. 42The Region accounts for more than half of the worldwide dengue burden, and the presence of all four serotypes has made these nations hyperendemic.To review the Regional dengue situation regularly and advise WHO and Member States on evidence-based strategies to accelerate prevention and control of dengue and other arboviral diseases in the Region, WHO-SEARO reestablished the Regional Technical Advisory Group (RTAG) on Dengue and other arboviruses in 2021. 42HO also launched the Global Arboviral Initiative (GAI) in 2022 as an integrated strategic plan focusing on monitoring risk, pandemic prevention, preparedness, detection and response, and building a coalition of partners to strengthen the coordination, communication, capacity-building, research, preparedness, and response necessary to mitigate the growing risk of epidemics due to these diseases.Guided by the GAI, WHO-SEARO plans to lead development of a new Regional Strategic Plan for the prevention and control of dengue in the South-East Asia Region in 2023. Schistosomiasis in the Region is confirmed to be endemic only in a small region in Indonesia -two districts in Central Sulawesi, with about 20,000 populations at-risk.Sustained efforts through the Integrated Schistosomiasis Control Programme implemented in the last 3 decades helped reduce the prevalence of schistosomiasis substantially.Indonesia launched the 2018-2025 National Roadmap for Schistosomiasis Eradication Programme, focused on integrated control programme encompassing mass drug administration, veterinary public health and vector (snail) control with environmental management.Having sustained the low prevalence in the last many years, the country is moving towards interruption of transmission. In 2021, WHO launched a Strategic Framework for integrated control and management of skin-related NTDs, advocating for an integrated approach for diagnosis, control and management of skin NTDs. 34ntegrated approach not only improve efficiency and cost-effectiveness of interventions and service delivery but also reduce stigma and discrimination and improve community and patients acceptance of interventions and services.Skin-related NTDs of public health importance in the Region includes scabies and mycetoma in addition to yaws, leprosy, cutaneous leishmaniasis and PKDL and LF.Scabies is widely prevalent across the Region whereas mycetoma has been reported in India and Thailand, but limited data on true burden and geographical distribution of these diseases exist. 43,44HO recommends MDA using ivermectin as a public health control strategy against scabies, and the progressive scale up of triple drug therapy MDA using ivermectin for elimination of LF in the Region provides an opportunity to assess and bring ancillary impacts on scabies. 45WHO-SEARO is developing a Regional integrated skin NTD toolkit by regional adaptation of the Global Strategic Framework to work with Member States and partners to scale up integrated skin NTD approach across the Region. Finally, one of the most neglected NTDs are other parasitic zoonoses such as taeniasis/cysticercosis, echinococcosis and foodborne trematodiases.Taenia solium is known to be the cause of over 30% of epilepsy cases through neurocysticercosis in many endemic areas where adequate sanitation is lacking and people and roaming pigs live in close proximity. 46Human infection with Echinococcus granulosus leads to the development of one or more hydatid cysts located most often in the liver and lungs.Echinococcosis is often expensive and complicated to treat and may require extensive surgery and/or prolonged drug therapy.Opisthorchis viverrini, the liver fluke, one of the parasites causing foodborne trematodiasis, is classified as carcinogenic agents as they may cause bile duct cancer (cholangiocarcinoma).All such diseases are reported in some countries in the South-East Asia Region but data on disease endemicity and their public health importance remains limited.Control and prevention of these diseases require intersectoral collaboration among the public health, animal health, food safety and WASH sectors.In 2018, the Regional Tripartite, composed of WHO-SEARO, WHO WPRO, FAO and WOAH in the Asia Pacific jointly organized the Meeting to accelerate prevention and control of neglected foodborne parasitic zoonoses to bring together national focal points from various sectors. 47Following the meeting, the Regional Tripartite closely collaborated and developed a series of resource materials to accelerate disease mapping in a standardized manner and control and prevention through One Health approach targeting human health, animal health, environment and food safety sectors. 48gain, the effective control and ultimate elimination of such diseases necessitate a comprehensive, multisectoral approach that encompasses a range of integrated actions.These encompass robust surveillance mechanisms, facilitating real-time disease monitoring and forecast utilizing disaggregated data, and seamless data sharing across sectors.Additionally, strategies encompassing veterinary public health, an expansion of water supply and sanitation coverage, and the augmentation of vaccine and antisera accessibility are vital components.Emphasizing food safety and implementing vector control interventions alongside impactful social and behavioral change communication strategies are imperative for safeguarding vulnerable and affected populations. However, the significance of holistic multisectoral engagement extends beyond the immediate containment and elimination of NTD transmission.It's essential to recognize that numerous NTDs result in lingering morbidities and disabilities that persist even after elimination targets have been achieved by individual countries.Thus, the provision of public health interventions and services must be fortified through collaboration with other programmes and sectors.This includes reinforcing vector control efforts, bolstering water and sanitation initiatives, promoting health education, and enhancing disability and psychosocial care provisions.Such endeavors must be underpinned by unwavering and sustained multisectoral partnerships. The way forward The WHO South-East Asia Region has a unique opportunity to demonstrate continued success in the control and elimination of NTDs that would fundamentally change the global NTD landscape.Being the Regional Flagship Priority in the South-East Asia Region, Member States have accorded strong commitment and high priority to them within their national public health agenda.The Region holds strong R&D potential based on the size of the population and the wealth of entities advancing R&D in technology and medicines. 49,50Countries in the Region also have relatively higher health system capacity compared to developing nations in many other Regions. A new global roadmap for eradication, elimination and control of NTDs-Ending the neglect to attain the Sustainable Development Goals: A roadmap for neglected tropical diseases 2021-2030-set out updated global targets and milestones to prevent, control, eliminate or eradicate 20 diseases and disease groups as well as cross-cutting targets aligned with the Sustainable Development Goals. 34 NTDs of public health importance in the Region.Most importantly, the Regional Strategic Framework proposes the strategic priorities of action at country level under each of the three strategic pillars of the Roadmap: (i) strengthen country ownership, leadership and stewardship, (ii) accelerate programmatic actions, and (iii) intensify integrated and cross-cutting approaches.These strategic priorities can be accomplished only through partnerships and cooperation among Member States, WHO, academia, industry and public and private institutions.With NTDs being diseases of those who are left behind, their elimination will be one more step towards achieving UHC and creating a more equitable world.Together, we can and we will, finish the task of eliminating NTDs. Contributors AY and SR conceptualized the outline and content of the manuscript; AY reviewed, analyzed and drafted the manuscript including tables and figures; SR, APD, MJ and ZL reviewed and provided comments; AY and SR synthesized and finalized the manuscript.All authors approved the final draft of the manuscript. Declaration of interests We declare no competing interests.AY, ZW, MJ and SR are staff members of WHO.APD is a Vice Chancellor of AIPH University.The authors alone are responsible for the views expressed in this publication, and they do not necessarily represent the decisions, policy, or views of WHO or AIPH University.The country names used do not imply the expression of any opinion whatsoever on the part of WHO or AIPH University concerning the legal status of any country, territory, city, or area, or of its authorities. new, infectious, serologically confirmed indigenous yaws cases for three consecutive years, supported by high coverage of active surveillance.21 Fig. 3 : Fig.3: The reported number of kala-azar cases in the WHO South-East Asia Region, 2010-2022. Table 1 : Neglected tropical diseases prioritized globally and endemic in the WHO South-East Asia Region. Table 2 : Public health interventions recommended by World Health Organization for elimination of priority NTDs of the Region. A Regional Strategic Framework for Sustaining, Accelerating and Innovating to end NTDs in the South-East Asia 2023-2030 is being developed by WHO SEARO in collaboration with Member States and partners by adapting the Roadmap 2030 in the context of the South-East Asia Region.It is intended to guide and coordinate efforts among Member States, WHO and partners in the South-East Asia Region to sustain progress, accelerate actions, and innovate approaches to effectively implement the Roadmap 2030 in the Region, encompassing all 15	2023-11-01T15:26:16.335Z	2023-10-01T00:00:00.000	{ "year": 2023, "sha1": "9dd21147d6c1a4bbbad732e6ce58e2d4f85154d1", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.1016/j.lansea.2023.100302", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "492916df466fa6151bcbfd9591ca18488fb5477c", "s2fieldsofstudy": [ "Environmental Science", "Medicine" ], "extfieldsofstudy": [] }
264935317	pes2o/s2orc	v3-fos-license	Faster Peace via Inclusivity: An Efficient Paradigm to Understand Populations in Conflict Zones United Nations practice shows that inclusivity is vital for mediation to be successful in helping end violent conflict and establish lasting peace. However, current methods for understanding the views and needs of populations during dynamic situations create tension between inclusivity and efficiency. This work introduces a novel paradigm to mitigate such tension. In partnership with collaborators at the United Nations we develop a realtime large-scale synchronous dialogue process (RLSDP) to understand stakeholder populations on an hour timescale. We demonstrate a machine learning model which enables each dialogue cycle to take place on a minute-timescale. We manage a key risk related to machine learning result trustworthiness by computing result confidence from a fast and reliable estimation of posterior variance. Lastly, we highlight a constellation of risks stemming from this new paradigm and suggest policies to mitigate them. Introduction Violent conflict leads to deaths, poor health, degraded education, inequality, and economic loss.Since 2011, conflicts worldwide have killed up to 100,000 people a year [1], with indirect deaths estimated to be 3-15 times greater [2].By 2030, over half of the world's poor is projected to be living in countries affected by high levels of violence [1].Migration, malnutrition, destroyed infrastructure, and distressed environments due to conflict lead to poor health [3], increased infant mortality [4], and decreases in the quality of childhood education [5].Conflict tends to disproportionately impact those with lower socio-economic status, leading to increased economic inequality [6].And, the overall economic losses due to conflict have doubled over the last decade to an estimated $1.02 trillion per year ($3 billion per day) [7]. While ceasefires are helpful in preventing escalation once a conflict has begun, long-term peace often requires a successful dialogue and mediation process between the parties.Consequently, over $27b is spent annually on peacebuilding efforts [7].Article 33 of the Charter of the United Nations stresses that "the parties to any dispute, the continuance of which is likely to endanger the maintenance of international peace and security, shall, first of all, seek a solution by negotiation, enquiry, mediation, conciliation, arbitration, judicial settlement" [8].UN practice shows that for a mediation and dialogue process to be successful, inclusivity is vital.The UN defines "inclusivity" as the extent and manner in which the views and needs of conflict parties and other stakeholders are represented and integrated into the process and outcome of a [conflict] mediation effort [9].Given the large number of stakeholders in most conflicts, mediators are often hard-pressed for ways to poll the positions, needs, and interests of stakeholder populations.Further, the reality that the positions of these populations tend to shift, as the dynamics of a conflict or dialogue process evolve, presents a unique challenge. Consequently, mediators have to grapple with potential tensions between inclusivity and efficiency [10].Additionally, while machine learning has the potential to increase efficiency, it is often viewed as too risky for high-stakes decisions [11], suffering from a lack of result trustworthiness [12].As a means to decreasing these tensions, in partnership with collaborators at the UN, we develop a realtime large-scale synchronous dialogue process (RLSDP) enabled by machine learning and manage risk related to result trustworthiness by computing result confidence from a reliable estimation of posterior variance.Further, we highlight risks stemming from this new paradigm and suggest policies to mitigate them. Methods Dialgoue process.Prior methods to gather the views and needs of stakeholder populations for inclusivity include asynchronous dialogue platforms, surveys, focus groups, and social media analysis [10].However, these methods manifest a tradeoff between conversational agility and statistical reliability.We aim to bridge this tradeoff by designing an approach with the agile dynamics of live conversation at a statistically relevant scale. We specify a RLSDP to be characterized by a continuous sequence of minute-scale exchanges between dialogue moderators and a large participating population.Each minute-scale cycle of the dialogue process includes four phases: (i) An open ended question is sent by the dialogue moderators.(ii) The participating population responds with natural language answers then votes on other participants' responses.(iii) Results are computed which include the fraction of the participating population predicted to agree with each answer and the confidence of each prediction.(iv) The dialogue moderators review the results and derive learnings which can inform the next dialogue cycle. Model.During phase (ii) of the RLSDP we employ two types of voting exercise: agreement and pair choice.In the first, a participant is shown a response and asked if they agree with it.We denote the the event that participant i reports agreement with response j by a ij , and disagreement by d ij .In the second, the participant is shown two responses and asked which they prefer.We denote the event that participant i reports preference of response j over response k by c ijk .The minute-scale timing of the cycle constrains the total number of exercises per person to be on the order ten.However, the number of participants and thus the number of expected responses is much larger.As a consequence, the first challenge is reminiscent of compressed sensing: given a sparse sampling of exercise data, maximize accuracy in predicting the fraction of participants which agree with each response. As a baseline, we consider the representative agent model of choice where each response has a utility (independent of the participant) m j , so that p(a ij \|m j ) = σ(m j ) = q j [13].We will refer to this as the binomial model of choice.A prominent example is of individual-level choice is probabilistic matrix factorization (PMF) [14], popularized by its success in the famous Netflix prize competition.In this context, choice is modeled at the individual level, at the cost of making strict assumptions on the number of factors involved in decision making.This amounts to an upper bound on the rank of the participant-to-item utility matrix.A closely related approach, matrix completion, drops these assumptions and instead uses the low-rank promoting nuclear norm prior.[15] Letting and denoting the logistic function by σ, yields the likelihood We now define the nuclear norm \|\| • \|\| * .Let X ∈ R d1×d2 be any arbitrary matrix with singular value decomposition X = U ΣV * .Then \|\|X\|\| * := tr(Σ) is the sum of the positive singular values As insig for nuclear norm minimization in low-rank matrix completion, we remark that one may also view \|\| • \|\| * as the L1 norm of the vector of singular values and thus one may roughly translate the sparsity inducing effects of L1 regularization into rank minimizing effects of the nuclear norm. We apply a uniform prior over the nuclear norm ball of radius τ .Hence the posterior is The second challenge is to estimate the posterior variance needed to compute confidence in predictions within the live timescale.Sufficiently simple models such as the binomial model may admit exact computation of posterior variance.In the case of more realistic models of individual choice, it is necessary to make an approximation.The canonical approaches to approximation of posterior variance and other probabalistic quantities of interest are variational inference [16], and Markov chain Monte Carlo methods (MCMC) [17].Recently, stochastic weight averaging (SWA) [18] has emerged as a promising local approximant to the posterior.Here, we compare SWA and MCMC methods to obtain confidence estimates in the context of a RLS dialog.For sampling techniques, it suffices to instead work with the unnormalized posterior p(A, B \| M )1\|\|M \|\| * < τ .The stochastic weight averaging begins from an MLE or MAP estimator, and runs stochastic gradient descent at a high learning rate η = 1.The iterates are taken as a local surrogate for samples from the posterior. Data and Implementation Experiments were conducted in a low risk environment.Data was obtained from a sample of 110 participants solicited through the Amazon Mechanical Turk crowdworking platform.The task description was "participate in a live online conversation about your experience as a crowdworker."Participants were instructed to "write a brief, thoughtful response, and evaluate and compare each others' responses."They were asked five questions [A1].For each question, an average of 136 responses, 1537 agreement exercises, and 1453 pair-choice exercises were collected over the course of 4 minutes. We implemented the SWA model in PyTorch, and used a Hamiltonian Monte Carlo (HMC) model with the same loss function using Tensorflow and the Tensorflow probability library.As a proxy for repeating the data collection with the same questions and varying duration, the data was split fifty times into training sets including between 5% and 95% of the data.We used samples from each model to approximate the posterior standard deviation of the marginal distribution of agreement for each response 1 N i σ(m ij ).We analytically computed posterior standard deviation for the binomial model with an uninformative Beta( 12 , 1 2 ) prior for agreement with each response. Results We compute the accuracy of the model on a holdout set of agreement voting exercise as the total datapoints per participant used to train the model is varied [figure 1(a)].An equal number of agreement and choice tasks was determined to be near optimal [A2] and thus used at training time.Accuracy increases linearly initially and the levels off at around 15 data points per participant to about 70-80% using SWA.We note that the parameters obtained via SWA are significantly better predictors than their HMC counterparts over all ranges, most likely due to their relative proximity to the MLE. Confidence is probed by computing the posterior standard deviation of the marginal distribution of population agreement for each response.We compare our model trained using SWA and HMC with a representative agent baseline over a range of data points per participant [figure 1(b)].Over the entire range explored our model significantly outperforms the baseline for both training methods.At 15 data points per person our model yields a one statndard deviation confidence range of ±1.5%. The average mean absolute error (MAE) between confidence from SWA and HMC is computed for different ranges of data points per person [table 1].At less than 2 data points per person SWA appears to significantly overestimate confidence.As the number of data points per person increase confidence estimates from SWA and HMC converge, with mean absolute error converging to about 0.2% at 15 data points per participant.Over this range, SWA runtimes are observed to be less than 1/100 th as long as those from HMC.We conclude that at 15 data points per person our ML model predicts the fraction of a participating population which agrees with a response with reasonably high confidence and that SWA can be used to reliably estimate confidence in seconds.This translates to feasibly executing phases ii and iii of an RLSDP in about 2 minutes.At this timescale, many cycles of an RLSDP can reliably take place over the course of a one hour dialogue process.Our next step is to pilot an RLSDP in the context of a mediation process, and impose de-risking policies, before deploying it into mediation of active conflicts.If successful deployment of an RLSDP is able to decrease the time to resolve one major conflict by one week, we estimate it could save 230 lives and $380 million in economic loss [A3]. Any deployment of a realtime large-scale synchronous dialogue process (RLSDP) warrants managing at least three categories of risk.The first risk is inaccurate results due to non-representative data.This can result from bias in questions, disengaged participants, or a participating population which is not representative.The latter can be the result of poor sampling or malicious actors.A policy to manage this first risk may include requirements that (a) dialogue moderators be trained in asking unbiased questions, (b) an appropriate population sampling and participant validation scheme be applied, and (c) randomized human verification of general data quality be conducted.The second risk is inaccurate results due to poor performance of a prediction model.This can result from a faulty machine learning model or programming errors.A policy to manage this risk may include the requirement for appropriate model verification to take place on production deployments.The third risk is inaccurate conclusions due to misinterpretation of results.This can happen because the results are interpreted in the absence of proper context or confidence in the results are miscalibrated.A policy to manage this risk may include the requirements that (a) proper context be identified and integrated into the interpretation of results and (b) all ML-based results include estimates of confidence. Figure 1 : Figure 1: Training results as a function of data per participant.Ranges shown for validation accuracy and posterior standard deviation.Means denoted as solid line.(a) Model validation Table 1 : Comparison of runtimes and mean absolute error (MAE) between confidence estimates for SWA and HMC across ranges of data per participant (DPP).	2023-11-03T06:41:17.988Z	2023-11-01T00:00:00.000	{ "year": 2023, "sha1": "f81ed6c56276c36eef5b900528466332b1e88500", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "f81ed6c56276c36eef5b900528466332b1e88500", "s2fieldsofstudy": [ "Political Science", "Computer Science", "Sociology" ], "extfieldsofstudy": [ "Computer Science" ] }
11704811	pes2o/s2orc	v3-fos-license	The effect of thermal processing in oil on the macromolecular integrity and acrylamide formation from starch of three potato cultivars organically fertilized Starches from three organically produced cultivars of potato tuber (Lady Rosetta, Spunta and Voyager) have been studied in relation to (i) acrylamide production (ii) macromolecular integrity after frying with extra virgin olive oil, soybean oil and corn oil. During cultivation, a treatment involving the combination of nitrogen, phosphorus and potassium fertilization under organic farming was applied (N1, P2, K1 where Ν1 = 1.3 g Ν per plant, P2 = 5.2 g P2O5 per plant, Κ1 = 4.0 g K2O per plant).Potatoes fried in olive oil retained the highest glucose concentrations for all cultivars 0.85 ± 0.2 mmol/kg, followed by 0.48 ± 0.2 for those fried in corn oil and 0.40 ± 0.1 mmol/kg for those fried in soybean oil. The highest average fructose concentration was recorded for the samples fried in corn oil as 0.81 ± 0.2, followed by 0.80 ± 0.2 and 0.68 ± 0.3 mmol/kg for the samples fried in olive and soybean oils, respectively. Asparagine was the most abundant free amino acid in the three varieties tested, followed by glutamine and aspartic acid. The mean initial concentration of asparagine in raw potatoes tubers was 42.8 ± 1.6 mmoles kg−1 for Lady *Corresponding author: Theo Varzakas, Department of Food Technology, Technological Educational Institute of Peloponnese, Kalamata, Greece E-mail: theovarzakas@yahoo.gr PUBLIC INTEREST STATEMENT This paper addresses the effect of thermal processing in oil on the macromolecular integrity and acrylamide formation from starch of three potato cultivars organically fertilized. This paper is quite significant since it is a farm to fork approach addressing the effect of organic fertilization and different cultivars used by the Greek food industry on the acrylamide production and macromolecular integrity, sugars and amino acid contents as well as oil profile (extra virgin olive oil, corn oil and soybean oil). It brings new knowledge to the field since it will help the industry to adopt practical, effective and innovative ways to further reduce the levels of acrylamide to help satisfy market and consumer pressures. The predominant effect of reducing sugar concentration on the acrylamide formation in potato products was confirmed here. Finally, it was found that there was no significant effect of the type of frying oil on the levels of acrylamide. Introduction The formation of acrylamide, a neurotoxic compound and possible carcinogen, in heated foodstuffs has been a global health concern to markets and consumers, as well as a matter of investigation by research groups and food safety committees (Halford et al., 2012). Acrylamide is formed naturally in plant-derived and starch-rich foodstuffs such as potato-and grain-based foods that are cooked at high temperatures (Pedreschi, Mariotti, & Granby, 2013). Acrylamide is formed in heated foods, generally starchy, through the process of the Maillard reaction, in which sugars react with the amino acid asparagine, the role of which has been well established over recent years (Koutsidis, De la Fuente, Dimitriou, Kakoulli, & Wedzicha, 2008). However, the relative importance of different sugars and/or carbonyls as reactive species, as well as the conditions employed, may play a crucial role in its formation (Koutsidis et al., 2009). It is suggested that both molecular mobility and sugar reactivity would determine the relative effect of sugars on the acrylamide formation, whereas temperature may also play an important role in determining the relative reactants (Wedzicha, Mottram, Elmore, Koutsidis, & Dodson, 2005). There are uncertainties regarding the link between acrylamide and cancer risk in humans: some results suggest a link between human exposure and cancer, whilst others do not support such a conclusion (Hogervorst, Schouten, Konings, Goldbolun, & Vad den Brandt, 2007). Exposure data, which are required to evaluate the link between acrylamide and cancer, are limited. In 2002, the European Commission began collecting occurrence data on the levels of acrylamide in foods. In February 2005, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) carried out a safety evaluation on acrylamide in food concluding that the issue poses a human health concern (Food and Agricultural Organisation/World Health Organisation, JECFA, 2005). This conclusion was consistent with an opinion published by the Scientific Committee on Food (SCF) in 2002. In co-operation with Member States, the European Food Safety Authority (EFSA) took over responsibility for this task in 2006. National food safety authorities in the EU Member States, academia and food manufacturers have sought to better understand acrylamide and to reduce its levels in foods. Many countries continue to contribute to the growing body of research and data. Workshops on this issue have been organized by EFSA in 2003 and the European Commission jointly with the European food and drink industry association (CIAA) in 2006 (European Food & Drink Industry Association (CIAA)-Food Drink Europe, 2011). CIAA has published an "Acrylamide Toolbox" based on existing knowledge in the food industry, which is regularly updated. Efforts have been made by food manufacturers to modify recipes and processes to reduce acrylamide occurrence in foods such as French fries, snacks and crisps. Acrylamide formation is also interrelated with quality attributes of heated foodstuffs that arise due to the involvement of common Maillard reaction intermediates. The addition of amino acids has been proposed as a mitigation strategy to reduce the levels of acrylamide in crisps, flat breads, and bread crust, while glycine has received particular attention as an additive that could potentially reduce the acrylamide formation by either competing for available Maillard reaction intermediates or reacting with acrylamide itself through Michael addition type reactions (Low et al., 2006). Inevitably, the application of any reduction technique must be a cost-effective solution since it is likely any additional expense will be transferred to the consumers. Numerous pathways that may lead to the formation of acrylamide have been depicted in the literature, for instance the formation of acrolein by the degradation of oxidative lipid, leading to ultimately acrylic acid, which in turn reacts with asparagine to form acrylamide (Yasuhara, Tanaka, Hengel, & Shibamoto, 2003). However, several research groups have confirmed that the main route for acrylamide in food is via the Maillard reaction, which is a complex series of non-enzymatic reactions between amino groups and reducing sugars that often determine the colour, flavour and texture of cooked foods (Halford et al., 2012). In the present study, the effect of organic fertilization on sugar and amino acid contents has been examined with respect to the acrylamide formation in potato tubers of three chipping varieties of potato Lady Rosetta, Spunta and Voyager after frying in either extra virgin olive oil, soybean or corn oil was studied. A key question was which oil is most appropriate for frying, and which frying conditions assure quality and safety of the fried products, particularly potato chips (Andrikopoulos, Kalogeropoulos, Falirea, & Barbagianni, 2002;Lolos, Oreopoulou, & Tzia, 1999;Moreira, Castell-Perez, & Barufet, 1999;Tareke, Rydberg, Karlsson, Eriksson, & Törnqvist, 2002). We consider the effects of different frying vegetable oils on the creation of acrylamide in potato chips and the estimation of reactants and residuals involved in the acrylamide formation as well as oil uptake of potatoes after frying (Willinger, 1964). In addition, the evaluation of the macromolecular integrity of potato amylose and amylopectin is also investigated, in an attempt to correlate their content and configuration with acrylamide formation. The results will help the industry to adopt practical, effective and innovative ways to further reduce the levels of acrylamide to help satisfy market and consumer pressures. Sample materials For the investigation of the influence of oil type on the acrylamide formation, eight samples of potato tuber were prepared and fried in three different types of oils: olive oil, corn oil and soybean oil. Potato varieties have been chosen according to suggestions of the Ministry of Agriculture, Greece and the conditions favouring their production in the area of Messinia and those cultivated for use by the major food industries in Greece. (http://varieties.potato.org.uk). They were collected from local companies like Compo Hellas. Treatments with two different concentrations of nitrogen, phosphorus and potassium have been used plus the control. So for each variety, eight different fertilization treatments have been used. For each treatment, three repetitions have been made with 5 plants each. Experiment was three factorial with two levels for each factor and due to the statistically significant interaction between the factors regression was carried out for each factor separately. The t-test tested the significance at a level of p ≤ 005. Each treatment was applied to five replicates of four plants each and the experiments were carried out at the Technological Educational Institute of Peloponnese (Kalamata, Greece) between March and May 2013. General procedure Tubers were washed and cut into pieces (~1 × 1 × 3 cm) with a French fry-shaped cutter (An industrial peeler (model M591E4, IMC, England) was used. A circular cutting mould was used to provide chips with a diameter of 40 mm. There was no colour change of potato while cutting as confirmed by a Hunter chroma meter. Potatoes were cut and then immediately transferred to the deep frier. Potato slices were rinsed immediately after cutting for 1 min in distilled water to eliminate some starch material adhering to the surface prior to frying. Two litres of oil was weighed and placed in a 5-L domestic deep-fryer (electrical fryer, Greek local company) with a thermostat and preheated. The fryer was equipped with a stirring mechanism to ensure a homogeneous temperature in the oil bath. French fries were weighed and 250 g of each fried at 180 ± 1°C for 8 min. After frying, potatoes remained in the fryer basket mesh, for one min, to remove the excess of surface oil. Then they were placed in pre-weighed filter paper to cool. The oil used in the fryer was refilled after 8 consecutive frying times. Absorbed oil measurement After four consecutive frying times, the oil was allowed to cool and then weighed to calculate the amount of oil absorbed by the potatoes during frying. The results were expressed as g absorbed oil per 100 g raw potatoes. Starch sample preparation and characterization Starch was extracted from raw potato and fried chips using a combined protocol first described by Willinger (1964) and Rosenthal and Espindola (1969). Samples were rinsed with distilled water, peeled, diced and disintegrated in a domestic blender in 1% sodium metabisulphite (Sigma Aldrich, UK, analytical grade) solution to prevent oxidation. The resulting slurry was sifted through a 0.25 mm sieve, rinsed and sieved again. Samples were centrifuged at 6,000 rpm for 10 min to remove excess solvent and degrease the sample. Peletted material was suspended in 0.1 M sodium hydroxide solution (Fisher Scientific, UK, analytical grade), re-centrifuged and repeated with 0.15 M sodium hydroxide solution. The remaining pellet was rinsed with deionized water until at neutral pH and with 70% ethanol (Fisher Scientific, analytical grade). The sample was dried in an oven at 45°C. For hydrodynamic characterization, samples were dissolved in 90% DMSO (Fisher Scientific, analytical grade). Intrinsic viscosity [η] was measured using an Ostwald capillary suspended in a water bath temperature controlled to 25.00 ± 0.01°C. Data were analysed using a linear extrapolation of the Huggins (1942) and Kraemer (1938) equations as reported by Harding (1997): where (η r ) is the relative viscosity (flow time of solution divided by flow time of solvent), (C) is the mass concentration, and (k H ) and (k K ) are the Huggins and Kraemer constants, respectively. The hydrodynamic radius (r H ) was estimated using Dynamic Light Scattering (DLS) from a ZetaSizer NanoZS (Malvern Instruments, UK) and analysed using Zetasizer Software (v6.20, Malvern Instruments, UK). Intensity fluctuations in the solution were measured at 20.0°C, with backscatter intensities from the solutions from a 632 nm laser registered at an angle of 173°. Sample preparation and analytical procedures The fried potatoes were retained at room temperature until equilibration. Twenty g was weighed and 150 ml H 2 O was added. The mixture was homogenized for 2 min and the solution allowed to stand for 10 min before being centrifuged for 15 min at 4,000 rpm. The supernatant solution was filtered on a filter paper with a pore size of 0.45 μm. The collected liquid was stored in glass vials and placed in a −20°C freezer. The titratable acidity (TA) and the reducing sugar concentration (RS) were measured in the filtered solution. TA was determined according to EC guidelines (Commission Regulation (EC) No 1065/97, Commission Regulation (EC) No 2568/91). Reducing sugars were determined by the DNS method (Miller, 1959). Preparation of the extracts from raw and French fries samples The samples were prepared using a modification of the method reported by Halford et al. (2012). Four raw potatoes were washed, peeled and cut into small pieces. Thereafter, two g of each was weighted onto a plastic plate and was put into a falcon tube followed by adding 5 ml of an aqueous methanol 50%. The sample was then blended and homogenized using a DI 25 basic dispersing device (KIKA-WERKE GMBH and Co. KG, Staufen, Germany) for 15 min at low speed at room temperature or until uniform, and was then centrifuged by benchtop Jouan C3i centrifuge (Thermo-Fisher Scientific, USA) at 4,000 rpm for 20 min. The aqueous supernatant was transferred to a screw top bottle (being careful to avoid solid particles) and filtered twice through Minisart filter units (0.45 μm), collecting approximately 4 ml of the filtrate, which was stored at −80°C. Analysis of the extracted samples using LC-MS The prepared samples were analysed by LC-MS using an Agilent 1100 high-performance liquid chromatography (HPLC) instrument (Agilent Technologies, Wokingham Berkshire, UK), with a quadrupole mass spectrometer operated in negative electrospray ionization mode (ESI). An isocratic separation was performed at room temperature using Luna 5μ NH2 100A 250 × 2.0 mm (Phenomenex Inc., Macclesfield, UK). The parameters of the instrument were as follows: the column temperature was 40°C. The mobile phase was ACN 80% in purified water, and the flow rate was 0.7 ml/min. The injection volume was 5 μl. Each extract was analysed in triplicate (technical replicate). The run time was 8 min, and the data were collected from 2 to 8 min. Fructose, glucose and sucrose eluted at around 2.07, 2.48 and 3.31 min, respectively. The mode of MS monitoring was selected ion recording (SIR), MassLynx software was used to obtain data and to transform them into usable results. The chromatogram readings of the gradient mixed standard sugar solutions (10-60 μg/ml) were used to plot a calibration curve for each sugar; glucose, fructose and sucrose, which then were used to determine the unknown concentration corresponding to each of these sugars in a dilution series (50% water/ methanol) of the samples under investigation. Preparation of the extracts from raw and French fried samples Raw potatoes were peeled after washing, cut, then 1 g of each placed in a falcon tube, to which 4 ml of hydrochloric acid (0.01 mol/L) was added to each tube and the sample was mixed for 15 min at low speed at room temperature until it became homogenized using a DI 25 basic dispersing device (KIKA-WERKE GMBH and Co. KG, Staufen, Germany). The homogenate was then centrifuged by benchtop Jouan C3i centrifuge (Thermo-Electron Corporation, USA) at 4,000 rpm for 20 min. The aqueous supernatant was carefully transferred to a screw top bottle (avoiding the solid partials), followed by filtration using Millipore Millex-HN filter units (0.45 μm) twice, collecting approximately 3 ml of the filtrate and stored at −80°C in a freezer. The amino acids in 25 μl of the filtrate were then derivatized using the EZ-Faast amino acid derivatization kit (Phenomenex, Torrance, CA, USA). Twenty-five microlitre of each filtrate was mixed with 100 μl of norvaline (equivalent to 20 nmol of norvaline) which acted as an internal standard (IS). This mixture was passed through the EZ: fast solid phase extraction sorbent (contained within a pipette tip) which was thereafter washed with 200 μl propanol. A solution of propanol and sodium hydroxide (200 μl) was then used to remove the sorbent (and the amino acids retained on it) from the pipette tip. Fifty microlitre of chloroform was added to the solution followed by the addition of 100 μl of isooctane to derivatize the amino acids. The amino acids were recovered in the upper organic layer, this was dried down using Nitrogen Evaporator and the sample was redissolved in in 100 μl isooctane: chloroform (80:20 v/v). The prepared samples were stored at 4°C in a fridge for 24 h prior to analysing by GC-MS (ThermoQuest, Manchester, UK). With the exception of the washing and peeling steps, the same steps of the preparation of extracts from raw potato samples have been followed for preparing extracts of French fries. Preparation of a linear gradient of standards solutions A linear gradient of standards solutions was prepared by mixing 10 μl of amino acid calibration standards solution with 100 μl of the internal standard IS, and then the EZ-Faast amino acid derivatization technique was used to derivatize the amino acids. The same steps were followed to prepare 25 and 50 μl standards. Analysis of extracted samples using GC-MS The GC-MS (ThermoQuest, Manchester, UK) was used to analyse the prepared samples. One microlitre of the sample was injected in the splitless mode (split closed for 10 s) using an AS3000 autosampler (ThermoQuest, Manchester, UK), where three technical replicates were performed for each sample. The injector of the Trace GC Ultra gas chromatograph (ThermoQuest, Manchester, UK) was maintained at 250°C, with an initial oven temperature of 90°C which was increased to 320°C at 20°C/ min (transfer line from the oven to mass spectrometer, 300°C). Helium (8psi) was used as the carrier gas to elute the amino acids from the ZB-AAA amino acid analysis column (10 m × 0.25 mm ID). The DSQ II mass spectrometer (ThermoQuest, Manchester, UK) was run in selected ion mode recording ions 84,101,114,116,130,144,146,155,156,158,172,180,184,243 and 244 with a dwell time of 0.03 s and the running time was 11.50 min. Calibration was achieved by comparison of peak areas for the amino acids in standard and sample runs after adjustment for variation in the peak area of the IS. Data were reported on a dry weight basis. Xcalibur program was used for data acquisition, processing and results delivery, and the data were then exported to excel for the statistical analysis, and the following equation was applied in order to calculate the concentration of amino acids in the sample under investigation: where PA = Peak area; Conc std = nmols added per unit of IS (one unit = 100 μl). Determination of acrylamide Acrylamide (0.5-50 μg/ml) in methanol (10%v/v) and formic acid (0.01-1% v/v) solution were prepared as standards. For sample analysis, 2 g of homogenized potato samples (approx. 0.3 mL) were mixed with a 13 C-acrylamide internal standard solution to make the isotope ratio match 1:1 and extracted with 40 mL water. The samples were analysed by LC-MS by means of an electrospray ionization interface. Sample extracts and the isotope ratio standard solution were injected (5 μL) on to a Symmetry 300 C4 column (150 mm length, 4.6 mm i.e. 5 μm particle size). The analytical separation was performed using an isocratic mixture of 10% methanol in water containing 0.01% formic acid at a flow rate of 0.2 mLmin −1 . The mass spectrometer was operated in selected reaction monitoring (SRM) mode at m/z 72→m/z 55 and m/z 75→m/z 58 for acrylamide and 13 C3-acrylamide, respectively. Statistical analysis T-tests and ANOVA were used to evaluate the significance of the effect of the type of frying oil on the concentration of acrylamide precursors. Means of interest were compared using p-values at level of significance 0.05. Moisture content For the three varieties, the mean moisture contents of raw potatoes were 81.8 ± 0.6% for Spunta, 78.4 ± 1.5% for Lady Rosetta and 81.7 ± 0.9% for Voyager (Figure 1). During frying, a decrease in the moisture content was observed. The final moisture contents were 45.5 ± 1.3% for Spunta, 44.0 ± 2.1% for Lady Rosetta and 46.5 ± 1.0% for Voyager. The percentage decrease in moisture varied from 30 to 65%. The moisture decrease differed significantly (p < 0.05) between soybean oil and olive oil as a mean of the three varieties, but not in any other comparison between frying oils. Absorbed oil in French fries Potatoes fried in olive oil, absorbed significantly (p < 0.05) smaller oil quantities compared to the other two oils, regardless of the potato variety tested. This result agrees with the findings of other studies (Kita & Lisińska, 2005) and is most likely due to the fact that the absorption of olive oil in fried potatoes is mainly localized to the surface of potatoes contrary to other vegetable oils where the absorption is usually diffused in the whole mass of the tubers. On the other hand, there was no statistically significant difference between the other two oils (corn oil and soybean oil). The mean value for olive oil absorbed was 4.5 ± 0.1 g/100 g fried potatoes, for corn oil it was 4.9 ± 0.3 g/100 g fried potatoes and for soybean oil it was 5.4 ± 0.3 g/100 g fried potatoes (Figure 2). Hydrodynamic characterization The polymeric integrity of the starches in response to the processing was assessed by viscometry and dynamic light scattering, and the relevant parameters are summarized in Table 1 and Figure 3. Intrinsic viscosity [η] evaluations were consistent between the Huggins and Kraemer extrapolation methods to zero concentration, and standard errors did not exceed 12%. Raw potato starch samples ranged between 420 and 535 ml/g, a range consistent with previously published intrinsic viscosity values for starch (Millard, Dintzis, Willett, & Klavons, 1997), however no significant difference was found in the extrapolation to infinite dilution (p > 0.1) between potato varieties. Upon cooking, starch extracts significantly reduced in intrinsic viscosity (p < 0.05), in all varieties, ranging from 220 to 360 ml/g. Distributions of the hydrodynamic radius r H were estimated from measurement of the translational diffusion coefficient from dynamic light scattering, DLS (see Harding, Satelle, & Bloomfield, 1992). The polydispersity of all samples was too large for the high-resolution analysis of the distribution, which was expected from an unfractionated preparation of high molecular weight polysaccharide. The z-average values for r H each processed starch are presented in Table 1. Raw starch samples ranged between 180 and 210 nm, again consistent with other starches (Millard, Wolf, Dintzis, & Willett, 1999), with a standard error no more than 6% and with no significant difference between potato varieties (p > 0.14). When fried, the starch significantly increased in radius (p < 0.05) in all varieties. These results are contrary to results found by Roger, Bello-Perez, and Colonna (1999) who found that heating starch using a microwave reduced the hydrodynamic radius of in vitro solutions. The difference observed in these results may be explained by the facts that frying is a fundamentally different heating technique to microwave-heating, and that these starch samples were heated in situ in the cells of potatoes. OLIVE OIL SOYBEAN OIL CORN OIL Although usually an increase in hydrodynamic radius would accompany an increase in the intrinsic viscosity, the opposite correlation was observed between uncooked and cooked starch samples. This would suggest that there has been a conformational change in the starch. Intrinsic viscosity is more sensitive to shape than hydrodynamic radius, thus the starch granules would have swollen with the uptake of water, and become more spherical as a result. Titratable acidity of oils When potatoes were fried in olive oil, the acidity was lower (p < 0.05), compared to corn and soybean oils in spite of the potato variety tested. However, there was no significant difference between the other two oils. The mean value for acidity for olive oil was (0.34 ± 0.10) g/100 g of fried potatoes, (0.45 ± 0.20) g/100 g for corn oil and (0.51 ± 0.10) g/100 g for soybean oil (Figure 4). On the other hand, Voyager variety showed higher acidity (p < 0.05), than the other two varieties (Spunta and Lady Rosetta) regardless of the oil being used. Determination of sugar concentration using LC-MS The means and the standard deviations of the concentrations of the glucose, fructose in the extracts of the French fries samples originating from three potato cultivars and fried in three different types of oils are shown in Figure 5. The mean of glucose concentrations of all the cultivars was greater when fried in olive oil (0.85 ± 0.2 mmol/kg), compared with 0.48 ± 0.2 mmol/kg for those fried in corn oil and 0.401 ± 0.1 mmol/kg for those fried in soybean oil. The highest average fructose concentration was recorded for the samples fried in corn oil as 0.81 ± 0.2, followed by 0.80 ± 0.2 and 0.68 ± 0.3 mmol/kg for the samples fried in olive and soybean oils, respectively. Reducing sugars are consumed in the Maillard reaction between them and the amine groups of the free amino acids Asn and Gln. Of these, the quantities of Asn are largely responsible for the acrylamide formation in French fries (Friedman, 2003;Taeymans et al., 2004). Frying in oil reduces the concentrations of all sugars. Notably the importance of reducing sugar concentration as the primary determinant of the acrylamide formation is well established, with free asparagine concentration contributing to the variance in some varieties/conditions. Determination of free amino acid asparagine concentration using GC-MS Concentrations of all free amino acids that occur naturally in potato cultivars were measured using GC-MS. Free amino acid concentrations were measured in raw and French fry potato chips. As demonstrated previously, asparagine was the most abundant free amino acid in the three varieties tested here, followed by glutamine and aspartic acid (Halford et al., 2012). The mean initial concentration of the key amino acid in the acrylamide formation, asparagine, in raw potatoes tubers was 42.8 ± 1.6 mmol kg −1 for Lady Rosetta, 34.6 ± 1.2 mmoles kg −1 for Spunta and 36.2 ± 2.0 mmol kg −1 for Voyager. Lady Rosetta contained a significantly higher concentration of asparagine compared to the other two varieties (p < 0.05). The residual asparagine contents of French fries were measured after frying in various oils. The asparagine contents of the French fries are not uniformly affected by the choice of oil but notably in these experiments corn oil differentially reduced the asparagine content of Lady Rosetta compared to other varieties in French fries. The ASN concentrations in the tested samples from the three potato varieties fried in three types of oils are graphically presented in Figure 6. Acrylamide formation Acrylamide was undetectable in raw potatoes (<20 μg/kg). However, acrylamide formation was detected in all oil fried samples. The greatest quantity of acrylamide was observed in French fries Voyager L ady Rosetta S punta ACIDITY (g oleic acid / 100 g fried potatoes) OLIVE OIL SOYBEAN OIL CORN OIL derived from the potato variety Lady Rosetta, followed by Spunta and Voyager. There is a significant reduction in the formation of acrylamide in the variety Voyager compared to the others (p < 0.05). Voyager potatoes contained the least quantity of the reducing sugar glucose and exhibited the smallest reduction upon frying. Glucose would be a substrate for the Maillard reaction leading to the acrylamide formation. Soybean oil generated the greatest quantity of acrylamide (2,600 ± 440 μg/kg) with Lady Rosetta variety and corn oil the least (1,920 ± 320 μg/kg), but overall the findings suggest that the type of frying oil has no significant impact upon acrylamide formation (Figure 7). This observation was found to be consistent with those from the various previous studies. In terms of acrylamide, Lady Rosetta>Spunta>> Voyager. This does not appear to be simply down to the asparagine levels. However, if the secondary impact of sugars is taken into account, the sugars (most notably glucose) are much lowest in voyager, and Lady Rosetta is higher than Spunta. The acrylamide looks like a sugars-asparagine effect. The acrylamide levels are extremely high for the processing method that was used, i.e. the production of French fries, but there is no comparison with other, similar studies or with the European Commission indicative level for French fries of 600 μg/kg. This suggests that reducing sugar concentration in the tubers was very high as shown in Figure 8. For example, Mestdagh et al. (2005) reported that the source of frying oils did not appear to influence the formation of acrylamide in fried potatoes and, therefore, the type of frying oil (cotton, olive, peanut, shortening, canola, soybean, seed, and sunflower) is not a significant variable for the acrylamide formation. Mestdagh et al. (2005) investigated the influence of the type of oil on acrylamide formation using a model system based on a closed stainless steel tubular reactor. Their study was also based on the addition of different types of frying oils, including olive oil, corn oil and soybean oil, separately to different artificial potato powder mixtures. These mixtures were heated at 170°C for 5 min. Although the results obtained indicated that the acrylamide levels were not equal for these different oils, no significant differences could be inferred between the tested oils. Heating fats or vegetable oils at high temperatures produce acrolein, resulting from the thermal degradation of glycerol or lipid. Oxidation of acrolein yields acrylic acid, which then reacts with a nitrogen source to form acrylamide (Skog & Alexander, 2006;Umano & Shibamoto, 1987;Yasuhara et al., 2003). Ehling, Hengel, and Shibamoto (2005) has reported the generation of high levels of acrylamide when fats or oils are heated with amino acid ASN, especially if these oils contain high concentrations of polyunsaturated fatty acid or linoleic acid. In this context they have low frying stability, suggesting that the highest concentration of acrylamide is formed when using soybean oil, followed by corn oil and olive oil. In this pathway, the oxidation of acrolein to acrylic acid seems to be a critical step in the acrylamide formation (Biedermann, Grob, Gutsche, & Weisshaar, 2002; Notes: Mean acrylamide concentrations extracted from French fries (μg/kg) fried in either olive, soybean, or corn oil originating from the potato varieties: Lady Rosetta, Spunta and Voyager. The error bars indicate the standard deviation from the means. Significant differences are indicated by asterisk p < 0.05. Taeymans et al., 2004). The inactivity of acrolein in Maillard reaction may be attributed to the fact that this acid immediately reacts with different food components instead of undergoing oxidation to acrylic acid. In addition, the common frying temperatures were found to be above the boiling degree of both acrolein (51°C), and acrylic acid (140°C). Therefore, these compounds that resulted from lipids degradation at elevated temperatures may quickly transform to their gaseous state, and thus do not participate in the formation of acrylamide in fried potatoes (Ehling et al., 2005). In general, although lipids and oils may play a minor role in the acrylamide formation and provide a desirable medium for heat transfer, these findings suggest that they do not appear to be significant precursors for the acrylamide formation as compared with the major acrylamide-sugar pathway. The present study was developed to address the effect of the type of frying oils on the concentrations of acrylamide precursors and the formation of acrylamide in French fries. The influence of the different types of frying oils (olive, corn, and soybean) on the concentrations of these precursors in French fries samples derived from three different potato cultivars (Lady Rosetta, Spunta, and Voyager) was investigated, and taken as an indicator of the acrylamide formation in these samples. As reported previously, differences in the acrylamide precursor concentrations were observed between the different varieties of potato. The variety Lady Rosetta having the highest concentrations of the precursor sugars and free asparagine, which correlated with the highest levels of acrylamide observed in the fried products. The lowest levels of acrylamide were observed in the Voyager variety independent of the oil used to fry the potato. Finally, it was observed that the acrylamide looks like a sugars-asparagine effect. This means that the predominant effect of reducing sugar concentration on the acrylamide formation in potato products is well established as confirmed by many studies. Contrary to the hypotheses formulated, it was found that there was no significant effect of the type of frying oil on the levels of acrylamide. The lowest acrylamide concentrations were observed in corn oil. Funding This work is funded by the European Social fund and national resources through Research program Archimedes III.	2017-11-03T00:56:52.066Z	2016-05-06T00:00:00.000	{ "year": 2016, "sha1": "40c284c69a0e9684e9d8a51486a1a978b9b0f88d", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1080/23311932.2016.1180950", "oa_status": "GOLD", "pdf_src": "ScienceParseMerged", "pdf_hash": "118b8981d8026815112054d4fedac63e1d631a99", "s2fieldsofstudy": [ "Agricultural And Food Sciences" ], "extfieldsofstudy": [ "Materials Science", "Chemistry" ] }
234336111	pes2o/s2orc	v3-fos-license	Ab initio simulation of the effects of substitutional doping of oxygen in the semifullerene (C30) Using first principles calculations, we studied the effects of substitutionally doping a 30-atom semifullerene (C30) with oxygen atoms, within the Density Functional Theory (DFT) and the pseudopotential formalism. We analyzed two C30 structures, one with a pentagonal and one with a hexagonal base. In both cases, the substitutionally absorbed oxygen atoms break the bonds between pentagons and hexagons of the semifullerene. We show that each semifullerene breaks into two geometrically flat structures, one of them being consistent with a known ring isomer of the fullerene C20. We used the Quantum ESPRESSO package code and the generalized gradient corrected Perdew-Burke-Ernzerhof (PBE) approximation for the exchange and correlation functional. Introduction Since their discovery, fullerenes have shown interesting physical and chemical properties [1]. Many studies have focused on their geometry, energetic stability, spectra, and interactions with other molecules. But theoretical and experimental investigations of fullerene molecules can be quite challenging, due in part to the large number of fullerene geometrical isomers. Hence, it is important to fully understand the factors determining the structures and stabilities of smaller carbon clusters [2][3][4]. Differences in geometry and topology between various fullerene isomers can lead to great differences in their chemical and physical properties [5][6]. For example, fullerene C20, -the smallest possible-has three isomers: cage, bowl and ring, with different properties [4]. In this work, we used DFT calculations to study a proposed heterofullerene compound [7] that includes oxygen atoms. We started with the C30 semifullerene, which results from the splitting in half of a C60 fullerene. After a structural relaxation of the pentagonal-based semifullerene (figure 1a), the bond length between a pentagon and a hexagon, labeled as (5,6), is equal to 1.425 Å. The bond length between two adjacent hexagons, labeled as (6,6), is 1.360 Å. Structural relaxation of the hexagonal-based semifullerene (figure 1b) results in a different structure consisting of four rectangle-shaped rings, in addition to the pentagons and hexagons already present. The bond length between a pentagon and a rectangle, labeled (5,4), is 1.540 Å, while the bond length between a hexagon and a rectangle, labeled (6,4) We also found for this second structure, that the bond length between two hexagons (6,6) and the bond length between one pentagon and one hexagon (5,6), were 1.436 Å and 1.485 Å respectively. All the former are average values over the set of available bonds. The experimental average values are 1.391 Å for the (6,6) and 1.455 Å for the (5,6) bond lengths [8]. Computational methods Self-consistent field calculations were used in this study, based on the Density Functional Theory (DFT) [9,10] and using the pseudopotential formalism. The Quantum Espresso code was used [11]. This package considers a plane-wave expansion for the electronic wave functions, and periodic boundary conditions. For the carbon and oxygen atoms, we used Martin-Troulliers norm-conserving pseudopotentials [12]. The Perdew-Burke-Ernzerhof (PBE-GGA) exchange and correlation functional was used [13]. The following valence electronic states were considered: for carbon C, 2s 2 2p 2 and for oxygen O, 2s 2 2p 4 . The Monkhorst-Pack [13] scheme for k-point sampling was used for integration over the first Brillouin zone. The Kohn-Sham energy functional was directly minimized using the conjugate gradient method. For the k-point sampling, a 4×4×4 grid was used, as well as a cut-off energy for the plane-wave set of 80 Ry (≈1100 eV). The convergence threshold for the energy was 1.2 × 10 -5 eV. Previous studies [14], lead us to use non-relativistic, non-polarized spin calculations, as they produce reliable results for this type of carbonbased materials. For crystal visualization purposes, we used the XCrySDen package [15]. Semifullerene bowl with a pentagonal base We considered the pentagonal-based semifullerene of 30 carbon atoms. This corannulene-like structure forms part of the C60 cage [3]. We first doped it substitutionally with five oxygen atoms, symmetrically placed around the carbon atoms forming the pentagonal base (see figure 2). After a structural relaxation of the system, we found that it splits into two separate systems. One of those new systems could be identified as the ring isomer of the fullerene C20. Semifullerene bowl with a hexagonal base The second case considered was the system with a hexagonal base. This sumanene-like structure forms part of the C60 cage as well [16]. We doped it substitutionally with 9 oxygen atoms as shown in figure 1. After being geometrically optimized, the structure breaks up again into two systems, one of which is a ring formed by 9 carbons (figure 5). The two structures obtained are separated by 3.80 Å (figure 5). As in the previous case, their planes are parallel and here the bond lengths of the smaller ring, composed of nine carbon atoms, range between 1.28 Å and 1.33 Å. The results obtained in this work, particularly for the C20 isomer structure, are consistent with previous studies [17,18]. These types of small carbon clusters are also potential candidates for carbon-based superconductors [19].	2021-05-11T00:06:54.337Z	2021-01-01T00:00:00.000	{ "year": 2021, "sha1": "2d0ca488ffe92cc7358a884dd283065cf93886ec", "oa_license": null, "oa_url": "https://doi.org/10.1088/1742-6596/1723/1/012054", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "8a05e5876c6a3b37a38e824f1933d26000446249", "s2fieldsofstudy": [ "Materials Science", "Physics", "Chemistry" ], "extfieldsofstudy": [ "Physics", "Materials Science" ] }
268569857	pes2o/s2orc	v3-fos-license	The mechanism of solid acid-catalyzed bamboo sawdust liquefaction under polyol systems Solid acid catalysts are widely used in the field of biomass catalytic conversion owing to their advantages of low environmental pollution, easy separation and reusability. Nevertheless, there are relatively few studies on the mechanism of solid acid liquefaction for biomass. In this study, the effect of acid strength and acid amount of various solid acids on the liquefaction efficiency has been investigated using waste bamboo sawdust generated from the pulp and paper industry as the raw material. In addition, the physicochemical changes of cellulose, hemicellulose and lignin during the reaction process of bamboo sawdust have been studied, and the liquefaction mechanism of bamboo sawdust under the action of various solid acids has been concluded. As a result, the liquefaction efficiency of bamboo sawdust under the polyol system of PEG400/propanetriol is mainly related to the acid strength of the solid acid, and the greater the acid strength of the solid acid, the better the catalytic effect on the bamboo sawdust, in which the residual amount of bamboo sawdust liquefaction catalyzed by the SPA catalyst is only 17.72%. Noteworthy, the most difficult component to liquefy is the crystallization of natural cellulose I into cellulose II during the reaction process, which is the primary obstacle to the complete liquefaction of bamboo sawdust by solid acid. Overall, these findings are valuable for the high value utilization of waste bamboo sawdust in the pulp and paper industry, as well as the application of solid acid catalytic technology for biomass. Introduction Biomass liquefaction is an effective method of utilizing biomass and is the basis for the preparation of various bio-based materials.In this process, acid is usually added as a catalyst, since it can effectively reduce the reaction temperature and improve the liquefaction efficiency of biomass (Pan, 2011).The commonly used homogeneous acids are sulfuric acid, hydrochloric acid, and phosphoric acid (Jensen et al., 2010;Kumar et al., 2017;Zhang et al., 2019), whose drawbacks are the difficulty of catalyst recycling, the amounts of wastewater in the production process and the corrosion of the equipment, while the solid acid catalysts have the advantages of low environmental pollution, easy separation.In addition, although inorganic acids are cheaper than solid acids, the higher costs of environmental protection equipment, production and maintenance do not provide a clear economic advantage in terms of the overall process. Solid acid-catalyzed liquefaction of biomass has been extensively reported.For example, Santander et al. showed that solid phosphoric acid supported by mesoporous silica catalyzed the liquefication of cellulose, with levogluconone detected in more than 85% of the liquid product (Santander et al., 2019).In addition, the hydrothermal pyrolysis of SO 4 2-/ Fe 2 O 3 , SO 4 2-/Al 2 O 3 , SO 4 2-/TiO 2 , SO 4 2-/ZrO 2 and SO 4 2-/SnO 2 was studied by Yang et al. (2015), with the highest acidic strength, SO 4 2-/SnO 2 , yielding 11.0% and 26.8 %for HMF and glucose, respectively.Li et al. (2020) reported a multifunctional catalyst (Cl-MCMB-SO 3 H) for the hydrolysis of cellulose.Compared with the catalyst MCMB-SO 3 H, the Cl − in the catalyst Cl − MCMB-SO 3 H favored the formation of hydrogen bonds, which reduced the activation energy of cellulose hydrolysis by about 20 kJ/mol, improving the hydrolysis efficiency with a high reducing sugar yield of 70.3%.Additionally, Liu et al. (2011) presented a multifunctional solid acid catalyst (Ru/Cs 3 PW 12 O 40 ) to catalyze cellulose after ball milling, with sorbitol yields up to 43%, which showed good catalytic performance even after 5 times of reuse.From the current research, the solid acids used for biomass catalysis are mainly concentrated in the similar class of solid acids.In particular, there is a lack of systematic research on the compositional relationship between different types of solid acids for biomass liquefaction, and the structure of solid acids and biomass liquefaction, which leads to the selection and preparation of solid acids for biomass liquefaction mainly relying on experience. The utilization of bamboo for pulp and paper production is a promising development in Asia, especially in China.It is expected that in 5 years, bamboo-based pulp and paper products will reach 10 million tonnes per year.Accordingly, a million tonnes of bamboo sawdust will be generated, which cannot be utilized at high value.The aim of this study was to investigate the effects of acid strength and acid quantity of various solid acid types on the liquefaction efficiency of bamboo sawdust.In addition, the physicochemical changes of cellulose, hemicellulose and lignin during the reaction process of bamboo sawdust were also taken as the focus.Finally, the liquefaction mechanism of bamboo sawdust under various solid acid conditions was proposed. Catalysts preparation ZrO 2 , SnO 2 , TiO 2 and Fe 2 O 3 were obtained by sintering Zr(OH) 4 , Ti(OH) 4 , Fe(OH) 3 and Sn(OH) 4 in air, respectively.ZrO 2 , SnO 2 , TiO 2 and Fe 2 O 3 were impregnated with 1 mol L -1 sulfuric acid for 1 h (metal oxide: sulfuric acid = 1 g: 15 mL), and then calcined in a muffle furnace.Typically, the calcination temperature was 500 °C for Fe, Sn, Ti and 600 °C for Zr, as well as the calcination time was 3 h for Zr, Fe and 4 h for Ti, but 5 h for Sn.The final prepared superacid catalysts were named as SO 4 2− / Fe 2 O 3 , SO 4 2− /TiO 2 , SO 4 2− /ZrO 2 and SO 4 2− /SnO 2 , respectively.5.0 g of diatomaceous earth (Chengdu Kolon Chemicals Ltd.) and 28.8 g of 85% phosphoric acid (Chengdu Kolon Chemicals Ltd.) was mixed and stirred to form a viscous paste mixture, which was then calcined in a muffle furnace at 300 °C for 48 h to yield white solid.Afterwards, the resulting product was cooled to room temperature in dry environment and ground into powder to produce solid phosphoric acid "SPA". All catalysts have been passed through a sieve with 0.15 mm aperture. Solid acid-catalyzed liquefaction of bamboo The heterogeneous catalytic degradation reaction system was formed in an oil bath at 170 °C for 18 h.5 g bamboo sawdust was added to a mixture of 30 g of PEG 400 and 10 g of glycerol, and 1.5 g solid acid catalyst was added.In addition, the filtrate was filtered and the residue was washed with 1,4-dioxane/water (4:1, v/v) until the reacted detergent was colorless.After that, the detergent was recovered by rotary evaporation at 60 °C and reused.The filter residue was dried at 105 °C for 8 h and then calcined at 550 °C before calculating the liquefaction conversion.The mass burned off is the mass of organic matter in the bamboo sawdust residue, and this calculation ignores the effect of solid acid loss. Where the R represents the residue yield (%), m r refers to the residue mass (g), m c stands for the residue mass after burning (g) and m 0 is the bamboo sawdust mass (g). NH 3 -TDP analysis The acid strength and acid content of the catalysts were determined by ammonia temperature-programmed desorption (NH 3 -TPD) over a temperature range of 100 °C-700 °C.And the measuring instrument is an AutoChem II 2920 from Mike Company. Chemical component analysis Concentrations of dextran, xylan, and lignin in solid samples and post-reaction residues were assessed using standard technique from the Renewable Energy Laboratory (NREL).Sugar content was assessed using an Agilent high performance liquid chromatography (2160II, CA, United States) and a Bio-rad Aminex PHX-87H sugar column, with signal detection using an oscillometric refractive index detector.Acid-soluble lignin was identified using an UV spectrophotometer (UV-Vis, T6 series, Beijing spectrum analysis General Instrument Co.). FT-IR analysis The IR spectra of the samples were measured in the wavenumber range of 400-4,000 cm -1 (Nicolet iS10, Thermo Fisher Scientific, Waltham, MA, United States).Solid samples were prepared as 1 wt% KBr pressed tablets, and IR spectra of liquid samples were determined by removing the solvent (PEG 400/glycerol = 3/1) background. XRD analysis X-ray diffraction (XRD) patterns of the samples were obtained on a Bruker D8 Focus X-ray diffractometer utilizing Cu K radiation.In detail, the samples were scanned in 4 °increments with a counting time of 1 min per spot over a range from 10 °to 80 °. Thermogravimetric analysis Samples were collected on a TA SDT650 thermogravimetric analyzer.The samples were ramped from room temperature to 900 °C in N 2 atmosphere (40 mL/min) at a rate of 10 °C/min. Solid acid and liquefaction rate The liquefaction effects of different solid acids on the same mass (1.5 g) of bamboo sawdust are shown in Table 1.Different kinds of solid acids show great differences on bamboo sawdust, among which the residue rates of loaded acid SPA, heteropolyacid PTA and SO 4 2-/ M x O y are 17.72%, 36.7% and 56.96%-68.02%,respectively.Therefore, among these catalysts, SPA exhibits the greatest effect on the liquefaction of bamboo sawdust.However, from the NH 3 -TPD profiles (Figure 1), it can be seen that the heteropolyacid PTA is the most acidic as it has obvious NH 3 resolution peaks in the high temperature region relative to SPA and SO 4 2-/M x O y .According to previous studies, the stronger the acidity, the better the effect on biomass liquefaction (Lin et al., 2022).Nevertheless, for PTA and SPA, two typical proton solid acids, the opposite result is observed.NH 3 -TPD curves of catalysts samples. Frontiers in Bioengineering and Biotechnology frontiersin.org The "free P 2 O 5 " on the surface of SPA readily interacts with low molecular organic matter.As a result, SPA is susceptible to forming complexes with low molecular organic matter, which makes it easier to come into contact with the substrate.In addition, water is produced during the reaction, which hydrolyzes some of the solid acid on the surface of the SPA to produce liquid phosphoric acid (Santander et al., 2019).Then, the liquid phosphoric acid comes into contact with the reactants and catalyzes the liquefaction of bamboo sawdust.As a result, the residue rate is the lowest under SPA catalyzation.These results indicate that the loaded solid acid has obvious advantage in catalyzing the liquefaction of bamboo sawdust in polyol alcohol system.The desorption temperature is related to the acid strength of the solid acid.Generally, a higher desorption temperature indicates a more difficult desorption of NH 3 and therefore a higher acid strength of the solid acid (Liu et al., 2014).In the lowtemperature region of 100 °C-200 °C, all types of solid acids show resolved peaks, which are mainly due to the physical adsorption of NH 3 on the surface of solid acid as well as its chemisorption on the weakly acidic sites.In the middle and high temperature region, compared with other solid acids, PTA solid acid has a strong absorption peak in the high temperature region around 550 °C, and thus has the highest acid strength.In addition, the NH 3 resolution peak of SPA is higher than that of SO 4 2-/SnO 2 in the middle and high temperature region.Therefore, the order of acid strength of these three solid acids is PTA > SPA > SO 4 2-/SnO 2 .In Table 1, the residue rates of bamboo sawdust after liquefaction of the three types of solid acids were ranked as follows: SPA (17.72%) > PTA (36.72%) > SO 4 2-/SnO 2 (56.96%).The acidity of PTA is higher than that of SPA, but in contrast, the residual rate after PTA treatment is higher.This is probably because SPA is a loading acid that produces small molecules of water during the liquefaction of bamboo sawdust, which hydrolyzes part of the solid acid on the surface of SPA to produce liquid phosphoric acid (Santander et al., 2019).In addition, the liquid phosphoric acid comes into contact with the reactants and catalyzes the liquefaction of bamboo sawdust under their joint action with the lowest residual rate.However, the residual rate of bamboo sawdust liquefaction catalyzed by PTA is smaller than that of catalyzed by SO 4 2-/SnO 2 , which may be attributed to the fact that PTA is more acidic than SO 4 2-/SnO 2 . As shown in the NH 3 -TPD profiles (Figure 1 /Fe 2 O 3 were 61.59%, 68.02% and 65.21%, respectively.These results indicated that the stronger the acidic site of solid acid, the better the liquefaction effect of bamboo sawdust in polyol system.In addition, the liquefaction residue rate of SO 4 2-/Fe 2 O 3 was essentially the same as that of SO 4 2-/ZrO 2 and SO 4 2-/TiO 2 , suggesting that the weak acid sites of the solid acid had no impact on the catalytic effect of bamboo sawdust in the polyol system. In summary, the liquefaction efficiency of bamboo sawdust under the polyol system of PEG400/glycerol is mainly related to the acid strength of the solid acid and the type of the solid acid with less correlation to the total amount of the acid in the solid acid: (1) the stronger the solid acid, the better the catalytic effect; and (2) the loading of solid phosphoric acid, the better the catalytic effect.This is because water molecules will be generated during the reaction process, and the water molecules will lead to the hydrolysis of solid phosphoric acid to generate liquid phosphoric acid, which will increase the reaction rate and make the bamboo sawdust liquefied more fully under the polyol system. Analysis of chemical composition of residues In order to investigate the effect of different solid acids on the liquefaction process of bamboo sawdust, the biomass fractions in the solid acid-catalyzed residue were determined.As shown in Figure 2, the relative content of cellulose in the residue increased in all catalyst groups compared with that of bamboo sawdust, with the most significant increases in the SPA and PTA groups.These results indicate that cellulose is the most difficult part to liquefy during the liquefaction of bamboo sawdust.Acid-soluble lignin was fully reacted in all catalyst groups, while the content of acid-insoluble lignin varied.In addition, the hemicellulose content varies among the catalyst groups, with complete reaction of hemicellulose in the SPA and PTA groups, and changes in the hemicellulose content of the other groups compared to the bamboo sawdust.In order to quantify the effect of solid acid types on the content of three major components in the liquefaction of bamboo sawdust, the relative contents of these components in bamboo sawdust (before liquefaction) and their residues (after liquefaction) were quantified (Table 2).The relative content of the three major elements (X) in the residue (after liquefaction) as a percentage of the pre-liquefaction bamboo sawdust was calculated according to the following equation (Niu et al., 2011): where M 1 represents the relative content of the three major elements in the residue and R refers to the yield of the liquefied residue. Cellulose is most difficult to liquefy because of its crystalline structure, which makes it difficult for the liquefying agent to act on it.In Table 2, the cellulose content in bamboo sawdust was 58.86%, which decreased under all conditions by adding solid acid.The cellulose contents of bamboo sawdust decreased to 39.51%-47.23%catalyzed by solid acid SO 4 2-/M x O y .In comparison, the cellulose content in bamboo sawdust was 27.29% using PTA with higher acid strength as catalyst.Therefore, the increase in solid acid strength facilitated the liquefaction of cellulose in bamboo sawdust.When SPA was used as a catalyst, the cellulose content was only 13.82%, which was attributed to the hydrolysis of solid phosphoric acid on the surface of SPA to small molecule phosphoric acid, promoting the hydrolysis of cellulose (Espinosa et al., 2013).Comparative analysis of the data show that cellulose in bamboo sawdust is the most difficult to be liquefied under the polyol system of PEG400/glycerol, which is the main obstacle restricting the complete liquefaction of bamboo sawdust by solid acids. The hemicellulose content of bamboo sawdust is 9.41%, which is relatively low.Moreover, the addition of various solid acid catalysts can promote the liquefaction of hemicellulose in bamboo sawdust.Notably, there is no hemicellulose in the residues after solid acid catalysis by PTA and SPA.Therefore, the hemicellulose in bamboo sawdust can be completely liquefied by solid acid catalysis with sufficient acidity in the polyol system. The lignin in bamboo sawdust could be categorized into acidsoluble lignin and acid-insoluble lignin, with contents of 4.43% and 27.30%, respectively.The addition of various solid acids completely liquified the acid-soluble lignin in bamboo sawdust, indicating that the acid-soluble lignin in bamboo sawdust was more easily liquefied.Moreover, the solid acid catalysts could also promote the liquefaction of acid-insoluble lignin in bamboo sawdust.Among them, the liquefaction effect of SPA was the best, which could reduce the acid-insoluble lignin content in bamboo sawdust from 27.30% to 4.25%.The lignin content in bamboo sawdust was extremely low.Studies have shown that lignin is more easily liquefied by acid and this incomplete liquefaction may be due to the long liquefaction reaction time.Additionally, the decomposed small-molecule lignin was further condensed to large-molecule compounds (Xie et al., 2014), which remain in the residue.Therefore, acid-insoluble lignin is present in the residue. Conclusively, the addition of solid acids promoted the liquefaction of the three main elements in bamboo in the polyol system of PEG400/glycerol.Acid-soluble lignin was the easiest to liquefy, and the addition of all types of solid acids resulted in its complete liquefaction.This was followed by hemicellulose and acidinsoluble lignin, which could be completely liquefied by solid acids with increasing acidic strength.Frontiers in Bioengineering and Biotechnology frontiersin.org Cellulose is the most difficult to be liquefied, which may be due to the fact that cellulose molecules are aggregated in the form of micro protofibrils, in which the crystalline and amorphous regions alternate, and the amorphous portion of cellulose is easily hydrolyzed during acid hydrolysis, leaving the crystalline fraction behind (Li, 2004;Kuthi and Badri, 2014).Meanwhile, during the liquefaction process, the liquefaction products undergo condensation reactions with lignin, condensation products tend to remain on the crystalline cellulose in the residue (Xie et al., 2014).These factors make it difficult for the residual cellulose to come into contact with the liquefiers for further reaction. FT-IR analysis of liquefaction residues In order to compare the changes of functional groups on the surface of residues after liquefaction with different catalysts, the liquefaction residues, bamboo sawdust and the mixture of bamboo Infrared spectra of BS, BS and catalyst mixture as well as liquefaction residue. sawdust and solid acid with 1.5 g of catalyst were characterized by IR spectroscopy, and the results were shown in Figure 3. The infrared absorption peaks of the main components of BS (cellulose, hemicellulose and lignin) can be seen in the infrared spectra.The broad peaks between 3,300 and 3,500 cm-1 belong to the stretching vibration peak of -OH.Moreover, the absorption peak at 2,926 cm-1 is attributed to the vibration absorption peak of -CH3, while the absorption peak appearing at 1733 cm-1 should be attributed to the uncoupled C=O stretching vibration, which is also the absorption peak of characteristic functional groups of hemicellulose (Wen et al., 2011;Xie et al., 2014).Notably, the peak intensity is not high due to the small content in BS.The absorption peaks at 1,629 cm-1, 1,518 cm-1 and 1,457 cm-1 belong to the benzene ring skeleton vibration, corresponding to the lignin structure in bamboo sawdust (Zhang et al., 2010).Absorption peak at 1,384 cm-1 is attributed to C-H bending vibration, and the peak at 1,053 cm-1 belongs to C-O stretching vibration, which is also a characteristic absorption peak of cellulose and hemicellulose in the infrared (Latif et al., 2014).One of the peaks at 896 cm-1 belongs to β-glucoside bond vibration, which is also the infrared characteristic absorption peak of cellulose (Chen et al., 2011).Meanwhile, the mixture of BS and solid acid catalyst were characterized by IR spectroscopy.It was found that the addition of solid acid catalyst had no effect on the position and intensity of the IR characteristic absorption peak of BS (cellulose, hemicellulose and lignin).Therefore, the effect of solid acid-catalyzed liquefaction of BS could be evaluated by comparing the changes in the IR characteristic absorption peak of BS and liquefaction residue. As shown in Figure 3, the IR spectral curves of pure BS, the mixture of BS and catalyst, and the residue of solid acid liquefaction are shown from top to bottom, respectively.Solid acids such as SO42-/ZrO2, SO42-/TiO2, SO42-/Fe2O3 and SO42-/SnO2 can catalyze the liquefaction of bamboo sawdust, but the residue rate is high.The liquefaction rate was low, and only part of cellulose, hemicellulose and lignin were liquefied (Figure 2; Table 2).As a result, the position of IR characteristic peaks was basically unchanged and the intensity did not change much, which was consistent with the detection results of residue compositions. When PTA with high acidity was used as catalyst, the IR absorption peak of hemicellulose at 1733 cm-1 basically disappeared compared with the BS spectrum, indicating that hemicellulose had been completely reacted.In addition, the absorption peaks of cellulose and hemicellulose at 1,053 cm-1 also disappeared, while new absorption peaks, i.e., the IR absorption peaks of cellulose, appeared at 1,080 cm-1, 1,053 cm-1 and 896 cm-1.Meanwhile, a new IR absorption peak was observed at 812 cm-1, which should be attributed to α-glycosidic bond vibrations (Wen et al., 2011), indicating an increase in cellulose content in the residue.This result is consistent with the detection of the residue compositions. When SPA with better liquefaction effect was used as catalyst, the IR absorption peak at 1733 cm-1 of hemicellulose was found to be completely disappeared when compared with the BS IR spectrum, indicating that hemicellulose was completely liquefied.At the same time, a new peak reappeared at 1719 cm-1, which was attributed to the absorption peak of conjugated C=O in aromatic nuclei.After liquefaction, the surface lignin was enriched on the residue surface by the products of polycondensation.In addition, the IR characteristic absorption peaks of cellulose and hemicellulose at 1,053 cm-1 of bamboo sawdust catalyzed by SPA disappeared, and two new peaks at 1,187 cm-1 and 1,094 cm-1 were formed with high peak intensity.These two peaks are the infrared characteristic absorption peaks of cellulose (Wen et al., 2011), indicating that the content of cellulose in liquefaction residue is increased.On the other hand, the intensities of the IR characteristic peaks of 1,629 cm-1, 1,518 cm-1 and 1,457 cm-1 in the liquefied residue were reduced compared with those in the BS, indicating lignin degradation, which was consistent with the results of residue components. Above infrared analysis results showed that with the increase of the acidity of the solid acid, the hemicellulose and lignin in bamboo material disappeared or weakened, while the characteristic absorption peak of residual cellulose was enhanced and the liquefaction effect was improved.All these indicated that the stronger the acid strength of solid acid, the better the liquefaction effect of bamboo sawdust under the polyol system of PEG400/glycerol. XRD analysis of liquefaction residue Figure 4 shows the diffraction intensity curves of the liquefaction residue and BS under the condition of 1.5 g catalyst dosage. According to the XRD pattern of BS, the main diffraction angles 2θ °are 16.06 °, 22.05 °and 26.30 °, among which 16.06 °and 22.05 °are the diffraction peaks of natural cellulose Ⅰ (Langan et al., 2001;Nishiyama et al., 2002).In addition, all samples have a narrow diffraction peak at 26.30 °, which may be the absorption peak corresponding to the BS containing SiO2 (Tymchyshyn and Xu, 2010). Compared with BS, the liquefaction residue of SO42-/ZrO2, SO42-/TiO2, SO42-/Fe2O3 has diffraction peaks of cellulose Ⅰ at 22.05 °and 16.06 °.It is noteworthy that the shape of the peak remains almost unchanged, but there is a decrease in strength, which may be due to the liquefaction of some cellulose in the bamboo.In addition, this may also be due to the presence of a large amount of solid acid catalyst in the residue. With the increase of liquefaction efficiency, the diffraction absorption peak belonging to cellulose Ⅰ is disappeared at 16.06 °when SO42-/SnO2 is used as catalyst, indicating that part of cellulose Ⅰ also reacts. When PTA is used as a catalyst, the characteristic diffraction peaks of natural cellulose at 22.05 °disappear, while the diffraction peaks of cellulose II at 21.96 °and 20.80 °appear (Langan et al., 2001), which may be attributed to the transfer of the metastable structure of natural crystalline cellulose Ⅰ from bamboo sawdust to thermodynamical stable cellulose Ⅱ (Isogai et al., 1989;Langan et al., 2001). A similar phenomenon is observed in the SPA liquefaction residue when the residue rate is only 17.72%.The new diffraction absorption peaks are appeared at 21.96 °, 20.80 °and 22.96 °. Meanwhile, the new characteristic diffraction peaks belonging to cellulose are sharp, indicating that the crystalline cellulose on the surface of the residue has been reacted, so the crystallinity of the remaining cellulose has been improved, which explains the difficulty of cellulose liquefaction in bamboo sawdust. From the above XRD analysis results, it can be inferred that with the increase of solid acid acidity, the amorphous part of cellulose will be liquefied firstly during the catalytic process, leaving the crystalline part of natural cellulose, due to the alternation of crystalline and amorphous regions of natural cellulose.Then, when the acidity of the solid acid is further enhanced, part of the natural crystalline cellulose will be liquefied.Part of it is converted into the thermodynamically more stable cellulose II, which is the reason why bamboo sawdust is difficult to be completely liquefied. Thermogravimetric analysis of liquefaction residue Figure 5 shows the TG curve of liquefaction residue and the BS to remove the inorganic salts as well as the remaining solid acid from the residue at a catalyst dosage of 1.5 g.Based on the data in Figure 3, the weight loss of all the samples can be calculated and the results are shown in Table 3. From Figure 5, it can be seen that the weight loss zone of bamboo sawdust can be divided into three parts.In stage a, the weight loss rate is 4.85%, which is ascribed to the evaporation of water from the residue.In stage b, the weight loss rate is as high as 90.81%, which is mainly due to the pyrolysis of lignin, cellulose and hemicellulose according to the compositional analysis in Figure 2. In stage c, the weight loss rate is only 4.34%, which is primarily ascribed to the pyrolysis of charcoal residue after bamboo pyrolysis. The weight loss of SO 4 2-/MxOy solid acid, SPA and PTA liquefaction residue is basically the same as that of the bamboo sawdust at stage a in the weight loss zone, which is mainly due to the evaporation of water from the residue, which is mainly due to the evaporation of water from the residue. Compared with the weight loss of bamboo sawdust in the b-zone stage, the significant weight loss temperature range of all solid acid residues in this stage is shortened, with significantly faster weight loss and lower weight loss rate (Figure 3), which is inversely correlated with the acid strength.This is attributed to the liquefaction of cellulose, hemicellulose and lignin in bamboo cuttings catalyzed by solid acid.Notably, the greater the acid strength, the more complete the liquefaction, and the more complete the destruction of the main structure of bamboo sawdust.Therefore, in stage b, the temperature interval of weight loss is significantly shortened, and the rate of weight loss is significantly accelerated, while the weight loss rate is reduced. For stage c, the TG curves of all catalyst liquefaction residues and bamboo cuttings are relatively flat, and the weight loss rate is relatively slow, which might be due to the pyrolysis of the more thermally stable organic matter in the liquefaction residues. As for stage d, it can be seen from Figure 5 that only the liquefaction residue of SO42-/SnO2, SPA and PTA have thermogravimetric regions in the high-temperature region at this stage.The relationship is as follows: the range of weight-loss temperature, weight loss rate and weight loss rate of the residue in stage d increase with the decrease of the liquefaction residue rate, which indicates that the three elements in the liquefied bamboo sawdust produce products that are difficult to be pyrolyzed with the increase of the acidity of the solid acid.By analyzing the weight loss curve in stage d, it is found that the weight loss rate of SPA liquefaction residue is 35.84%.From Figure 2, it can be seen that the composition of SPA liquefaction residue is 21.96% acidinsoluble lignin and 78.04% cellulose.Therefore, it can be judged that the material of heat weight loss at this stage is mainly cellulose.Meanwhile, XRD results show that the cellulose in the residue is mainly composed of cellulose Ⅱ with relatively high crystallinity. Previous studies have shown that cellulose is less prone to weight loss during low-temperature pyrolysis when the crystallinity of cellulose is high (Mukarakate et al., 2016) cellulose is less prone to weight loss during low-temperature pyrolysis when the crystallinity of cellulose is high.This explains well why the liquefied residues of SO42-/SnO2, PTA and SPA increase by an interval of weight loss at 700 °C. Conclusion In conclusion, the effects of different solid acids as catalysts on the liquefaction rate of bamboo biomass have been investigated.Consequently, the liquefaction efficiency of bamboo sawdust was mainly related to the acid strength of the solid acid under the polyol system of PEG400/glycerol.Notably, the higher the acid strength of solid acid, the better the catalytic effect on bamboo sawdust.After the liquefaction of bamboo sawdust, the residual amount of SPA catalyst was only 17.72%, which catalyzed the liquefaction of the majority of the biomass in bamboo sawdust.Additionally, the most difficult component to liquefy was the crystalline conversion of natural cellulose I to cellulose II during the reaction process, which was the main obstacle limiting the complete liquefaction of bamboo sawdust by solid acid. FIGURE 2 FIGURE 2Effects of different catalysts on the mass fractions of chemical composition in liquefied residues. FIGURE 5 TG FIGURE 5TG curves of bamboo sawdust and solid acid liquefaction residues. TABLE 1 Influence of catalysts on liquefaction effect. TABLE 2 Chemical composition of bamboo sawdust and liquefaction residues. *The chemical composition were calculated according to Eq. 2. TABLE 3 Weight loss rates for all samples at different temperature ranges.Wu et al. 10.3389/fbioe.2024.1372155	2024-03-22T15:28:33.698Z	2024-03-20T00:00:00.000	{ "year": 2024, "sha1": "9b674fbc6e98128dc5197daad95ca42135c33fd2", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fbioe.2024.1372155/pdf?isPublishedV2=False", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "2ff2e7674c8372d569c123a32a5b7d4eb061e199", "s2fieldsofstudy": [ "Environmental Science", "Chemistry", "Materials Science" ], "extfieldsofstudy": [ "Medicine" ] }
17885321	pes2o/s2orc	v3-fos-license	KMS States, Entropy and the Variational Principle in full C-dynamical systems To any periodic, unital and full C-dynamical system (A, \alpha, R) an invertible operator s acting on the Banach space of trace functionals of the fixed point algebra is canonically associated. KMS states correspond to positive eigenvectors of s. A Perron-Frobenius type theorem asserts the existence of KMS states at inverse temperatures equal the logarithms of the inner and outer spectral radii of s (extremal KMS states). Examples arising from subshifts in symbolic dynamics, self-similar sets in fractal geometry and noncommutative metric spaces are discussed. Certain subshifts are naturally associated to the system and the relationship between their topological entropy and inverse temperatures of extremal KMS states are given. Noncommutative shift maps are considered. It is shown that their entropy is bounded by the sum of the entropy of the associated subshift and a suitable entropy computed in the homogeneous subalgebra. Examples are discussed among Matsumoto algebras associated to certain non finite type subshifts. The CNT entropy is compared to the classical measure-theoretic entropy of the subshift. A noncommutative analogue of the classical variational principle for the entropy of subshifts is obtained for the noncommutative shift of certain Matsumoto algebras. More generally, a necessary condition is discussed. In the case of Cuntz-Krieger algebras an explicit construction of the state with maximal entropy from the unique KMS state is done. Introduction Let A be a unital C * -algebra endowed with a 2π-periodic automorphic action α of R. In algebraic statistical mechanics elements of A represent kinematic observables of an infinite quantum system, and α is the time evolution of the system. Equilibrium states of the system are states on A which satisfy the KMS condition with respect to α. We shall assume throughout the paper that α is 2π-periodic, so it factors through an action γ of the circle T, and also that γ is full , in the following sense. Let A k be the spectral subspace of elements a ∈ A such that γ z (a) = z k a. We assume that for all k ∈ Z, the closed linear span of {xy, x ∈ A k , y ∈ A −k } is the fixed point algebra A 0 . An example is given by a crossed product C * -algebra A = B ⋊ β Z by a single automorphism β, endowed with the dual action γ =β. All KMS states on A are tracial, and are given by T-invariant extensions to A of β-invariant tracial states on B. More generally, if (A, γ) is not a dual C * -dynamical system, nontracial KMS states arise. Interesting examples are, in increasing generality, the Cuntz-Krieger algebras [CK], the Matsumoto algebras associated with a subshift [M], and the Pimsner algebras associated with a full finite projective Hilbert C bimodule [P], all endowed with the canonical gauge action. While KMS states for the first two classes of C -algebras are now well understood (see [MFW], [E], [MYW]), the third class of C * -algebras is the main motivation of the present paper. (We shall see that Pimsner C * -algebras are in fact a typical example, in the sense that any unital, full and periodic C * -dynamical system is isomorphic to a system constituted by a Pimsner C * -algebra associated to a full, finite projective Hilbert bimodule, though not unique, and its canonical gauge action. ) We point out that, on the other hand, various authors, regarding the Cuntz-Krieger algebras as examples of noncommutative topological dynamical systems, have computed the Voiculescu topological entropy [V] of the so-called 'canonical endomorphism' σ ( [Ch], [BG]). However, among these, to the authors' knowledge, it is only for the case of the Cuntz algebras O d that a close relationship is known between KMS states, Voiculescu's topological entropy and measure-theoretic entropy in the sense of [CNT], see [Ch]. Choda's result states that the CNT entropy of σ computed with respect to the unique KMS state equals the topological entropy of σ, which is, in turn, log(d). This result can be regarded as a pivotal example of noncommutative dynamical system for which a variational principle for the entropy holds. Our ultimate goal is that of investigating the variational principle for the entropy and its relationship with KMS states, in full periodic C * -dynamical systems. KMS states for 2π-periodic actions have already been considered in the literature by several authors. Olesen and Pedersen gave in [OPI] an existence and uniqueness theorem for KMS states of the Cuntz algebras. This was generalized to the case of Cuntz-Krieger algebras by Enomoto, Fuji and the second-named author in [EFW] and by Evans in [E]. In [BEH] Bratteli, Elliott and Herman construct, for any closed subset F of the extended real line, a simple C * -algebra A F endowed with a 2π-periodic one-parameter group, for which F is precisely the set of inverse temperatures of KMS states, and such that for each β ∈ F , A F has a unique KMS state at inverse temperature β. We remark that if F ⊂ R, then the T-action is full, and if F ⊂ (0, +∞), A F is purely infinite (cf. section 2). In a subsequent paper [BEK] Bratteli, Elliott and Kishimoto show that even the set of KMS states with a specified inverse temperature can be fairly arbitrary. In a recent paper [MWY] Matsumoto, Yoshida and the second-named author study KMS states for the Matsumoto C * -algebras associated with a subshift in symbolic dynamics. They develop a Perron-Frobenius type theorem for a suitable positive operator naturally acting on a certain subalgebra, and show that the logarithm of its spectral radius arises as the inverse temperature of some KMS state. Furthermore they show a connection with the topological entropy of the underlying subshift. Our approach is close to that of [MWY], in that we emphasize the Perron-Frobenius theory. The starting point is that to any periodic and full C dynamical system we associate certain completely positive maps on the underlying C -algebra, which we interpret as being Perron-Frobenius type operators. KMS states correspond then to the positively scaled tracial states on the fixed point algebra. We study the problem of existence of KMS states, thus proving a Perron-Frobenius type theorem, and the relationship with the variational principle in ergodic theory. The paper is organized as follows. In the first section choose finite subsets {y i } and {x j } of A 1 such that j y j * y j = I and i x i x i * = I. Such multiplets exist because the group action is full. They can be regarded as playing the role of the canonical unitary in B ⋊ β Z implementing β. We then consider two completely positive (cp) maps: T {yi} : T → i y i T y i * and S {xj } : T → j x j * T x j on A 0 , and also, by transposition, operators t ′ and s ′ which are inverses of one another on the Banach space of trace functionals on A 0 . These operators are independent of the choice of the multiplets {y i } and {x j }. KMS states for the system at finite inverse temperatures then correspond to tracial states on A 0 which are positively scaled by those cp maps, or, equivalently, to tracial state eigenvectors of s ′ . In the next section we show that, under the necessary condition that the fixed point algebra has a tracial state, the inner and outer spectral radii of s ′ correspond to inverse temperatures of 'minimal' and 'maximal' KMS states (see Theorem 2.5 and Corollary 2.6). This can be regarded as a Perron-Frobenius theorem. The key point in the proof is that one needs to consider the trace functionals of the enveloping von Neumann algebra of A 0 , endowed with its order structure. In sections 3-5 we discuss some examples. In section 3 we apply our results to the Pimsner C * -algebras generated by finite projective Hilbert bimodules, and we thus deduce a criterion for existence of KMS states which applies, in particular, to the case where the coefficient algebra is simple, unital and has a tracial state. In section 4 we construct Hilbert bimodules, and hence full C * -dynamical systems, via Pimsner's construction, naturally arising from two different situations: subshifts of symbolic dynamics and self-similar sets in fractal geometry. In both cases the coefficient algebra is commutative, and the corresponding Hilbert bimodules are described by a finite set of endomorphisms. We show that in the former situation Pimsner's construction yields the Matsumoto C * -algebras, while in the latter one gets a genuine Cuntz algebra. We also discuss a generalization of the latter example to noncommutative metric spaces introduced by Connes [Co]. It is interesting to compare our discussion with the papers by Jørgensen-Pedersen [JP] and Bratteli-Jørgensen [BJ], where the authors consider a relationship between Cuntz algebras and multiresolutions in wavelet and fractal analysis. In section 5 we look more closely at the subclass of the so called Cuntz-Krieger bimodules (and the corresponding C * -algebras). These are bimodules for which the coefficient algebra is a finite direct sum of unital simple C algebras. The leading and simplest example is, of course, that of Cuntz-Krieger algebras, where each summand algebra is a copy of the complex numbers. We show in particular that if each of the summands has a unique trace and the defining {0, 1}-matrix A is irreducible then the associated Pimsner C -algebra has a unique KMS state at inverse temperature log(r(A)), where r(A) is the spectral radius of A. In the next section we associate to each pair ({y i }, {x j }) of finite subsets of A 1 as above, a pair of one-sided subshifts, (ℓ {xj} , ℓ ′ {yi} ), which roughly correspond to the operator s ′ and its inverse t ′ . We show that, under certain conditions, the topological entropies of these subshifts are precisely the minimal and maximal inverse temperatures of KMS states. Furthermore we give a criterion for approximating such extremal temperatures with arbitrary (a priori non KMS) tracial states satisfying suitable conditions. In section 7 we introduce a ucp map σ {xj } : T → j x j T x j * implemented by a multiplet {x j } of grade 1 as above, which should be compared with the map S {xj } . The main result of this section is the estimate where the l.h.s. is the Brown-Voiculescu topological entropy [B], [V] of σ {xj } and the second summand at the r.h.s. is the topological entropy, suitable defined, of the set of contractions φ xi,xj : T → x i * T x j of the homogeneous C subalgebra A 0 . Both summands at the r.h.s. of this inequality are necessary. Indeed when A = A 0 ⋊ α Z then the associated subshift is trivial so its entropy is zero, and the above inequality, when combined with monotonicity of topological entropy ( [B], [V]) leads to Brown's result ht A (Ad(u)) = ht A 0 (α), where u ∈ A is a unitary implementing α [B]. Another extreme case is that of the Cuntz-Krieger algebras O A . Now the second summand vanishes and the previous estimate yelds the result by Boca and Goldstein [BG] that ht(σ {xj } ) = ht(ℓ {xj } ) = log(r(A)), see Corollary 7.8. We then focus our attention on those algebras for which ht({φ xi,xj }) = 0 and we show that if the x j 's have pairwise orthogonal ranges ht(σ {xj } ) = h top (ℓ {xj } ). We next discuss new examples of this occurrence among Matsumoto algebras [M] associated to a subshift. Our assumption of orthogonality introduces a certain trivialization to the classical situation. In fact, in this case the algebra of continuous functions on the one-sided subshift ℓ {xj} , together with the endomorphism induced by the left shift epimorphism, sits naturally inside the noncommutative dynamical system (A, σ {xj } ). Therefore monotonicity of topological entropy implies that the topological entropy of ℓ {xj} is ≤ the topological entropy of the noncommutative subshift σ {xj } , thus leading to the equality, see Theorem 7.7. We discuss new examples of this occurrence among the Matsumoto algebras associated to certain non finite type subshifts. In section 8 we investigate the CNT dynamical entropy of σ {xj } . We show that, under the orthogonality assumption where µ is the shiftinvariant probability measure on ℓ {xj } obtained restricting φ. We also find a condition on σ {xj } under which any such µ arises as the restriction of some φ. This enables us to obtain a variational priciple for certain systems (A, σ {xj } ) for which ht({φ xi,xj }) = 0. More precisely, our variational principle asserts, for those systems, the existence of σ {xj } -invariant states of A with respect to which the CNT dynamical entropy equals the Voiculescu topological entropy of σ {xj } . In the last section we establish a closer relationship between KMS states and states with maximal entropy. The main point is that a KMS state ω is to be understood as a quasi-invariant measure for the noncommutative shift σ {xj } , as ω • σ {xj } and ω are equivalent. In classical ergodic theory, measures with this property are called conformal, and play an important role, as they lead to measures with maximal entropy. We thus show an explicit general way of constructing σ {xj } -invariant measures from KMS states. We then consider basic examples, which we may think of as being noncommutative Markov shifts: systems (A, γ) containing some Cuntz-Krieger algebra O A in a way that γ restricts to the canonical gauge action on O A . We show that the σ {xj } -invariant state φ previously derived from a KMS state with maximal entropy restricts, on the algebra of continuous functions on the classical Markov subshift ℓ A , to the unique invariant measure µ with maximal entropy. We thus conclude that if ht({φ xi,xj }) = 0, This yields a generalization of Choda's result [Ch] to the Cuntz-Krieger algebras, and Matsumoto algebras associated to certain non finite type subshifts. The scaling property Recall that a state ω over a C -algebra A endowed with a one-parameter automorphism group α is called a KMS state at inverse temperature β ∈ R if for all a, b in a dense * -subalgebra of A α , the set of entire elements for α (which is in fact a dense * -subalgebra). We will only consider 2π-periodic one-parameter groups, i.e. groups for which α comes from an action γ of T by α t := γ e it . Furthermore, in view of applications to the algebras generated by Hilbert bimodules, we will assume that A is unital and that the group action is full, in the sense explained in the introduction. Then we note that the spectral subspace A k , for positive k, is in fact the linear span of the product set of k copies of A 1 . Moreover, by definition of full C * -dynamical system, there exist, for all n ∈ N, finite subsets {y i } and {x j } of A n such that i y i * y i = I and j x j x j * = I. We define correspondingly, for any tracial state τ on A 0 , We shall usually write δ(τ ) and ǫ(τ ) for δ 1 (τ ) and ǫ 1 (τ ) respectively. 1.1. Lemma Let (A, γ, T) be a full C * -dynamical system, with A unital, and let τ be a tracial state on A 0 . Then δ n (τ ) and ǫ n (τ ) do not depend on the finite subsets {y i } and {x j } of A n satisfying the above relations. If in paricular {y i }, {x j } ⊂ A 1 , one has, for all n ∈ N, Proof We shall only prove the statements relative to {y i }, those relative to {x j } can be proved similarly. Let {z 1 , . . . , z q } ⊂ A n be another multiplet satisfying k z k * z k = I, and write z k = i a k,i y i , where a k,i := z k y i * ∈ A 0 . Then Note that δ n (τ ) ≤ i y i y i * . Furthermore The conclusion follows choosing subsets in A n of the form {y i1 . . . y in } and {x j1 . . . x jn }, where {y i }, {x j } ⊂ A 1 and satisfy i y i * y i = I and j x j x j * = I. Let, for λ > 0, T S λ be the set of tracial states τ on A 0 for which λτ (xy * ) = τ (y * x), x, y ∈ A 1 . (1.2) We shall show that a state of A satisfies the KMS condition w.r.t. α if and only if its restriction to A 0 is an element of some T S λ . We start with the following characterization of T S λ . The following result characterizes the set of tracial states on A 0 which gives rise to KMS states for (A, γ). Let F 0 : A → A 0 denote the projection onto the fixed point algebra obtained overaging over the circle group action. Proposition The maps ω → ω ↾ A 0 , τ → τ • F 0 set up a bijective correspondence between the set of KMS states ω for (A, γ) at inverse temperature β and the set T S e β . Proof If ω is a KMS state at inverse temperature β then the KMS condition (1.1) can be formulated, equivalently, for any pair x, y in the dense linear span of the A k 's. Therefore the restriction τ of ω to A 0 is a tracial state such that ω = τ • F 0 since ω is γ-invariant. Furthermore, if x, y ∈ A 1 , ω(y * x) = ω(xα iβ (y * )) = e β ω(xy * ), therefore τ ∈ T S e β . Conversely, if this condition is satisfied by some tracial state τ on A 0 , then ω := τ • F 0 is an extension of τ to a state on A, and it is not difficult to check that ω(y * x) = ω(xα iβ (y * )) for x, y ∈ A 1 , and hence inductively for x, y ∈ A 1 . . . , and the proof is complete. Remark A simple argument shows that the sequences a n := inf{ǫ n (τ ), τ ∈ T S(A 0 )} and b n := sup{ǫ n (τ ), τ ∈ T S(A 0 )} are respectively supermultiplicative and submultiplicative, therefore the sequences a n 1/n , b n 1/n converge, and lim n a n 1/n = sup a n 1/n and lim n b n 1/n = inf b n 1/n . We next give a criterion of faithfulness for KMS states. 1.5. Proposition Let (A, γ) be a full periodic C * -dynamical system with A unital, and consider the * -monomomorphism α : y i * y i = I and defined by α(a) = (y i ay j * ). If A 0 has no proper closed ideal I such that α(A 0 ) ∩ M p (I) = α(I) (e.g. A is simple), then any KMS state of (A, γ) is faithful. Proof Let τ be the restriction of a KMS state ω to A 0 . Then we have, by (1.2), that xay * ∈ I if a ∈ I. This yields α(I) ⊂ M p (I)∩α(A 0 ). We show the reverse inclusion. Let a ∈ A 0 be such that y i ay j * ∈ I, i, j = 1, . . . , p. Then δ(τ )τ (a * y i * y i ay j * y j ) = τ ((y i ay j * ) * y i ay j * ) = 0. Hence, summing up, we see that a ∈ I, and this shows that α(A 0 ) ∩ M p (I) ⊂ α(I). It follows from our assumption that I = {0}, as, clearly, I = A 0 . Now the canonical conditional expectation F 0 : A → A 0 is faithful, so ω is faithful. We conclude this section recalling from [GP] a criterion for pure infinity of unital C * -algebras, which we shall need in the sequel. We refrain from giving here the proof. We only point out that the arguments essentially go back to Cuntz' proof of pure infinity of O d [C]. Also, the result is a generalization of Rørdam's result (cf. [R]) about pure infinity of crossed products by proper corner endomorphisms. Following [R], we say that a C * -algebra B has the comparability property if B has at least one tracial state, and furthermore a projection e ∈ B is equivalent to a subprojection of f if τ (e) < τ (f ) for all tracial states of B. Assume that our C * -algebra A has a nonunitary isometry S in some A n , n > 0, and also that the fixed point algebra A 0 has the comparability property. Then for all tracial states τ on A 0 one has, by Lemma 1.1, ( 1.6. Theorem [GP] Let (A, γ, T) be a full C * -dynamical system, and assume that A 0 is unital, simple, separable, of real rank zero, and that every M n (A 0 ) has the comparability propery. If, for some n > 0, sup{δ n (τ ), τ ∈ T S(A 0 )} < 1, then A n contains a nonunitary isometry. Furthermore, A is simple and purely infinite. A Perron-Frobenius theorem We associate to each pair of finite subsets Let T ′ , S ′ : A 0 * → A 0 * denote the Banach space adjoints of T and S respectively, and let T (A 0 ) ⊂ A 0 * the Banach subspace of trace functionals. Then one has the following result. Proposition For any finite subset {y i } (resp. {x j }) of A 1 satisfying i y i * y i = I (resp. j x j x j * = I) the associated operator T ′ (resp. S ′ ) leaves T (A 0 ) stable. Let t ′ (resp. s ′ ) be the restriction of T ′ to T (A 0 ). Then t ′ (resp. s ′ ) does not depend on the set {y i } (resp. {x j }) satisfying y i * y i = I (resp. x j x j * = I). Furthermore t ′ and s ′ are inverses of one another. Proof It is easy to check that T ′ transforms trace functionals into trace functionals. Let {y ′ k } be another finite subset of A 1 such that Finally note that by that trace property of τ . Likewise, Note that KMS states of (A, γ) correspond precisely to the tracial state eigenvectors for s ′ (or t ′ ). The following result, which has its own interest, explains why δ(τ ) = ǫ(τ ) −1 when τ corresponds to a KMS state. Proposition is a homeomorphism of T S(A 0 ) endowed with the weak * -topology. KMS states of (A, γ) correspond, as in Prop. 1.3, to fixed points of h. The inverse of h is the map We have: is a positive valued continuous function on a compact set, therefore h is continuous. For the same reason k is continuous on T S(A 0 ), and, by the trace property, h and k are inverses of one another. Our next aim is to look more closely at the spectrum σ(s ′ ) of s ′ . The previous proposition shows that 0 / ∈ σ(s ′ ) = σ(t ′ ) −1 . So we can define the inner and outer spectral radius of s ′ : We give some estimates for r min (s ′ ) and r max (s ′ ). Proposition Let {y i }, {x j } ⊂ A 1 satisfy i y i * y i = I and j x j x j * = I. Then one has Taking the n-th root and passing to the limit, one gets the spectral radius of T : Similarly, one has The proof is completed recalling that s ′ is the restriction of S ′ to a closed subspace, so r max (s ′ ) ≤ r(S ′ ) = r(S) and similarly We next show that the inner and outer spectral radii of s ′ correspond to inverse temperatures of KMS states, or, in other words, that they are in the point spectrum of s ′ , with corresponding positive eigenvalues. The fact that the outer spectral radius is in the point spectrum was first proved in [MWY] for the Matsumoto C * -algebras associated with subshifts [M]. The key point in our situation is that one needs to consider the trace functionals of A 0 endowed with the order structure which arises when we extend such traces to normal traces on the enveloping von Neumann algebra. We anticipate the following, possibly known, lemma. Lemma Let n φ n be a series of normal linear functionals on a von Neumann algebra M weakly convergent to φ. If each of the the absolute values \|φ n \| is tracial then \|φ\| ≤ n \|φ n \|. Proof Let τ be a positive tracial linear functional, then \|τ (xy)\| ≤ \|\|y\|\|τ (\|x\|) (see for example [T]). Consider the polar decompositions Then, for a positive x, we have We are now in the position of proving our main result of this section. 2.5. Theorem Let (A, γ, T) be a full C * -dynamical system, and assume that A is unital, and that A 0 has a tracial state. Then r min (s ′ ) and r max (s ′ ) are eigenvalues of s ′ with corresponding tracial state eigenvectors. Proof We first show that r max (s ′ ) is a spectral value for s ′ and then that it is in fact an eigenvalue with a tracial state eigenvector. A similar argument will prove that r(t ′ ) is an eigenvalue for t ′ = s ′ −1 with a tracial state eigenvector. By the uniform boundedness theorem, there exists a sequence {z n } of complex numbers such that \|z n \| → r max (s ′ ) + and R(z n )τ 0 → ∞ for some τ 0 ∈ T (A 0 ), where R(z) is the resolvent of s ′ in z. Since T (A 0 ) is linearly spanned by its tracial states, we may assume that τ 0 is a tracial state. Consider, for \|z\| > r max (s ′ ), the Neumann series: (2.1) By the previous lemma, on the enveloping von Neumann algebra of A 0 , so R(\|z n \|)τ 0 → ∞, and this shows that r max (s ′ ) ∈ σ(s ′ ). (2.1) also shows that R(λ)τ 0 is a nonzero positive functional for λ > r max (s ′ ), hence, arguments similar to those of Lemma 3.1 in [MWY] prove that is a sequence of tracial states such that every weak * -limit point of it is a tracial state eigenvector with eigenvalue r max (s ′ ). The previous theorem can be considered as an analogue of the Perron-Frobenius theorem for matrices with nonnegative entries. 2.6. Corollary Let (α, R) be a 2π-periodic one-parameter automorphism group of a unital C * -algebra A, such that the induced T-action γ is full. If s ′ is defined as above, relatively to γ, then the set of inverse temperatures of KMS states is a closed subset of the interval [log(r min (s ′ )), log(r max (s ′ ))] containing the extreme points. Proof The subset of T S(A 0 ) corresponding to KMS states is weakly * -compact by Prop. 2.2, furthermore the map ǫ : T S(A 0 ) → R + defined at the beginning of section 1 is weakly * -continuous. It follows that the set of elements of the form log(ǫ(τ )), when τ ranges over all tracial states on A 0 corresponding to KMS states, is compact. Now this set is precisely the set of possible inverse temperatures by Prop. 1.3. The rest follows from the previous Theorem. A KMS state of (A, α) at inverse temperature β min := log(r min (s ′ )) or β max := log(r max (s ′ )) will be called extremal. Let β be the inverse temperature of a KMS state, and set, as in the previous section, Then for all n, a n 1/n ≤ e β ≤ b n 1/n , so lim n 1/n log(a n ) ≤ β ≤ lim n 1/n log(b n ). It is then natural to ask for which tracial states τ , the sequence 1/n log(ǫ n (τ )) approximates the maximal or the minimal inverse temperature. In section 5 we shall give a sufficient condition. We conclude this section with the discussion of two examples known in the literature. The first example, arising from ergodic theory, shows that in general, at a fixed inverse temperature, there may be more than one KMS state. Example Let (X, T ) be a topological dynamical system: X is a compact metric space endowed with a homeomorphism T . We suppose that X is not a finite set. Then it is well known that the C * -algebra Tracial states on C(X)⋊ αT Z are in one-to-one correspondence with T -invariant probability measures on X, while there is no nontracial KMS state on (A, γ). The operator s ′ therefore has spectrum contained in the unit circle. However, s ′ is the Banach space adjoint of α T , so its spectrum is the same as that of α T which must be equal to T by simplicity of A [OP]. There is an important example, due to Furstenberg, of a minimal analytic diffeomorphism T of T 2 with nonunique invariant measures (see, e.g, [Ma]), which thus leads to an example of nonuniqueness of tracial states on the simple crossed product C * -algebra C(T 2 ) ⋊ αT Z. The next example shows that the set of inverse temperatures can in general be an arbitrary closed subset of R. Example In [BEH] Bratteli, Elliott and Herman construct an example of a simple C * -algebra B endowed with T-action for which the set of possible inverse temperatures can be any arbitrary closed subset F of R ∪ {+∞, −∞}. For each temperature the corresponding state is unique. More in detail, B is obtained by cutting down the crossed product A ⋊ α Z of an AF-algebra by some projection P in A. If neither +∞ nor −∞ belongs to F , A itself is simple, and this implies that the T-action is full since P is a full projection. If moreover F ⊂ (0, +∞), one can choose α so that α(P ) < P , see [BEH], hence α(B 0 ) ⊂ B 0 . Then ρ := α ↾ B 0 is a proper corner endomorphism of B 0 , and one has B = B 0 ⋊ ρ N. Now by a result of Rørdam [R], B is purely infinite. KMS states of the Pimsner algebras In this section we discuss an application of the results of the previous section to the C * -algebra O X associated to a Hilbert C * -bimodule X over a C * -algebra B. We refer the reader to [P] for the construction of O X . We just recall that both X and B embed isometrically respectively as a Hilbert bimodule in and a C * -subalgebra of O X . We shall always assume that X is finite projective and full, and that B is unital. Therefore any finite basis x ∈ X. Therefore we can conclude that (O X , γ) is a full periodic C * -dynamical system. We start proving that systems of this form are typical examples, in the sense that we can always easily associate to any unital, full, periodic C * -dynamical system (A, γ), a finite projective Hilbert C * -bimodule X such that A = O X and γ is the canonical T-action. We should note, however, that the Hilbert bimodule X and its coefficient C * -algebra are, in general, not unique. In fact our construction leads to a maximal Hilbert bimodule. In applications, it may be more convenient to start with smaller Hilbert bimodules. It is well known the case of Cuntz-Krieger algebras discussed in [P], where the coefficient algebra is finite-dimensional. In section 4 we shall discuss the more general situation of Matsumoto algebras, and we will construct natural minimal generating Hilbert bimodules. 3.1. Theorem Let (A, γ) be a full C * -dynamical system over T and assume that A is unital. Then there exists a full finite projective Hilbert C * -bimodule X over a unital C * -algebra B such that (A, γ) can be identified with O X , endowed with its canonical gauge action. Proof Choose finite subsets {y i } and {x j } of A 1 such that i y i * y i = I and X be the natural conditional expectations. Since ϕE = F 0 ϕ and E is faithful, ϕ is injective by a well known argument. Thus ϕ is the desired isomorphism. A KMS state on (O X , γ) at some inverse temperature log(δ) ∈ R restricts to a tracial state τ on the coefficient algebra B satisfying, for a ∈ B, τ ( i < x i , ax i >) = δτ (a), with {x i } a right basis of X. We show that conversely any such trace extends to a KMS state. Lemma with δ > 0, extends uniquely to a KMS state on O X at inverse temperature log(δ). This extension is faithful if τ is faithful. Proof We first prove uniqueness. A KMS state for (O X , γ) is determined by its restriction to the homogeneous C * -subalgebra, and, by the trace-scaling property of the operator S {xj} on that subalgebra, it is in fact determined by its restriction to the coefficient algebra B. Conversely, if one is given a tracial state τ 0 on B as required then it is easy to check that is a sequence of tracial states (faithful if τ 0 is faithful) such that τ n+1 ↾ L(X n ) = τ n , which thus gives rise to a tracial state τ on the homogeneous subalgebra positively scaled by S {xj } . Therefore τ extends to a KMS state on O X at inverse temperature log(δ). We now apply the results of the previous section to the Pimsner C * -algebras. 3.3. Theorem Let B be a unital C * -algebra with a tracial state, and let X be a full finite projective Hilbert C * -bimodule over B. Assume that for every tracial state τ on B, and any basis Then O X has a KMS state. (1) Let s ′ : Then the set of possible inverse temperatures is a closed subset of [log(r min (s ′ )), log(r max (s ′ ))] containing the extreme points. ( is weakly * -continuous, so, by the Schauder-Tychonov fixed point theorem, there is a tracial state τ on B such that We can now extend such a τ to a KMS state on (O X , γ), by the previous lemma. (1) follows from Corollary 2.6. Since O X is T-simple, O 0 X has no ideal of the kind described in our faithfulness criterion, Prop. 1.5, hence (2) follows. (3)-(5) follow from Corollary 1.4. Finally, if B has a unique tracial state then so does O X 0 since it is an inductive limit of C * -algebras stably isomorphic to B itself. Therefore O X has a unique KMS state. Examples of full dynamical systems arising from subshifts, self-similar sets and noncommutative metric spaces In this section we continue our discussion of examples of full C * -dynamical systems obtained via Pimsner's construction. We start considering two different examples of Hilbert bimodules both described by families of * -endomorphisms on commutative C * -algebras, arising respectively from symbolic dynamics and fractal geometry. We shall also discuss a generalization of the latter example to noncommutative metric spaces. Subshifts in symbolic dynamics and Matsumoto algebras We recall the construction of the Matsumoto algebra O Λ associated with a two-sided subshift Λ, [M]. Fix a finite discrete set Σ = {1, 2, ..., d}, and let Σ Z be the infinite product space endowed with the product topology. We denote by σ the shift homeomorphism on Σ Z defined by (σ(x)) i = x i+1 . For a shiftinvariant closed subset Λ of Σ Z , the topological dynamical system (Λ, σ ↾ Λ ) is called a subshift. We denote by Λ + the set of one-sided sequences x ∈ Σ N such that x appears in Λ. For example, Σ N = Σ Z + . We shall still denote by σ the left shift epimorphism of Σ N . The dynamical system (Λ + , σ ↾ Λ+ ) is called the onesided subshift associated to Λ. A finite sequence µ = (µ 1 , . . . , µ k ) of elements µ j ∈ Σ is called a word. We denote by \|µ\| the length k of µ. For k ∈ N, let Λ k = {µ\|µ is a word with length k appearing in some x ∈ Λ}, Λ l = l k=0 Λ k and Λ * = ∞ k=0 Λ k , where Λ 0 denotes the set constituted by the empty word. Let {e 1 , ..., e d } be an orthonormal basis of a d-dimensional Hilbert space H = C d . Let F 0 be the one dimensional space CΩ spanned by a normalized vector Ω, called the vacuum vector, and let F k be the Hilbert space spanned by the vectors e µ = e µ1 ⊗ · · · ⊗ e µ k for µ = (µ 1 , . The unital C * -subalgebra T Λ of the algebra of bounded linear operators on F Λ generated by {T i \|i = 1, ..., d} is called the Toeplitz algebra associated with Λ, and contains the algebra K(F Λ ) of compact operators on F Λ . The Matsumoto algebra O Λ associated with the subshift Λ is the quotient algebra It is generated by the quotient image {S i \|i = 1, ..., d} of {T i \|i = 1, ..., d}. The unitary representation of T on F Λ defining the grading implements an automorphic action of T on T Λ leaving K(F Λ ) stable. We thus obtain an automorphic T-action γ on O Λ such that and with no zero rows or columns: Then each ρ i preserves C(Λ A+ ). For Markov subshifts, Matsumoto's construction yields the Cuntz-Krieger algebras, see [M],: , the Hilbert bimodules X and Y are isomorphic. Now a standard argument shows that We shall choose, to this aim, the smallest such C * -algebra, which is related to the Krieger left cover of a sofic subshift and the past equivalence relation considered by Matsumoto. Let the following notion of past equivalence relation. Two points x and y ∈ Λ + are called l-past equivalent, x ∼ l y, if Λ l (x) = Λ l (y). The corresponding set of equivalent classes is denoted by Ω l := Λ + / ∼ l . For µ ∈ Λ l , if x ∼ l y, then q µ (x) = q µ (y). Thus q µ defines a functionq µ ∈ C(Ω l ). The set {q µ ∈ C(Ω l ); µ ∈ Λ l } separates the points in Ω l , thus it generates C(Ω l ). A l (Λ + ) is precisely the set of functions in ℓ ∞ (Λ + ) which have the same value on each l-past equivalent class and we have an isomorphism between A l (Λ + ) and C(Ω l ). We see directly that the , the Hilbert bimodules X and Y are isomorphic by the identification of B with A Λ . The universality of the Matsumoto algebra and the Pimsner algebra immediately shows that Contractions of compact metric spaces We next dicuss an example associated with a self-similar set in fractal geometry. Let Ω be a (separable) complete metric space and let {γ 1 , ..., γ d } be a finite family of nonzero proper contractions of Ω with Lipschitz constants c i = Lip(γ i ) < 1. We assume that d ≥ 2. Then there exists a unique nonempty compact set K ⊂ Ω satisfying the (exact) invariance condition The above invariance condition shows that the compact set K is self-similar in a weak sense. For example the Cantor set, the Koch curve and the Sierpinski gasket are typical examples of self-similar sets. We refer the reader to the book of Hutchinson [H] for more information on fractal geometry. The topological dimension of K is dominated by the Hausdorff dimension of K, and the Hausdorff dimension of K is dominated by the similaritiy dimension of K, which is a finite number D satisfying i c D i = 1. Thus K has a finite topological dimension. Consider the C * -algebra B = C(K) and the canonical Hilbert right B-module In [H] Hutchinson shows that there exists a unique regular Borel probability measure µ on K satisfying, for any measurable set F , Consider the trace τ 0 on B corresponding to the probability measure µ. Then Since < x i , x i >= 1 for i = 1, . . . , d , O X has a KMS state at the inverse temperature β if and only if β = log d. Moreover the uniqueness of the probability measure implies that the corresponding KMS state is also unique. We shall show that the algebra O X is in fact the Cuntz algebra O d . Before proving this, we study a more general situation to include standard d-times around embeddings. Let K be a compact metric space. Consider the C * -algebra B = C(K) and the state space S of B. Let Lip(K) be the space of Lipschitz functions, and let Lip(f ) denote the Lipschitz constant of f ∈ Lip (K). In [H] Hutchinson considers the following metric L on S: Then (S, L) is a complete metric space, and the topology defined by L is precisely the weak * -topology of S. Consider the canonical Hilbert right B-module X = B d and any injective unital * -homomorphism φ : Then the Pimsner C * -algebra O X is isomorphic to the universal C * -algebra generated by B = C(K) and the Cuntz algebra O d satisfying the relations aS j = i S i φ ij (a) for a ∈ B and i, j = 1, . . . , d. Proposition In the above situation, let Ψ be the unital positive map If the Banach space adjoint Ψ * induces a proper contraction on S with respect to the metric L, then O 0 X has a unique tracial state. Moreover O X has a KMS state at the inverse temperature β if and only if β = log d and the corresponding KMS state is also unique. Proof We identify L(X n ) with B ⊗ M d n (C). Using the commutation relation, it is easy to see that the inclusion map Φ n : where α and β run the set {1, . . . , d} n of words with length n. We remark that the present situation is similar to that of the Cuntz-Krieger algebras associated to aperiodic matrices. In fact, for any state ω ∈ S, Ψ * n (ω) converges to the unique ω 0 in S with respect to L. This resembles the Perron-Frobenius Theorem for aperiodic matrices. Example In the fractal case, each proper contraction γ i induces an en- i is a proper contraction on S with respect to the metric L. Example We next study the example of standard d-times around embed- where (u t ) t is a continuous path of unitaries in M d (C) such that u 0 = I and such that u 1 is the unitary matrix corresponding to the operator taking vectors e 1 , . . . , e d of the canonical basis of C d to e d , e 1 , . . . , e d−1 respectively. We regard φ as a map φ : B → L B (X) = B ⊗M d (C). Then Ψ * = 1 d i φ i * induces a proper contraction on S with respect to the metric L. In fact We assume that d = 2 for the simplicity of notation. For f ∈ Lip(T) and x, y ∈ T, choosing carefully the nearest pairs between { x 2 , x+1 Thus L(Ψ * (ϕ 1 ), Ψ * (ϕ 2 )) ≤ 1 2 L(ϕ 1 , ϕ 2 ). 4.8. Remark One can easily show, using known results, that under suitable circumstances O X is simple and purely infinite. Indeed, assume that K is totally disconected or connected and has a finite topological dimension. If O 0 X = lim − →n L(X n ) is simple, then O 0 X is of real rank zero by [BDR], since O 0 X has a unique trace. By a result of Martin and Pasnicu [MP], O 0 X has the comparability property on every matrix algebra. Thus we can apply a result by [GP] (Theorem 1.6) and conclude that the O X is simple and purely infinite. For example, in the case of a standard d-times around embeddings all the assumptions are satisfied. In fact O 0 X = lim − →n L(X n ) is a Bunce-Deddens algebra. Note that we have naturally embedded an AT algebra into a purely infinite simple C * -algebra. Again, in the fractal case , it is easy to show that O 0 X is simple. So we can apply the preceding argument. However, it is not difficult to show that in this case O X is canonically isomorphic to the Cuntz algebra O d (a fact which will be later generalized to noncommutative metric spaces): We identify L(X n ) with C(K, M d n (C)). Then the inclusion map Φ n : L(X n ) → L(X n+1 ) is described by block diagonal matrices (Φ n (f ))(t) = diag(f (γ 1 (t)), . . . , f (γ d (t))). By the uniform continuity of f , Φ n+k,n (f ) is approximated by a constant matrix up to ε for a sufficient large k. Thus O 0 X is a UHF algebra M d ∞ and O X is exactly the Cuntz algebra O d generated by the original operators {S 1 , . . . , S d }. Contractions of noncommutative metric spaces The preceding argument suggests a generalization to noncommutative metric spaces introduced by Connes in [Co]. Our setting will be the following. Let A and B be unital C * -algebras. Suppose that B is a Banach bimodule over A. Let δ : A ⊃ Dom(δ) → B be a densely defined * -derivation with kerδ = CI. Let S be the state space of A. Consider the following metric L on S: The metric is allowed to take the value ∞. In [RiI], Rieffel considers the question of whether the metric topology agrees with the underlying weak * topology on the state space. His setting is, however, more general, as he works with normed vector spaces endowed with seminorms not necessarily arising from * -derivations. We assume that {a ∈ Dom(δ) ; δ(a) ≤ 1}/CI is bounded in A/CI. (4.1) By Proposition 1.6 in [RiI] this condition is equivalent to the fact that the metric L on S is bounded. Let {φ 1 , . . . , φ d } be a finite family of unital * -endomorphisms on A, with d ≥ 2. Recall that the crossed product C * -algebra C * (A; φ 1 , . . . , φ d ) of A by {φ 1 , . . . , φ d } is the universal C * -algebra generated by the image of a C homomorphism π : A → C (A; φ 1 , . . . , φ d ) and the Cuntz algebra O d with the generators S 1 , . . . , S d satisfying the relations π(a)S i = S i π(φ i (a)) for a ∈ A and i = 1, . . . , d. We note that π is isometric if and only if ∩ i kerφ i = 0. In this case the crossed product C Proposition In the above setting, assume that the restrictions γ i of the Banach space adjoint φ * i : A * → A * to the state space S of A are proper contractions with respect to L. Then the endomorphism crossed product C algebra C (A; φ 1 , . . . , φ d ) is canonically isomorphic to the Cuntz algebra O d and has a unique KMS state at inverse temperature log d. Proof Let c be the maximum of the Lipschitz norms c i = Lip(γ i ), i = 1, . . . , d. We shall show that π(A) is included in the canonical UHF subalgebra M d ∞ of the Cuntz algebra O d . For any a ∈ Dom(δ) and ε > 0, there exists n ∈ N such that c n diam(S, L) δ(a) ≤ ε. Fix a state ω 0 ∈ S. Consider the diagonal matrix t = diag(ω 0 (φ α (a))) α ∈ M d n (C). Then for any state ω ∈ S, we have Remark M. Rieffel has kindly pointed out to us that in the proof of the previous Proposition we never use the fact that L comes from a * -derivation, but rather only that it is a seminorm satisfying the boundedness condition (4.1). Seminorms of this kind were studied in [RiII]. Let δ : A ⊃ Dom(δ) → B be the densely defined * -derivation of De Leeuw, given by where Dom(δ) is the set of Lipschitz functions in A. Then kerδ = CI. Let α be the * -endomorphism on A defined by α(f )(x) = f ( x 2 ). Then the restriction γ of the Banach space adjoint α * : A * → A * to the state space S is a proper contraction with respect to L. Therefore the endomorphism crossed product C * -algebra C * (A; α, α) is isomorphic to the Cuntz algebra O 2 . KMS states of Pimsner C * -algebras associated to Cuntz-Krieger bimodules In this section we illustrate Theorem 2.5 by examples. We shall discuss some situations where there is a unique KMS state, or more generally, where the set of KMS states can be easily characterized. The inspiring example is that of the Cuntz-Krieger algebras that we discuss here below. Some of the following facts are well known in terms of path algebras in subfactor theory. KMS states of Cuntz-Krieger algebras Let O A be the Cuntz-Krieger algebras associated to a matrix A = (a ij ) ∈ M d ({0, 1}). A KMS state ω for the canonical circle action restricts to a tracial state τ on the f.d. commutative subalgebra A generated by the ranges P 1 , . . . , P d of the generating partial isometries S 1 , . . . , S d . Let λ = (λ 1 , . . . , λ d ) ∈ R + n be defined by λ i = ω(P i ). Since ω is normalized, we have λ i = 1. The scaling property s ′ (τ ) = ǫ(τ )τ says, when checked on A, that λ is a nonnegative eigenvector of A, and hence, when A is irreducible, it is the unique normalized Perron eigenvector for A by the Perron-Frobenius Theorem [G]. Let us analyse more in detail the structure of the Banach space T (O A 0 ) and the spectrum of the operator s ′ . Let L r , r ≥ 1, denote the unital finite-dimensional C subalgebra of O A generated by elements of the form S i1 . . . S ir P k (S j1 . . . S jr ) . Set L 0 = CP 1 + · · · + CP d . Then the set of minimal central projections for L r is {σ r (P k ), k = 1, . . . , d}, where σ is the canonical endomorphism of L 0 Therefore d r,k is the sum of the entries in the k-th column of A r : The (j, i)-entry of inclusion matrix of L r ⊂ L r+1 can be computed by looking at the projection σ r (P i )σ r+1 (P j ) = σ r (S i P j S i * ) which 0 when a i,j = 0, otherwise it is the sum d r,i minimal projections of L r+1 σ r+1 (P j ). Thus the inclusion matrix of L r ⊂ L r+1 is A t . A tracial state on O A 0 is described by a sequence of positive traces {τ r }, r ≥ 0 on L r such that τ 0 is normalized and τ r+1 ↾ Lr = τ r . Therefore one needs to assign a sequence of nonnegative column vectors (t r ) ∈ R + d which will be the values that τ r takes on the minimal projections of L r . The coherence relations require that A(t r+1 ) = t r , r ≥ 0 while the positivity and normalization properties translates into: The latter implies, as expected, normalization of each τ r : Removing positivity and normalization, but requiring instead a norm bound for the sequence (t r ), one finds that T (O A 0 ) is described by The operator s ′ acts as: while its inverse is t ′ ((t r ) r≥0 ) = (t 1 , t 2 , . . . ). Proposition If is linearly spanned by elements of the form (λ −r t 0 ) r≥0 where t 0 is an eigenvector of A with eigenvalue λ and \|λ\| = r(A). Furthermore Proof Let t 0 be an eigenvector for A with eigenvalue λ such that \|λ\| = r(A). Set t r := λ −r t 0 , so that At r+1 = t r . We have: where E 0 is the rank one orthogonal projection onto the span of the Perron eigenvector. It follows that (t r ) ∈ T (O A 0 ). Furthermore (t r ) is also an eigenvector of s ′ with the same eigenvalue. The same argument shows that if t 0 ∈ T (O A 0 ) were an eigenvector with eigenvalue λ such that \|λ\| < r(A), and t r is defined as above, then A r \|t r \| 2 would be unbounded, so that (t r ) does not define an element of T (O A 0 ). With similar arguments one sees that T (O A 0 ) is linearly spanned by vectors of the form (λ −r t 0 ) where \|λ\| = r(A). We show that if either λ / ∈ σ(A) or λ ∈ σ(A) but \|λ\| < r(A) then s ′ − λ is invertible. We start assuming that λ / ∈ σ(A). Then s ′ − λ is clearly injective. We show that it is also surjective. For any matrix B with complex entries let B + stand for the matrix with entries the absolute values of the corresponding elements of B, and let M B be the maximum of the absolute values of its entries. Then which shows that (t r ) ∈ T (O A 0 ) and (s ′ − λ)(t r ) = v r . Assume now that λ ∈ σ(A) − {0} but \|λ\| < r(A). We have already noted that s ′ − λ is injective. Furthermore since for each (t r ) ∈ T (O A 0 ), any t r belongs to the range of A − λ, with similar arguments one shows that s ′ − λ is surjective. Remark If A is aperiodic, then the homogeneous subalgebra of O A has a unique trace. More generally, if one drops the assumption that A is symmetric, then, with a more extensive use of Perron-Frobenius theory, one can still show that eigenvectors corresponding to nonmaximal eigenvalues do not appear in the point spectrum of s ′ , hence our result shows that σ(s ′ ) ⊂ {λ : \|λ\| = r(A)}. Note also that if we more generally start with a reducible matrix A, then we are in a situation of nonuniqueness of KMS states for O A (corresponding to the minimal and maximal Perron eigenvalues of A). KMS states as Markov traces arising from inclusions of finite algebras with finite Jones index Let N ⊂ M be an inclusion of II 1 -factors with finite index or of finitedimensional C * -algebras such that Z(N ) ∩ Z(M ) = C. Let, in the latter case, A be the inclusion matrix. Let τ be a faithful tracial state on M , and consider the unique τ -preserving conditional expectation Endow X = M with the C * -bimodule structure over N as follows. The structure of N -bimodule is defined by left and right multiplication, while the Nvalued inner product is < x, y > N := E τ (x * y). Then X is full and finite projective as a right N -module ( [GHJ]). It is not difficult to check that is the the Jones tower. A KMS state at inverse temperature β for O X corresponds precisely to a Markov trace for the tower, which is unique, and one has β = log([M : N ]). See [K]. If N ⊂ M are finite factors, each term of the tower is a finite factor, hence its trace space is one dimensional, and spanned by the Markov trace. So dimT (O X 0 ) = 1, and s ′ acts multiplying by [M : N ]. If N ⊂ M are finite-dimensional C * -algebras, the inclusion matrix of L N (X r ) ⊂ L N (X r+1 ) is A t A, which is symmetric and irreducible ( [GHJ]). It is not difficult to show, with arguments similar to those of the previous example, that T (O X 0 ) is again linearly spanned by traces corresponding to eigenvectors of A t A with maximal eigenvalue. KMS states of Pimsner algebras associated with Cuntz-Krieger bimodules After these motivating examples, we consider, more generally, systems of the form (O X , γ), where X is what we call a Cuntz-Krieger Hilbert C * -bimodule and γ is the canonical gauge action. Such Hilbert bimodules, and simplicity of the corresponding C * -algebras O X , have been considered in [KPW]. Consider d ≥ 2 unital simple C * -algebras A 1 , . . . A d and a matrix A = (a i,j ) ∈ M d ({0, 1}) with no row and no column identically zero. Let, for any pair of indices i, j such that a i,j = 1, X i,j be a full, finite projective A i -A j Hilbert bimodule, and let X = ⊕ i,j:ai,j =1 X i,j be endowed with the natural structure of Hilbert bimodule over A := A 1 ⊕· · ·⊕A d . Then since no row of A is zero, left A-action is faithful, and since no column of A is zero, X is full. Clearly X is finite projective as a right module. We assume that there is a system of tracial states τ 1 , . . . , τ n on A 1 , . . . A n respectively satisfying, for each pair of indices for which a j,k = 1, for some λ j,k > 0. Here {x r k,j } r if a basis of X j,k . We set λ j,k = 0 if a j,k = 0. We will call {τ i } a coherent set of traces. Note that (λ j,k ) is irreducible precisely when A is. If each A j has a unique tracial state τ j , the set {τ j } is coherent. This is indeed the case of Cuntz-Krieger algebras. Let P j be the identity of A j . Recall from [KPW] that A ′ ∩ O X has a unique unital endomorphism σ such that σ(a)x = xa, x ∈ X, a ∈ A ′ ∩ O X . Recall also that if A is irreducible, O X is simple [KPW], and if A is aperiodic, A is separable and has real rank zero and all M n (A j ) have the comparability property, then O X is simple and purely infinite (cf. Theorem 1.6). For all r ≥ 1 L A (X r ) is a finite direct sum of unital simple C * -algebras, and its minimal central projections are σ r (P 1 ), . . . σ r (P d ). One has L A (X r )σ r (P j ) = L Aj (X r P j ), where X r P j is regarded as an A-A j Hilbert bimodule. Let τ 1 , . . . , τ d be a coherent choice of tracial states on A 1 , . . . , A d respectively. Let {u i (r),j } be a basis of X r P j . Then the positive functional is nonzero, tracial and independent of the basis. Let ǫ r (τ j ) denote its norm. After normalization, we get a tracial state T j (r) on L(X r P j ). If A j has a unique tracial state, T j (r) is the unique tracial state of L(X r P j ). Consider a tracial state τ r on L(X r ) which restricts to a multiple of T j (r) on L(X r P j ), and let t j (r) = τ (σ r (P j )), so t j (r) ≥ 0 and j t j (r) = 1. Then τ r restricts to τ r−1 if and only if for all a ∈ L(X r−1 P j ), and all j, So τ r restricts to τ r−1 if and only if . Then a solution is obtained choosing for v 0 the Perron-Frobenius eigenvector of the nonnegative matrix (λ j,k ) with the normalization v k 0 = 1, and iteratively v r = λ −1 v r−1 , where λ is a positive eigenvalue of (λ j,k ). Note that if A is aperiodic, then (λ j,k ) is aperiodic as well, so such a v 0 is the only possible solution. One can easily check that the tracial state τ thus obtained on O X 0 satisfies and therefore gives rise to a KMS state of O X . Summarizing, we have proved the following result. Theorem Let A = (a i,j ) ∈ M d (0, 1) be an irreducible matrix, A 1 , . . . , A d unital simple C * -algebras with a nonzero trace, and let, for any pair of indices for which a i,j = 1, X i,j be a full, finite projective, Hilbert A i -A j bimodule. Consider X = ⊕ i,j:ai,j =1 X i,j as a Hilbert bimodule over A 1 ⊕ · · · ⊕ A d . Then any normalized Perron-Frobenius eigenvector of the irreducible matrix (λ i,j ) associated to a coherent system of tracial states on A 1 , . . . , A d defines, as described above, a KMS state of O X at inverse temperature β = log r((λ i,j )). In particular, if each A j has a unique tracial state, then there is a unique system of coherent tracial states, and the corresponding KMS state is the unique KMS state of O X . Note that if each A j has a unique trace and A is aperiodic, then O X 0 has a unique trace, which correponds necessarily to the unique KMS state. Inverse temperatures and topological entropy The aim of this section is to establish a relationship between inverse temperatures of extremal KMS states and the topological entropy of certain subshifts naturally associated to (A, γ). Let {y i } and {x j } be finite subsets of A 1 \{0} satisfying i y i * y i = I and In the case of the Matsumoto C * -algebras O Λ , h(S) is the topological entropy of the shift homeomorphism σ ↾ Λ , and it was shown to coincide with the maximal inverse temperture of certain KMS states in [MWY]. Note that since ( y i1 . . . y in 2 ) 1/n ≥ 1 for all n and y i ≤ 1, for all i, and similarly, Note however that if A is a crossed product C * -algebra by a single automorphism α then h(α) = h(α −1 ) = 0. More generally, since the sequences defining h(T ) and h(S) converge to their greatest lower bounds, we see that these are positive if and only if one has repectively ( y i1 . . . y in 2 ) 1/n ≥ 1 + ε ( x j1 . . . x in 2 ) 1/n ≥ 1 + ε for all n and some ε > 0. We now associate to a fixed finite set of nonzero elements {x j } ⊂ A 1 such that x j x j * = I, a one-sided subshift ℓ {xj} , defined as follows. Set Σ := {1, . . . , d}, where d = Card{x j }. Then Clearly ℓ {xj} is a closed subset of Σ N mapped onto itself by the left shift homomorphism: Notice that if (A, γ) does not result from a crossed product by an automorphism, or a proper corner endomorphism, d ≥ 2. Replacing the T-action γ by the action z ∈ T → γ ′ z := γ z −1 , we see that we also have, for any finite subset {y i } ⊂ A 1 \{0} such that i y i * y i = I, a one-sided subshift: where Σ ′ is the set of the first d ′ := Card{y i } positive integers. We also introduce the following two-sided subshifts: and Λ ′ {yi} = {λ ∈ Σ ′ Z : y λr+s . . . y λr = 0, r ∈ Z, s ∈ N}. Remark Even though it would seem more convenient to work with two-sided subshifts, we should point out that these may be rather small, in the sense that a finite word (λ 1 , . . . , λ r ) occurring, e.g., in ℓ {xj } does not necessarily extend to a word in Λ {xj} . The following simple example well describes the situation. Consider the C * -algebra A = M 2 (C) ⊗ C (T), and define the following 2πperiodic automorphic action α of R: and Let e i,j , i, j = 1, 2, be a system of matrix units for M 2 (C), and define x 1 = e 1,2 ⊗I, x 2 = e 2,2 ⊗u, y 1 = e 1,2 ⊗I, y 2 = e 1,1 ⊗u, where u(z) = z, z ∈ T. Then all the above elements are in A 1 , and satisfy x 1 x 1 * + x 2 x 2 * = I, y 1 * y 1 + y 2 * y 2 = I, so (A, α) is a full C * -dynamical system. Note that x 1 2 = 0, x 1 x 2 = 0, x 2 x 1 = 0, x j x j * = I and let ℓ {xj} and Λ {xj} be the associated one-sided and two-sided subshifts respectively. If j x j * x j is invertible then Λ {xj} = ∅. Furthermore An analogous statement holds for ℓ ′ {yi} and Λ ′ {yi} . Proof Since j x j x j * = I, for any (i 1 , . . . , i r ) such that x i1 . . . x ir = 0 there is an i r+1 such that x i1 . . . x ir x ir+1 = 0, and therefore there is a sequence (i n ) n≥1 such that x i1 . . . x in = 0 for all n ∈ N. To complete the proof relative to the set {x j } it is now clear that it suffices to show, for any such (i n ), the existence of i 0 ∈ Σ such that x i0 x i1 . . . x in = 0 for all n ≥ 0. If this were not the case, for all k ∈ Σ there would exist n k such that x k x i1 . . . x in k = 0. Letting n = max{n k , k ∈ Σ}, we must have x k x i1 . . . x in = 0 for all k ∈ Σ, and therefore, k x k * x k being invertible, x i1 . . . x in = 0. This is now a contradiction. The statement relative to the set {y i } can be proved similarly. In particular, if A = (a ij ) is a {0, 1}-matrix with no zero row or column, then the generating partial isometries {S i } of the Cuntz-Krieger algebra O A satisfy both generally, if Λ is a nonempty subshift of Σ Z then the canonical set of generating partial isometries {S i } of the Matsumoto C * -algebra O Λ still satisfy the above conditions (see 4.1) and we have also in this case Λ {Si} = Λ and ℓ {Si} = Λ + . Another example is provided by the algebras generated by certain Cuntz-Krieger bimodules X = ⊕ (i,j):ai,j =1 X i,j as described in section 5. More precisely, if X i,j is the Hilbert bimodule defined by a unital * -isomorphism φ i,j : A i → A j , then one can define S i,j to be the identity of A j regarded as an element of X i,j . So S i = j:aij =1 S i,j are partial isometries of O X 1 satisfying the Cuntz-Krieger relations with respect to A = (a ij ). Therefore Λ {Si} is again the two-sided Markov subshift defined by the matrix A = (a i,j ). The example arising from fractal geometry discussed in section 4 is in the same spirit, in that the natural basis of the generating module are generators of a Cuntz algebras, so the associated one or two-sided subshifts are full. Note that all the examples above discussed have in common the fact that there is a multiplet {x i } ⊂ A 1 such that i x i x i * = I consisting of elements with pairwise orthogonal ranges (and therefore they are necessarily partial isometries). We start by establishing general estimates for the extremal inverse temperatures using the topological entropies of the associated subshifts. Recall [DGS] that for a one-sided (or two-sided) subshift (ℓ, σ ↾ ℓ ) the topological entropy can be computed as where θ n (ℓ) is the cardinality of the set ℓ r of distinct words of length n occurring in ℓ. Proposition Let {y i }, {x j } ⊂ A 1 \{0} be finite subsets such that i y i * y i = I and j x j x j * = I, and let ℓ ′ {yi} and ℓ {xj} be the corresponding one-sided subshifts, defined as above. Then Proof By Prop. 2.3 and the triangle inequality β min ≥ −h(T yi ) and β max ≤ h(S {xj } ). The rest follows from i1,...,in x i1 . . . x in 2 ≤ θ n (ℓ {xj } ), and its analogue for {y i }. We shall see that in the general situation if it is possible to choose the multiplets {y i } and {x j } carefully, then the topological entropies of the corresponding subshifts lead to the extremal inverse temperatures of KMS states. We first present an intermediate result, which gives a sufficient condition for h (T ) and h(S) to coincide with the topological entropy of the associated subshifts. Proof We shall prove only the first statement. Let θ n denote the number of words of length n occurring in ℓ ′ {yi} , i.e. the number on n-tuples (i 1 , . . . , i n ) ∈ Σ ′ n such that y i1 . . . y in = 0. Then given ε > 0, for infinitely many indices n, (1 − ε)θ n 1/n ≤ θ n 1/n l n 1/n ≤ ( y i1 . . . y in 2 ) 1/n ≤ θ n 1/n , hence taking the logarithm of the limit over n, The previous result applies whenever one is working with a multiplet consisting of partial isometries with mutually orthogonal ranges. We next show that, strengthening the hypotheses of the previous result, all the inequalities of Proposition 6.2 become equalities. More precisely, if the positive evaluations of a tracial state τ on the iterated basic monomials, e.g. (x i1 . . . x in ) * x i1 . . . x in , do not get too small when n increases, then the maximal inverse temperature β max can be approximated iterating the operator s ′ on τ . The proof is inspired by an analogous result in [MWY] for the Matsumoto algebras associated to subshifts. Theorem Let {y If lim sup n µ n 1/n (τ ) = 1 In particular, if τ is the restriction of a KMS state ω, then ω has minimal inverse temperature. If instead lim sup n ν n 1/n (τ ) = 1 If τ is the restriction of a KMS state ω, then ω has maximal inverse temperature. Proof We shall prove only the first statement. For infinitely many indices n, hence taking the n-th root and then the logarithm of the limit over a subsequence, by the arbitrariness of ε, we get lim sup The last equality follows from Prop. 6.3. Now δ n (τ ) 1/n = t ′ n (τ ) 1/n ≤ t ′ n 1/n → r(t ′ ), so h top (σ ↾ ℓ ′ {y i } ) ≤ −β min , which, together with Proposition 6.2, proves the first statement. If in particular τ arises from a KMS state ω at inverse temperature β, then δ n (τ ) = e −nβ , so β = β min . The previous result can be regarded as an analogue of the well known fact from Perron-Frobenius theory that for an irreducible nonnegative matrix A the maximal eigenvalue can be approximated by r(A) = lim sup n A n (τ ) 1/n , where τ is any vector with positive entries [G]. Corollary If there is a finite subset {y (2) (3) The algebra C generated by all finite products of the form is finite-dimensional, then the conclusions of the previous theorem hold for any faithful tracial state on A 0 . Proof Just note that under our assumptions any of the nonzero basic monomials is a projection, and therefore it majorizes a minimal projection in C. It follows that lim n µ n (τ ) 1/n = 1 (resp. lim n ν n (τ ) 1/n = 1) for any faithful trace τ on A 0 , so the previous theorem applies. In particular, this result applies to all the examples discussed at the beginning of this section. Topological entropy of canonical ucp maps Let {x j } be a finite set of a C * -algebra A of grade 1 such that j x j x j * = I. In the previous section we have associated to this set a one-sided subshift (ℓ {xj } , σ ↾ ℓ {x j } ) of the Bernoulli shift (Σ N , σ), where Σ is the state space of the first d positive integers, and d = Card{j : x j = 0}, in a way that, under suitable circumstances, its classical topological entropy equals an extremal inverse temperature of KMS states. One can also associate to the subset {x j } a unital completely positive map defined by In view of the results of the previous section, we ask whether there is a relationship between the the Voiculescu topological entropy of this map and the classical topological entropy of the subshift ℓ {xj } . If A = O n is the Cuntz algebra with generators S 1 , . . . , S n , Choda shows in [Ch] that the topological entropy of the canonical endomorphism σ {Si} is log(n), i.e. the topological entropy of the associated full shift. In the more general case where A = O A is a Cuntz-Krieger algebra, and {S j } is the canonical set of generating partial isometries, Boca and Goldstein [BG] have recently computed the Voiculescu entropy [V] of this map, and they have shown that it equals the logarithm of the spectral radius of A, or, in other words, the classical topological entropy of the underlying finite type subshift ℓ A [DGS]. However, special cases, although extreme from a certain point of view, of full periodic C * -dynamical systems are the crossed products by an automorphism α. Brown showed in [B] that ht A 0 ⋊Z (Ad(u)) = ht A 0 (α), where u is a unitary implementing α. It is obvious that the associated subshift is in this case trivial, so its entropy is zero. In Theorem 7.4 we give, for full periodic C * -dynamical systems, an upper bound for ht(σ {xj } ) which allows to recover the above discussed results as special cases. We will then apply this result to find new examples, among the Matsumoto algebras associated to non finite type subshifts, where In the beginning of this section the automorphic action of the circle plays no role, therefore we shall not assume that the x j 's are of grade 1. We define the associated one-sided subshift ℓ = ℓ {xj } as in the previous section. We now show that the ucp map σ {xj } can be understood as a noncommutative subshift. Let σ ↾ ℓ be the restriction of σ to ℓ and T ℓ the * -monomorphism of C(ℓ) obtained by transposing σ ↾ ℓ , i.e. Also, we will consider a natural basis of neighborhoods for ℓ. For each (i 1 , . . . , i r ) ∈ ℓ r , consider the cylinder set For a fixed r ∈ N, these constitute an open and closed cover of ℓ with cardinality θ r = Card ℓ r . 7.1. Proposition Let {x j } be a finite subset of A such that j x j x j * = I, and let (ℓ, σ ↾ ℓ ) be the associated one-sided subshift. Then there is a unique unital completely positive map Φ : C(ℓ) → A taking the characteristic function Moreover, if the sequence (m n ) n defined in Proposition 6.3 does not converge to 0, then Φ is faithful. Proof We first notice that, for each r ∈ N, the map Φ r : Σ N → Σ r projecting onto the first r coordinates takes ℓ onto the subset ℓ r of ⊂ Σ r consisting of θ r elements. Therefore there is a natural * -monomorphism φ r : C θr → C(ℓ) taking a θ r -tuple assuming value 1 on (i 1 , . . . , i r ) and zero elsewhere to the characteristic function of [i 1 . . . i r ]. Similarly, there are, for r ≤ s, natural monomorphisms φ r,s : C θr → C θs such that φ s φ r,s = φ r . Since the cylinder sets {[i i . . . i r ], r ∈ N} form a basis of closed and open sets for ℓ, we see that the image of all the φ r is dense. It follows that C(ℓ) is the inductive limit of the C θr 's under the maps φ r,s . We define the ucp map Φ r : C θr → A which takes the characteristic function of [i i , . . . , i r ] to the element x i1 . . . x ir (x i1 . . . x ir ) . Since Φ s φ r,s = Φ r s ≥ r, thanks to j x j x j * = I, we get a ucp map 0 Φ : ∪ r C θr → A. Since Φ r (f ) ≤ f , 0 Φ extends to a ucp map on C(ℓ), which is the desired Φ. The relation Φ • T ℓ = σ {xj } • Φ can be easily checked on the total set of characteristic functions of cylinder sets. We now construct a conditional expactation E r : C(ℓ) → C θr . Choose a faithful normalized Borel measure µ on ℓ, and associate to a function f ∈ C(ℓ) the θ r -tuple with coordinates for each (i 1 , . . . , i r ) ∈ ℓ r . One can easily check that E r (I) = I and that E r (f a) = E r (f )a, a ∈ C θr . Clearly (E r ) r converges pointwise in norm to the identity. Assume now that f ∈ C(ℓ) is a positive element such that Φ(f ) = 0. Then ΦE r (f ) converges to 0. On the other hand therefore if m r does not converge to 0, a subsequence of E r (f ) converges to 0, so f = 0. Note that if (m n ) n does not converge to 0 then lim sup n m 1/n n = 1 thus we are in the position of applying Proposition 6.3. A particularly important case is when the x j ' have pairwise orthogonal ranges. Corollary then the ucp map Φ : C(ℓ) → A constructed in the previous proposition is in fact a * -monomorphism. The restriction of σ {xj } to C(ℓ) corresponds to the one-sided subshift T ℓ . For any subshift Λ the Matsumoto C * -algebra O Λ satisfies the requirements of the previous result [M]. Let now (A, γ) be a full C * -dynamical system over T. Our next aim is to compare the topological entropy of the ucp map σ {xj } , when {x 1 , . . . , x d } is a finite subset of A 1 \{0} such that j x j x j * = I, with entropic properties of the canonical homogeneus C * -algebra. We refer the reader respectively to [V] for the notion of topological entropy for nuclear C * -algebras and to [B] for its generalization to exact C * -algebras, and to [BG] for the generalization of the topological entropy ht(P ) of a ucp map P on a unital exact C * -algebra. We will use the same notation as [B]. Thus in particular for a finite set Ω ⊂ A 0 , rcp(π, Ω, δ) is computed with respect to factorizations of completely positive contractions, not necessarily unital, from A 0 to B(H) via finite dimensional C algebras. Brown proves in [B] that rcp(π, Ω, δ) is independent of the choice of π. We will regard A faithfully represented on a Hilbert space H, and we take π to be the inclusion ι A 0 of A 0 in B (H). Moreover we will avoid indicating π in the above definitions. We anticipate, for later use, the following immediate consequence of the definition. (2) If ω is such that for some finite dimensional C -subalgebra D ⊂ A 0 which is the range of a conditional expectation, ω (n) ⊂ D except for finitely many n, then ht a ({φ xi,xj }, ω) = 0. Our aim is to show the following result. 7.4. Theorem Let (A, γ, T) be a full C * -dynamical system, with A 0 exact. Let σ {xi} and ℓ {xi} be the ucp map and the one-sided subshift associated to a set {x i } ⊂ A 1 satisfying i x i x i * = I. Then for any finite subset ω ⊂ A 0 and n 0 ∈ N 0 , We will prove this theorem combining appropriate analogues of arguments of Brown [B] for crossed product C * -algebras and Boca and Goldstein [BG] for Cuntz-Krieger algebras. Motivated by [B], we define certain cp maps. Let F ⊂ N 0 be a finite subset. Set Here I F := ∪ r∈F ℓ r , θ F = r∈F θ r , and, for k ∈ Z, m k : A → A k is the natural projection. Note that S F is contractive and cp. For a contractive cp map φ : which is contractive and cp. Let f ∈ ℓ 2 (N 0 ) have support in F , and define the cp map Note thatS F,f (I) = f 2 2 I. Again, for a contractive cp map ψ : So ψ F,f is cp contractive if f 2 ≤ 1. Finally, for a contractive cp map Λ : (H). Note that in particular, if φ and ψ are as above, which factors through the algebra M θF (B). One can easily show that for an element of fixed degree X ∈ A k , The following lemma is our analogue of Lemma 3.4 in [B]. Lemma Let ω be a finite subset of the unit ball of A 0 , n 0 ∈ N and δ > 0. Consider the set F ′ (ω, n 0 ) := {T x γ , T ∈ ω, \|γ\| ≤ n 0 }. Then there is a finite set F ⊂ N 0 which depends only on n 0 and δ and not on ω such that Proof We proceed as in the proof of Lemma 3.4 in [B]. Let f ∈ ℓ 2 (Z) be a function with finite support E, f 2 ≤ 1 such that \|f * f (n) − 1\| < δ/2, n = 0, 1, . . . , n 0 . Heref (p) = f (−p). Replacing f with a suitable translate if necessary, we may assume , which is a subset of N 0 . Note that F depends on δ and n 0 but not on ω. Consider contractive cp maps φ : A 0 → B and ψ : Let us choose B with minimal rank. Then the cp contractive map Φ ψ•φ,F,f factors through the finite dimensional algebra M θF (B), which has rank θ F rank (B). We are thus left to show that for a ∈ F ′ (ω, n 0 ), We write a = T x γ with T ∈ ω, \|γ\| ≤ n 0 . Then the l.h.s. is bounded by Now the computation of Φ ι A 0 ,F,f given before on elements with fixed degree shows that the second summand is bounded by We now evaluate the first summand. The last inequality follows from our choice of ψ and φ and from the fact that if a matrix A ∈ M h,k (A 0 ) has entries of norm bounded by c then A ≤ (hk) 1/2 c. Proof of Theorem 7.4. We apply the previous Lemma to the sets F ′ (ω (n+n0) , n 0 ) for fixed n 0 and ω and all n ∈ N. Note that the corresponding set F can be chosen independent of n. We can thus find for each n ∈ N a contractive cp map Λ n : A → B (H) factoring through a finite dimensional algebra B of rank Consider the ucp map Ψ m : M θm (B(H)) → B (H) taking the matrix (t µ,ν ) to the operator \|µ\|=\|ν\|=m π(x µ )t µ,ν π(x ν ) * . Corollary Consider the same situation as in Theorem 7.4. Let (ω α ) α∈A be a net of finite subsets of A 0 with total union. Then Proof This is a straightforward consequence of the fact that ∪ α,n0 F (ω α , n 0 ) ∪ F (ω α , n 0 ) * is total in A and of the Kolmogorov-Sinai property of the entropy of a ucp map, [V], [B], [BG]. Remark If in particular A = A 0 ⋊ α Z and u ∈ A 1 is a unitary implementing α on A 0 then ℓ u is a single point space, so its entropy is zero. By Lemma 7.3 and the previous Corollary, we recover Brown's result that ht A (Ad(u)) ≤ ht A 0 (α) (and therefore one deduces an equality by monotonicity of topological entropy [B].) The case where we can choose the x j 's with pairwise orthogonal ranges is of course of special interest, the Cuntz algebras, Cuntz-Krieger algebras and Matsumoto algebras belonging to this class. The next result shows that the estimate of the entropy can be made more precise in this case. 7.7. Theorem Let (A, γ, T) be a full C * -dynamical system, with A 0 exact. Let {x j } ⊂ A 1 be a finite subset such that j x j x j * = I, where (ω α ) α∈A is any net of finite subsets of A 0 with total union in A 0 . If in particular for some net (ω α ) α∈A ht 2 ({φ xi,xj }, ω α ) = 0, α ∈ A, then Proof The proof of the second inequality ≤ goes exactly as that of Theorem 7.4 with the only exception that entries of ρ m (σ l (x γ T )) are now already elements of F ′ (ω (n+n0) , n 0 ). We show that By monotonicity of topological entropy [B], [V] and Corollary 7.2, where, as before, T ℓ denotes the * -monomorphism of C(ℓ) implemented by the one-sided shift. We are thus left to show that ht (T ℓ The proof is similar to that of [BG], which in turn goes back to [V,Proposition 4.6]. Let µ be a σ-invariant probability Borel measure on the two-sided subshift Λ = Λ {xj} defined before Prop. 6.1, and let us restrict it to a σ-invariant probability measure on ℓ = Λ + . For any ucp map γ : M → C(ℓ), with M finite dimensional, let h µ,T ℓ (γ) be defined as in [CNT], by means of the function H µ (γ, T ℓ γ, . . . , T ℓ n−1 γ). Reasoning as in [V,Prop. 4,6] we see that h µ,T ℓ (γ) ≤ ht(T ℓ ). Choosing M = C θn and γ : C θn → C(ℓ) the natural inclusion, then one finds, thanks to [CNT,Remark III.5.2], that the classical measurable entropy H µ (σ ↾ Λ ) is ≤ ht (T ℓ ). Taking the supremum over all invariant measures we obtain the claim, by the classical variational principle for topological entropy [DGS,Theorem 18.8]. Remark It is natural to ask whether the upper bound for ht(σ {xj } ) described in the previous result can be further improved to h top (σ ↾ ℓ {x j } ) + ht 0 ({φ xi,xj }). We next show that the estimates above obtained are good enough to compute ht(σ {xj } ) in the case of Cuntz-Krieger algebras. This was first done by Boca and Goldstein [BG]. Proof The second equality is the well known computation of entropy of finite type subshifts of order two. See, for example, Proposition 17.12 in [DGS]. The first equality will follow from the previous Theorem provided we show that there is an increasing sequence ω p , p ∈ N of finite subsets of O A 0 with total union such that ht 2 ({φ Si,Sj }, ω p ) = 0, p ∈ N. It is clear that ∪ p ω p is total in the homogeneous subalgebra. One easily checks that, for fixed p and all n, ω p (n) is contained in the linear span of ω p , which is finite dimensional C * -algebra, so by Lemma 7.3 ht 2 (φ {Si,Sj } , ω p ) = 0. We next look at the class of Matsumoto algebras O Λ associated to a general subshift Λ ⊂ {1, . . . , d} Z [M]. See also subsection 4.1. Let {S i } d 1 be the canonical set of generating partial isometries. We recall from [M] a few properties of O Λ . First, the relations which easily imply that for any pair of words with the same length, q α,β := S α * S β =0 unless α = β. We will write q α for q α,α , α ∈ ∪ r Λ r . Note that these are projections. Furthermore the following commutation relations hold in O Λ : for µ, ν ∈ ∪ r Λ r : q µ S ν = S ν q µν , (7.3) q µ q ν = q ν q µ . (7.4) By (7.4) the algebra Q l generated by the projections {q α , α ∈ ∪ l k=0 Λ k } is commutative and therefore finite dimensional. These properties imply that the finite sets ω k,l := {S α ES β * , \|α\| = \|β\| ≤ k, E minimal projection in Q l } has total union in O Λ 0 . It follows that O Λ 0 is AF [M], so O Λ is a nuclear C * -algebra. Using properties (7.1)-(7.4) one can show with tedious computations that for all n ∈ N, This computation is aimed to give an estimate for ht a (φ Si,Sj , ω k,l ). The last inequality follows from the existence of a conditional expectation onto Q 2n+2 max(k,l) . The rest follows choosing δ of the form δ θn−1 a , taking the logarthm, dividing by n and passing to the lim sup. We combine the previous result with Theorem 7.6. Theorem If σ {Si} is the ucp map associated to the canonical set of generators of a Matsumoto C * We conclude this section discussing two examples of subshifts for which ht(σ {Si} ) = h top (Λ). The first example beyond Markov shifts is that of sofic subshifts, see [DGS] and therein quoted references. By [M] a subshift is sofic if and only if ∪ n Q n is finite dimensional. Then Theorem 7.10 yields the desired equality. Another example is that of β-shifts associated to β-expansion of real numbers [Re], [Par], [Bl]. In this case it is proved in [KMW] that if the β-shift is not sofic, dim(Q n ) = n + 1, and this leads again to the same conclusion. CNT dynamical entropy and variational principle In fact, if A = O n Choda shows in [Ch] not only that h top (σ) = log(n) but also that, if φ is the unique KMS state of O n , then h φ (σ) = h top (σ) = log(n), where the l.h.s. denotes the Connes-Narnhofer-Thirring dynamical entropy of σ [CNT]. This result has its own importance, as it exhibits a fundamental example where a noncommutative variational principle for the entropy holds true. (We refer the reader to [DGS] for a formulation of the variational principle for the entropy in ergodic theory for compact spaces.) It is an open problem whether a noncommutative variational principle for the entropy of C * -algebras holds. In this section we give a class of examples for which this is true, thus generalizing Choda's result. Examples will be the canonical ucp map of the Cuntz-Krieger algebras, or certain Matsumoto algebras associated to non finite type subshifts. In this section we show that the CNT dynamical entropy of the ucp map σ {xj } , defined in the previous section is ≥ the m.t. entropy of the associated subshift Λ {xj} , as defined, e.g., in [DGS], see Theorems 8.5, 8.6. This inequality looks similar to that of Theorem 7.7 relative to the topological entropy, but it goes in the reverse order. We start establishing the setting of the CNT entropy. Let A be a unital C * -algebra, and let γ i : A i → A, i = 1, . . . , n be ucp maps from finite-dimensional C * -algebras, and let φ be a state on A. Let us recall from [CNT] that an Abelian model for (A, φ, γ 1 , . . . , γ n ) is given by an Abelian finite-dimensional C * -algebra B, a state µ on B and subalgebras B 1 , . . . , B n of B for which there is a ucp map E : A → B with φ = µ • E. Consider first the entropy of the Abelian model (B, µ, B 1 , . . . , B n ) as defined in [CNT,III.3] and then the quantity H φ (γ 1 , . . . , γ n ), defined as the supremum of the entropies of all the Abelian models (see [CNT,Definition III.4]). The following result is an obvious consequence of the definition. is finite-dimensional and commutative and if there exists a conditional expec- where the r.h.s. denotes the classical m.t. entropy of the restriction of φ to is an Abelian model for (A, φ, γ 1 , . . . , γ n ). Let E i : B → B i denote the canonical conditional expectation associated to µ. Then E i • E • γ i : A i → γ i (A i ) conincides with γ i , which is a * -isomorphism, thus its entropy defect is zero (see [CNT], section II). It follows from [CNT,Definition III.4] that Let now σ be a ucp map of our C * -algebra A such that φ • σ = φ, and let γ : M → A be a ucp map from a finite-dimensional C * -algebra M . Define the m.t. dynamical entropy of γ with respect to φ to be and, finally, define the m.t. dynamical entropy of σ as h φ (σ) = sup γ {h φ,σ (γ)}, where the supremum is taken over all possible γ : M → A. Corollary Let A be a unital C * -algebra, φ a state of A, and let σ be a ucp map of A such that φ•σ = φ. Let γ : M → A be a unital * -monomorphism from a commutative finite-dimensional C * -algebra M . Assume that the smallest σstable C * -subalgebra C of A containing γ(M ) is commutative and that σ ↾ C is a * -monomorphism. If, for n ∈ N, there exists a conditional expectation where the r.h.s. denotes the classical m.t. dynamical entropy of the partition of the spectrum of C defined by γ(M ) with respect to σ ↾ C (see, e.g., [DGS,Def. 10.8]. It follows that where the r.h.s. denotes the classical m.t. entropy of the epimorphism of the spectrum of C defined by the restriction of σ ( [DGS,Def. 10.10]. Proof Just apply the previous lemma to γ 1 = γ, . . . , γ n = σ n−1 γ and then pass to the limit. The last assertion is a consequence of the classical Kolmogorov-Sinai property of the entropy. In order to apply the above result, one needs to know under which conditions on the system (A, σ, γ) as in Cor. 8.2 every invariant measure µ on C extends to a σ-invariant state φ on A fulfilling all the requirements of the previous Corollary. We start giving a well known sufficient condition for the existence of invariant conditional expectations. Then there is a unique conditional expectation E : Proof Set E(a) = φ(e) −1 φ(ae)e, and check that E is the desired conditional expectation. Uniqueness follows easily from faithfulness of φ on M . We next give a condition on (A, σ, γ) so that every invariant measure on the spectrum of C extends to a σ-invariant state on A containing C in its centralizer. In view of the previous Lemma, this would imply the existence of conditional expectations E n as in Corollary 8.2, satisfying all the necessary requirements. In particular, if µ is faithful, for every finite-dimensional C * -subalgebra M ⊂ C there exists a unique conditional expectations E M : A → M such that Proof Let C 1 ⊂ C 2 ⊂ . . . be an increasing sequence of unital finite-dimensional C * -subalgebras of C with dense union, and, for n ∈ N, let F n be the set of minimal projections of C n . Set which is a nonempty convex and compact subset of the state space S(A) in the weak * -topology. The function f 0 taking any state φ on A to the state a → e∈F1 φ(eae) is weakly * -continuous and leaves K 0 invariant, thus, by the Schauder-Tychonov fixed point theorem, the fixed point set is nonempty. Note that K 1 is still compact and convex. Define now the weakly continuous function f 1 : S(A) → S(A) taking φ to a ∈ A → φ( e∈F2 eae) and check again that K 1 is invariant under f 1 , so that is nonempty. We thus find iteratively a decreasing sequence K 0 ⊃ K 1 ⊃ K 2 . . . of nonempty compact convex subsets of S(A). Consider the compact convex set K := ∩ n∈N K n . A state φ is in K if and only if φ ↾ C = µ, φ(ae) = φ(ea), e ∈ ∪ n F n , a ∈ A, and therefore, being ∪ n C n dense in C, φ(ca) = φ(ac), c ∈ C, a ∈ A. Thus, applying again the Schauder-Tychonov fixed point Theorem, we find a fixed point φ of f σ , which is the desired extension of µ. The last assertion now follows from the previous lemma. We now collect all the results we have obtained, in form of a Theorem. 8.5. Theorem Let A be a unital C -algebra, endowed with a faithful ucp map σ, and let γ : M → A be a unital * -monomorphism from a commutative finitedimensional C * -algebra M . Assume that the smallest σ-stable C * -subalgebra C of A containing γ(M ) is commutative and that σ(ca) = σ(c)σ(a), c ∈ C, a ∈ A. Let µ be a faithful σ-invariant state on C extended to a σ-invariant state φ on A centralized by C, this being possible by Prop. 8.4. Then where the r.h.s. denotes the classical m.t. entropy of the epimorphism of the spectrum of C defined by the restriction of σ to C. We now go back to the situation where σ = σ {xj } is the ucp map implemented by a finite subset {x j } of A such that j x j x j * = I. 8.6. Theorem Let A be a unital C * -algebra, and let {x j } by a finite subset constituted by d nonzero partial isometries satisfying j x j x j * = I, j x j * x j is invertible, Also, σ is faithful as j x j * x j is invertible. The commutation relations between the domain projections and the range projections of the iterated products of the x j 's easily show that σ {xj } (ca) = σ {xj } (c)σ {xj } (a), c ∈ Φ(C(Λ {xj } + )), a ∈ A. In particular, σ {xj } is a * -monomorphism on Φ(C(Λ {xj } + )). Thus the previous theorem applies. If for example A = O A or, more generally, A = O Λ , the assumptions of the previous theorem hold true. Note that, with the notation and assumptions of the previous result, we know that, using also Corollary 7.6, where ω α is any net of finite subsets of A 0 with total union. The middle inequality is due to Voiculescu [V]. In classical ergodic theory, a probability measure m on a dynamical system (X, T ) such that m • T * = m is called an equilibrium measure, or a measure with maximal entropy, if it maximizes the entropy, i.e. if h m (X) = h top (X). It is well known that dynamical systems arising from subshifts admit equilibrium measures, see [DGS]. Applyling this fact to our subshift Λ {xj } , we see that there exists a shift-invariant measure µ on Λ {xj} with Combining with the previous inequality, we obtain, under the simplifying assumption that the second summand in (8.1) vanishes, an existence theorem of equilibrium states in the noncommutative situation above considered. 8.7. Corollary Consider the same situation as in Theorem 8.6. Let µ be a shift-invariant measure on Λ {xj } with maximal entropy, and let us extend it to a σ {xj } -invariant state φ on A centralized by Φ(C(Λ {xj} + )). Assume furthermore that for a net ω α of finite subsets of A 0 with total union, ht 2 ({φ xj ,xj }, ω α ) = 0. By virtue of the remark following Theorem 7.10, we obtain the following result. 8.8. Corollary Let Λ be a subshift of one of the following types: (1) Markov shift, (2) sofic subshift (3) β-shift. Let us extend an invariant measure µ on Λ with maximal entropy to a state φ on O Λ centralized by the canonical commutative subalgebra where {S i } is the canonical set of generating partial isometries of the Matsumoto algebra O Λ . From KMS states to equilibrium states In this section we make an attempt to show a closer connection between KMS states on full periodic C * -dynamical systems studied in sections 1-6 and equilibrium states considered in sections 7-8. To motivate the result of this section, we consider the classical situation of a topological dynamical system (X, T ) over a compact space X. A Borel probability measure m on X is called conformal if m • T * is equivalent to m. The study of conformal measures is of particular importance as it leads to equilibrium states of the system [DU]. Now in our noncommutative setting, where we replace X by a unital C * -algebra A endowed with a full action of the circle, and T by the ucp map σ {xj } , KMS states provide a natural class of states on A which play the role of conformal measures. Indeed we have the following immediate result. 9.1. Proposition Let (A, γ) be a full periodic C * -dynamical system over a unital C * -algebra A, and let {x j } be a finite subset of A 1 such that j x j x j * = I and j x j * x j is invertible. If ω is a KMS state at inverse temperature β then where a = e −β j x j * x j is obviously a positive and invertible element of A 0 . We show how to produce σ-invariant states on A from KMS states of the system (A, γ). Consider the completely positive map S {xj} : a ∈ A → j x j * ax j already considered in section 2. Let ω be a faithful KMS state for (A, γ) at maximal inverse temperature β max = log(λ max ). Then, for t > λ max , consider the series which we claim to be Cauchy for every seminorm p T , T ∈ A, where p T (a) = \|ω(aT )\|, a ∈ A. We show the claim. so ω(a t ) = 1, and define ω t := ω(a t · ). Then for any T ∈ A, for t → λ max . So any weak * -limit point φ of ω t for t → λ max is a σ {xj }invariant state on A. Note that if every x µ * x µ commutes with any x ν x ν * , for all multiindices µ and ν, then the Banach space generated by the x ν x ν * is in the centralizer of any such φ. The next result can be regarded as an example where the construction of equilibrium states out of KMS states is explicit. This is the noncommutative analogue of the well known relationship between the Perron-Frobenius Theorem and equilibrium states for Markov subshifts, see Proposition 17.14 in [DGS]. 9.2. Theorem Let (A, γ, T) be a unital, full, periodic C * -dynamical system, and let O A ⊂ A be a unital Z-graded inclusion of the Cuntz-Krieger algebra associated to an irreducible matrix A, in A. If ω is a faithful KMS state of A at maximal inverse temperature log(λ max ), then (1) λ max = r(A), (2) ω t is norm convergent, for t → λ max + , to a σ {Sj } -invariant state φ centralized by C(Λ A+ ), where {S j } is the canonical set of generators of O A . (3) φ restricts on C(Λ A+ ) to the unique probability measure µ for which In particular, if for a net ω α of finite subsets of A 0 with total union ht 2 ({φ xi,xj }, ω α ) = 0 Proof It is known that Markov subshifts defined by irreducible matrices have a unique maximal measure, see Theorem 19.14 in [DGS]. The elements a t , t ≥ λ max , defined as in (9.1) belong to the finite-dimensional C * -subalgebra of C(Λ A+ ) generated by S i S i * , the characteristic functions of the cylinders [i], i = 1, . . . d. Since ω(a t ) = 1 and ω is faithful, there exists a norm-limit point a of a t , for t → λ max + . Inspection shows that a is an eigenvector of S {Si} with eigenvalue λ max , and therefore it corresponds to a left eigenvalue (v i ) of A, normalized so that ω(a) = 1. In particular, a t is convergent. Since ω is a KMS state of A, and hence of O Λ w.r.t. the gauge action, evaluating ω on S i S i * gives the unique, up to a scalar, positive right eigenvector (u j ) of A. The normalization ω(a) = 1 yields i u i v i = 1. Evaluating φ on S i1 . . . S ir (S i1 . . . S ir ) * gives φ(S i1 . . . S ir (S i1 . . . S ir ) * ) = ω(aS i1 . . . S ir (S i1 . . . S ir ) * ) = v i1 ω(S i1 . . . S ir (S i1 . . . S ir ) * ) = v i1 λ r ω((S i1 . . . S ir ) * S i1 . . . S ir ) = v i1 λ r a i1,i2 . . . a ir−1,ir i j ω(a ir ,j S j S j * ) = v i1 u ir λ r−1 a i1,i2 . . . a ir−1,ir . If we now compare with the formula given in Prop. 17.14 in [DGS], we see that µ = φ ↾ C(ΛA + ) restricts precisely to the unique measure on Λ A+ with maximal entropy.	2014-10-01T00:00:00.000Z	1999-12-18T00:00:00.000	{ "year": 1999, "sha1": "659f355efc2464272d4ba403f2eed17170fca902", "oa_license": null, "oa_url": "http://arxiv.org/pdf/math/9912151", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "659f355efc2464272d4ba403f2eed17170fca902", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [ "Mathematics" ] }
259321659	pes2o/s2orc	v3-fos-license	Germline variant of CTC1 gene in a patient with pulmonary fibrosis and myelodysplastic syndrome Introduction Telomeropathies are associated with a wide range of diseases and less common combinations of various pulmonary and extrapulmonary disorders. Case presentation In proband with high-risk myelodysplastic syndrome and interstitial pulmonary fibrosis, whole exome sequencing revealed a germline heterozygous variant of CTC1 gene (c.1360delG). This "frameshift" variant results in a premature stop codon and is classified as likely pathogenic/pathogenic. So far, this gene variant has been described in a heterozygous state in adult patients with hematological diseases such as idiopathic aplastic anemia or paroxysmal nocturnal hemoglobinuria, but also in interstitial pulmonary fibrosis. Described CTC1 gene variant affects telomere length and leads to telomeropathies. Conclusions In our case report, we describe a rare case of coincidence of pulmonary fibrosis and hematological malignancy caused by a germline gene mutation in CTC1. Lung diseases and hematologic malignancies associated with short telomeres do not respond well to standard treatment. Introduction Human telomeres consist of long TTAGGG sequence repeats that are bound and stabilized by two nucleoprotein complexes required for the protection and replication of chromosome ends [1]. A protein complex "shelterin" is comprised of six telomerespecific proteins (TRF1, TRF2, POT1, Rap1, TIN2 and TPP1), and a heterotrimeric "CST" complex is composed of the proteins Ctc1, Ten1 and Stn1 involved in telomere maintenance [2,3]. Genomic alterations in these proteins are causative of a number of disorders known as telomeropathies. Telomeropathies are closely associated with premature aging and a reduction of cell ability to cope with recurrent damage. Several pulmonary, hematological, or liver diseases are associated with telomeropathies. These diseases include pulmonary fibrosis as idiopathic pulmonary fibrosis (IPF) and myelodysplastic syndrome (MDS). IPF manifests as bibasilar reticular abnormalities, bronchiectasis, honeycombing on high-resolution computed tomography, restrictive pulmonary function impairment, and decreased lung diffusion capacity for carbon monoxide. IPF is idiopathic interstitial pneumonia characterized by progressive fibrotic damage of lung parenchyma. MDS is a clonal hematopoietic tissue disorder manifesting as morphologic dysplasia in myeloid lineage and peripheral cytopenia [4]. Case report A 69-year-old Caucasian male, a non-smoker, presented with a mild dry cough and mild shortness of breath after exercises. He was sent for suspected interstitial pulmonary fibrosis. The patient's family history was negative regarding pulmonary and malignant diseases. Until then he was only treated for hypertension. The patient was an old-age pensioner. He worked as an engineer (dustfree office work). Clinical assessment (including screening questionnaire, antibodies against specific antigens) was performed in order to exclude hypersensitivity pneumonitis or autoimmune dis-ease. No exposure was found to cause exogenous allergic alveolitis. No causative antigens were found. Physical examination revealed clubbing fingers and bilateral end-inspiratory crackles in the lower and middle lung areas. The posteroanterior chest-X ray showed bilateral reticular pulmonary infiltrates (Figure 1). High-resolution computed tomography (HRCT) of lungs identified reticular opacities, bronchiectasis, and honeycombing changes (Figures 2 and 3). Pulmonary function testing revealed no ventilation defect [forced vital capacity (FVC) = 3.54 L, 100% of the predicted value (p.v.); total lung capacity (TLC) = 5.13 L; 82%] and moderate decrease of diffuse lung capacity for carbon monoxide (DLco; 4.23 L; 53%p.v.). Arterial blood gas analysis was also normal. Bronchoscopy with bronchoalveolar lavage showed an increased number of neutrophilic granulocytes and lymphocytes (13%, 74%). Lymphocyte subtypes were investigated: CD3+CD4+ = 69.20%, CD3+CD8+ = 27.00%. The ratio CD4+/CD8+ was 2.5. The neutrophilic and lymphocytic alveolitis was caused by a previous lung infection in a patient immunocompromised by hematological disease. There was no yield of transbronchial lung biopsy. Due to the patient's condition and hematological disease, histological verification by transbronchial cryobiopsy was not indicated. We considered other interstitial lung diseases (ILDs) including hypersensitivity pneumonitis (HP), but we did not find any exposure in either work or home. The HRCT finding was discussed with the Multidisciplinary Team (MDT). Other ILDs were excluded, autoantibodies were negative, and no exposure leading to hypersensitivity pneumonitis (HP) was found (negative questionnaires, specific IgG were negative). The diagnosis was concluded as usual interstitial pneumonia on HRCT of the thorax with honeycombing at dorsal basal regions. The patient had been treated with corticoids before he was referred to us. Radiological progression of lung fibrosis occurred during treatment. We gradually discontinued corticoids. The duration of this treatment was 6 months. We considered antifibrotic treatment, but we did not administer it, as the patient did not meet the indication criteria for antifibrotic therapy. At the same time, the patient developed thrombocytopenia (77 x 10 9 /L) and macrocytic anemia (hemoglobin 108 g/L; mean red cell volume 111 fL). Furthermore, high-risk MDS with blast excess (EB-2) was diagnosed (11.2% of myeloblasts in the bone marrow; deletion 5q in cytogenetics). Therapy with 5-azacytidine was initiated and the patient received 20 cycles of this treatment. Unfortunately, the patient was only temporarily stabilized by treatment and subsequently progressed into acute myeloid leukemia and died of severe COVID-19 pneumonia with respiratory failure. Because of suspected inherited predisposition leading to IPF and MDS, genetic testing focused on whole exome sequencing (WES) was performed. The WES variants evaluation process was aimed at the analysis of single nucleotide variants (SNV) and short indels (indel) within virtual genes panel associated with myeloid malignancies and predispositions to both myeloid and pulmonary disorders (Supplementary Table 1 Discussion Telomeropathies are very heterogeneous diseases. Therefore, patients with the same mutation can present different manifestations. This report describes a rare case of coincidence of pulmonary fibrosis and hematological malignancy caused by a germline gene variant in CTC1 (CST telomere maintenance complex component [1]. There is no similar case of the coincidence of these two diseases described in the medical literature. We did not find any articles describing the coincidence of the two diseases (aplastic anemia and pulmonary fibrosis). The CTC1 gene (OMIM #613129), located on chromosome 17p13.1, encodes a 1,217 amino acids nuclear protein Ctc1 which is a component of the conserved telomere maintenance CST complex along with the Stn1 and Ten1 proteins [3,6,7]. CTC1 variants alter the function of the CST complex, which may result in the shortening of telomeres and DNA damage responses [8,9]. In that case, telomeres are recognized as damaged DNA that can result in cell-cycle arrest, cell apoptosis, or senescence. The CST complex was initially proposed to play a role in telomere length homeostasis by reducing access of telomerase to telomeres in order to prevent excessive telomere lengthening [10]. Subsequently, the CST complex was shown to promote telomere replication [6]. Therefore, the Ctc1 protein is involved in the maintenance of telomeres. Telomeropathies are very heterogeneous diseases depending on the gene mutated and the specific variants, their penetrance, and the existence of anticipation effects. Therefore, patients with the same variant can present different manifestations. Some patients present severe symptoms early, such as those of dyskeratosis congenita (DC) or the related Hoyeraal-Hreidarsson syndrome, Resvesz syndrome, and Coats plus syndrome. Moreover, diseases associated with telomeropathies may occur at a younger age than usual in sporadic forms: AA median age 20-30 years at diagnosis [11], idiopathic pulmonary fibrosis (IPF) median age 40-60 years at diagnosis [12,13]. A study by Arias-Salgado et al. describes genetic analyses of aplastic anemia and idiopathic pulmonary fibrosis patients with short telomeres caused by CTC1 gene mutation but not as a coin- cidence of both diseases, but each disease separately. However, we did not find any articles describing the coincidence of the two diseases (aplastic anemia and pulmonary fibrosis) [13]. There is an increasing number of inherited disorders in which excessive telomere shortening underlies the molecular defect, with dyskeratosis congenita (DC) being the archetypal short telomere syndrome. Excessive telomere shortening can affect almost any organ system, so the clinical manifestations are protean, including developmental delay, cerebellar hypoplasia, exudative retinopathy, AA, acute myeloid leukemia, IPF, idiopathic hepatic cirrhosis, head and neck cancer and dental abnormalities, and may be multisystemic [14]. Unfortunately, there are no "age-specific" telomere length standards, because their length is reflected in an incredible number of individual factors (age is only one of them) [15]. For this reason, it is still difficult to use telomere length for routine diagnosis. Diseases and conditions associated with pathogenic variants of CTC1 include cerebroretinal microangiopathy with calcifications and cysts (CRMCC) and DC. The majority of CTC1 pathogenic variants found in either CRMCC or DC are compound heterozygotes of a missense variant and a truncation variant, with a few exceptions of compound heterozygotes for two missense variants [16,17]. Blood cells from CTC1 mutated patients were shown to exhibit shortened telomere lengths or telomere lengths at the lower range of normal [8,18]. However, one study reported no significant differences between the leukocyte telomere lengths of CTC1 mutant patients and those of controls, raising the possibility that the disease mechanism of CTC1 mutations may involve nontelomeric functions [19]. Short telomere syndromes associated with pulmonary diseases, particularly fibrosis, respond poorly to standard treatments, such as corticosteroids and bronchodilators. Androgen derivatives could be a potential therapeutic option able to re-elongate previously shortened telomeres. However clinical trials are needed to develop pharmacological agents aimed at correcting disease-causing genetic defects and determine if androgen therapy is effective for telomere-related interstitial lung diseases [20,21]. Lung transplantation may be required in end-stage disease. Higher rates of hematological, renal and infectious complications were seen, requiring reduced immunosuppressive regimens in all cases [14]. The prognosis of high-risk MDS is also poor. It is even worse in cases with a congenital predisposition. In our patient, there was only a temporary stabilization of the disease during 5-azacytidine therapy. Conclusions Telomeropathies are associated with a wide range of diseases and combinations of various pulmonary and extrapulmonary disorders. Lung diseases associated with short telomeres, such as pulmonary fibrosis, and also hematologic malignancies do not respond well to standard treatment. With our case, we want to draw attention to the fact that even apparently different diseases can have the same genetic basis in one patient.	2023-07-05T05:07:45.926Z	2023-01-17T00:00:00.000	{ "year": 2023, "sha1": "7d0eaf3494623ce38da5ae5fa45c979988911942", "oa_license": "CCBYNC", "oa_url": "https://mrmjournal.org/mrm/article/download/909/1405", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "7d0eaf3494623ce38da5ae5fa45c979988911942", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
139161127	pes2o/s2orc	v3-fos-license	Continuous prefabricated cross-beams manufactured by using off-shuttering molding technologies The paper describes a new constructive solution of prefabricated cross-beams manufactured be using off-shuttering molding technologies. Prefabricated elements of such cross-beams are pre-stressed trough-shaped parts. Together with hollow slab panels, stacked on the upper surfaces of their vertical walls, they serve as permanent shuttering for installing required crosswise and longitudinal (for support moments) reinforcement and for further pouring of cast concrete. The study provides a methodology and gives tests results of three series of prototype samples and of one full-sized prefabricated cross-beam. The purpose of the tests is to study the actual work of prefabricated cross-beams of the proposed construction. To analyze the test results, the authors performed check calculations of prototype samples according to the existing Russian design regulations of reinforced concrete structures. The obtained data analysis leads to a conclusion that the proposed construction of prefabricated crossbeams is quite reliable and makes it possible to give recommendations on their calculation and design. Introduction Now in Russia there are large-scale processes of the technical re-equipment of enterprises producing reinforced concrete structures. Among the most dynamically introduced technologies, it is possible to distinguish the off shuttering molding technology [1], which makes it possible to produce various designs of constant cross-section length, reinforced with pre-stressed high-strength reinforcement. In this case, the cross-section can be of almost any shape (rectangular, T-shaped, I-shaped, U-shaped, with and without pinholes, etc.), and the product length is limited only by the requirements for the load bearing capacity of the manufactured structure. The authors proposed the construction of precast reinforced concrete frame, where in addition to hollow-core slabs manufactured by the off shuttering molding technology with continuous precast crossbeams [2], precast parts of which are manufactured as hollow-core slabs according to off shuttering molding technology, unlike cross beams for a similar frame, considered in the paper [3] and manufactured by conveyor technology. An example of the use of beams of T-section, MATEC Web of Conferences 196, 02021 (2018) https://doi.org/10.1051/matecconf/201819602021 XXVII R-S-P Seminar 2018, Theoretical Foundation of Civil Engineering manufactured by the off shuttering molding technology, can be the construction of a precast cross beam, mentioned above [4]. The proposed crossbeam ( Figure 1) consists of pre-stressed concrete parts with complex geometric shape, installed between the columns, and monolithic concrete which combines prefabricated elements (columns and beam parts) into a continuous single precast monolithic structure. Thus, the part and the slabs stacked on it serve as permanent shuttering for concrete, the part being a kind of pre stressed reinforcing element. To prevent the concrete from flowing into the slab pinholes plugs are inserted. Span parts in the proposed precast cross beam are taken up by the longitudinal pre stressed reinforcement part. For the take up of support elements in the nearby zone, the upper non-tensioning longitudinal reinforcement passing through the column ( Figure 5) is installed. The shearing strains are taken by the vertical reinforcing bars of frames. When the span poi4-4nts obtained from the results of static analysis framework, exceed the maximum cross beam strength, and the cross beam is reinforced only by pre stressed reinforcement parts on the bottom part (before laying monolithic concrete) additional untensioned longitudinal reinforcement can be set. 1prefabricated beam part; 2 -monolithic concrete; 3hollow-core plank; 4plastic plug; 5, 6lower and upper pre-reinforced wire reinforcement with 5 mm diameter, сarbon steel wire for reinforcement (Russian standards -Вр1400: 7)additional (if needed) lower untensioned reinforcement А400: 8reinforcement frames from А400 cores. Materials and methods In order to study the function and destruction nature of the proposed precast cross beams and prefabricated pretension part acting as an independent structure, as well as to study their strength in normal and inclined cross sections, special actual size prototype samples were manufactured and tested. The three series of samples were made and tested (two samples of the first series and three samples of the second and third series) and a cross beam part. Before the tests, the actual strength of concrete of each sample was determined by the methods of destructive and non-destructive testing (model N Schmidt hammer) (in cubes 100x100x100 mm). The main geometric characteristics of the samples are shown in Figures 2 and 3, and the reinforcement parameters and strength characteristics of the concrete in the samples in Table 1. 1 -the prefabricated cross beam part; 2 -monolithic concrete; 3 -hollow core slab; 4 -plastic plug; 5, 6lower and upper pre-reinforced wire reinforcement with 5 mm diameter, сarbon steel wire for reinforcement (Russian standards -Вр1400: 7)additional (if needed) lower А400: 8reinforcement frames from А400 cores. The test load for all samples was applied in steps up to destruction with an exposure time of 10 minutes at each stage. In the process of testing, the observation of the appearance and development of normal, inclined and contact cracks was made (between the beam part and monolithic concrete). The deflection in the middle of the span and the drafts of the supports were measured by the PAO-6 flexometers, the deformation of the concrete by resistance strain gages with a base of 50 mm and the Hugenberger strainmeters with a base of 100 mm, deformation of the reinforcemen by strain gages with a base of 10 mm. Notations adopted in Table 1: A sp -the cross-sectional area of the lower pre stressed reinforcement of the cross beam part; a sp -the distance from the center of gravity of the A sp reinforcement to the bottom surface of the cross beam part; A s -the cross-sectional area of an unstressed stretched reinforcement (the lower in the span, the upper on the support); a s -the distance from the center of gravity of the reinforcement A s of the cross beam to its nearest horizontal side; q sw -the force in the clamps per unit length, determined by formula (8.59) [3], replacing R sw by 0.8 * R sn , where R sn = 400 MPa -the standard value of the resistance of reinforcement class A400; R bm , R btm -the average strength of concrete to axial compression and axial tension, respectively, taken as a result of determining the actual strength of concrete samples. crossbeams and when testing the crossbeam part at the top of its edges. The distance between the supports of the upper distributor beam is 1600 mm, the two lower ones -600 mm each, the calculated span of the samples and the beam part were taken equal to 6850 mm. Samples of the second and third series were intended to study the operation of the precast cross-beam in the joint zone with the column, with maximum shearing force and the reference torque. They were made of column fragments 1.4 m high and 1.4 m pre-stressed parts (Figure 3), connected by monolithic concrete and tested using a cantilever scheme ( Figure 5). Load N 0 was created by a hydraulic jack and transmitted directly to one of the consoles. The opposite console was fixed from vertical displacements in a place symmetrical to the place where the load N 0 was applied. The places of application of the load and the consoles fastening when testing of the samples of 2-d series are chosen in such a way that the samples destruction occurs from the action of the supporting zone and the samples of the 3-d series -from the action of shearing force. The distance from the column side to the center of application of the force N 0 for the samples of the second series was 1.3 m, and for the samples of the third series -1.0 m. Results The cross beam parts test results. The first normal cracks in the part appeared in the middle of the span in the lower zone at a load of N 0 = 4.2 tons (60% of the fracture). The width of the crack was 0.05 mm, the height of crack propagation from the lower face was 50-80 mm, the average distance between the cracks along the length of the part was 250 mm, the deflection in the middle of the span was 18.4 mm. With gradual load increasing, there was the development of existing cracks and the formation of new normal ones occurred between the initially formed cracks, with an average crack pitch of 125 mm. At a load of 6 tons (85.7% of the destructive one), the width of crack opening reached 0.3 mm, and the deflection in the middle of the span was 57.4 mm. At a load of N 0 u = 7 t, there was a brittle sample fracture in the middle zone over the compressed concrete. The results of the first series samples testing The sample 1-1-400. The destruction of the sample occurred at a load of N 0 u = 270 kN in compressed concrete in the cross section in the middle of the span. At a load of 140 kN (51.7% of the destructive), in the middle of the span in the lower stretched zone, the first normal cracks appeared with a width disclosure of 0.2 mm to a height of 50-85 mm. The distance between the cracks along the length of the specimen was 250 mm. Under this load, the deflection in the middle of the span was 7.8 mm. Later, the existing cracks and the formation of new normal cracks between the existing ones took place, with the minimum MATEC Web of Conferences 196, 02021 (2018) https://doi.org/10.1051/matecconf/201819602021 XXVII R-S-P Seminar 2018, Theoretical Foundation of Civil Engineering crack pitch being 125 mm, and the height development did not exceed 250 mm (ie, normal cracks did not develop into the upper monolithic part). With a load close to destructive one, the width of crack opening reached 0.2-0.4 mm, the deflection in the middle of the span was 32.8 mm. It should be noted that during the test of the sample there was no cracking between the beam part and monolithic concrete. Sample 1-2-310. The sample destruction occurred with compressed concrete at a load of N 0 u = 170 kN. Under the load of 9 kN (52.9% of the destructive), in the middle of the span in the stretched zone, the first normal cracks appeared with a 0.2 mm wide opening to a height of 50-85 mm. The distance between the cracks along the length of the specimen was 250 mm. The deflection in the middle of the span was 12.7 mm. Later, the existing and the formation of new normal cracks took place, with the minimum crack pitch being 125 mm, and the height development did not exceed 250 mm (that is, the formed normal cracks did not spread to the upper monolithic part and the height development did not exceed 250 mm). Under the load of 15 kN (88.2% of the destructive), the crack opening was up to 0.3 mm, the deflection in the middle of the span was 38 mm. As in the previous sample, the formation of cracks between the beam part and monolithic concrete did not occur. The sample 2-3-400 was destroyed under the load of N 0 u = 329 kN. The destruction nature ( Figure 6) was different from the previous two. By the time of destruction in the sample, a sharp increase in the width of the inclined cracks occurred, followed by concrete crushing in the middle zone of the upper monolithic part. More over the monolithic concrete was destroyed over the upper working armature. When the destructive loads were reached in all the samples of the third series, the inclined cracks were sharply opened, followed by the destruction of the concrete of the cross beam part (Figure 7). By nature, it is a typical destruction on concrete strips between inclined cracks. Discussion The results of tests and check calculations for all samples are presented in Table 2. The check calculations of the tested prefabricated cross beams for determining theoretical destructive loads as well as comparing them with experimental values are performed both for reinforced concrete T-sections in accordance with hardening tests 8.1.8 ÷ 8.1.13 and 8.1.32, 8.1.33 [5] with the replacement of the calculated resistances of concrete to the corresponding average values, and the calculated resistances of the reinforcement to the corresponding normative values. https://doi.org/10.1051/matecconf/201819602021 XXVII R-S-P Seminar 2018, Theoretical Foundation of Civil Engineering Analysis of the data given in Table 2 shows that the calculation of the strength of normal sections of a precast cross beams can be carried out with sufficient reliability by the	2019-04-30T13:08:16.837Z	2018-01-01T00:00:00.000	{ "year": 2018, "sha1": "81b0c048bd822a95a3124e25cf98a93e3d6a4b39", "oa_license": "CCBY", "oa_url": "https://www.matec-conferences.org/articles/matecconf/pdf/2018/55/matecconf_rsp2018_02021.pdf", "oa_status": "GOLD", "pdf_src": "Adhoc", "pdf_hash": "d312df77649e47f56dffd1c7d23b3103a0ed4236", "s2fieldsofstudy": [ "Engineering", "Materials Science" ], "extfieldsofstudy": [ "Materials Science" ] }
59511931	pes2o/s2orc	v3-fos-license	Original Labial lesions by human bite -- Introduction: A bite from another human is an unusual cause of maxillo-facial traumatology. Our objective was to describe the epidemiologic and clinical characteristics of labial lesions caused by this kind of bite. Patients and methods: A descriptive prospective study was carried out at the University hospitals of Yalgado OUEDRAOGO and Blaise COMPAORE in Ouagadougou in Burkina Faso between June 2012 and May 2017. Results: We collected a sample of 28 patients with an average age of 32.5 years. The sex ratio was about 0.3. All the bites had been in ﬂ icted by women. The victims generally came from an underprivileged socioprofessional background. The bite had occurred during a brawl in 26 patients (92.9% cases) and in some context of aggression among 2 patients. The lesions were located on the lower lip among 21 patients, the upper lip among 4 patients, and along the commissure among 3 patients and resulted in a loss of tissue among 23 patients (82.1%). The treatment was surgical in nature, using trimming and labialization with satisfactory results in all cases. Conclusion: The circumstances surrounding the bites are not always clear and the lesions are almost always serious and require delicate treatment. Introduction Bites are a cause of maxillofacial trauma, and they most commonly are the result of animal bites, especially in children [1,2]. The most common human bites are self-inflicted accidentally during falls, epileptic seizures, or tic disorders, and often affect the tongue or the malar mucous membrane [3,4]. Voluntary human bites are most commonly reported in children during play [5]. Human bites in adults are uncommon and are poorly described [6]; in fact, most human bites are believed to not be reported in the literature [1,2,6]. Such bites are very serious traumas from the outset and deserve greater attention. This is because human bites are very often deep and generally result in tissue loss of varying degrees [7]. In addition, bacteria and viruses in human saliva result in infections in most cases [8]. The majority of human bites involve the face, and the most affected parts are the ear, nose, and lips [4,6,7]. Labial bites inflicted by a third party were the focus of our study, which aimed to describe the circumstances of occurrence, the clinical lesions, and the essential steps for managing this pathology. Patients and methods We conducted a prospective descriptive study referring to the medical records of patients seen and treated for labial lesions resulting from bites inflicted by a third party. This study was conducted in the departments of Stomatology and Maxillofacial Surgery at the Yalgado OUEDRAOGO and Blaise COMPAORE University Hospital Centers over a period of 5 years from June 2012 to May 2017. Our study population was a comprehensive nonprobability sample. Included in the study were all cases of labial bites inflicted by a third party and treated at the aforementioned hospitals during the study period. We examined the patients during the initial consultation and completed data collection by referring to hospital records, operating records, and patient medical reports. The variables studied were epidemiological (age, sex, occupation), clinical, and therapeutic. The data were analyzed using the French language version of the August 2008 edition of the EPI INFO 3.5.1 software. Results Our sample included 28 patients who consulted the hospital for a labial bite inflicted by a third party. The sample consisted on 22 women and six men, with a sex ratio of 0.3. The mean age of the patients was 32.5 years, with a range of 16-61 years. The bites had been inflicted by women in all cases. Victims were most often from disadvantaged socioprofessional backgrounds (Table I). The bite occurred during a fight in 26 patients (92.9% cases) and during an attack in two patients. Actual circumstances have not always been accurately reported because of patient discomfort. The most frequent were brawling in drinking establishments, workplaces, and neighborhoods (Table II). The perpetrator was unidentified in two cases. The consultation time was <6 h for nine patients, 6-12 h for 13 patients, and >12 h for 6 patients. The location of the injury was the lower lip in 21 patients, upper lip in four patients, and the commissure in three patients. These were contused wounds with irregular borders and often showed the visual imprints of the teeth in the tissue. Lip tissue loss was noted in 23 patients (82.1%). These tissue losses did not exceed one-third of the lip (Table III). In one case, the lip tissue fragment amputated by the aggressor was ingested by the victim. The amputated fragment was brought to the hospital by 17 patients. All patients received antibiotic treatment (amoxicillin + clavulanic acid combined with metronidazole) and an analgesic (paracetamol). Anti-tetanus prophylaxis was also performed with tetanus antitoxin and tetanus vaccine in all patients. Prophylaxis against viral hepatitis was performed in a patient who consented to a viral hepatitis B screening test. The surgical treatment consisted of careful tissue debridement after disinfection with hydrogen peroxide and povidoneiodine. A simple suture was performed in lesions without tissue loss but almost always preceded by debridement of the wound margins. In lesions with tissue loss, a labioplasty were required with only advancement flaps. The results were favorable in all cases ( Figs. 1 and 2). Labioplasty was performed under local xylocaine anesthesia in 17 patients and under general anesthesia in the operating room in 11 patients (day surgery). Postoperative follow-up was performed on an outpatient basis. Discussion Human bite injuries inflicted by a third party are relatively rare in the clinical context but deserve special attention. The prevalence and incidence of human bites are unknown and probably underestimated because not all cases are reported [6,[9][10][11]. Human bites are the third most frequent cause of bites in the United States after dog and cat bites [12]. The most common intentional human bites involve the face and hands [6,13,14]. In our study, these lesions were primarily confined to young adults, who were often female (average age 32.5 years with an age range of between 16-61 years, and a sex ratio of 0.3). A Nigerian study of human bite lip lesions also found that these lesions were predominantly found in young adults (mean age 32.6 years with a range between 21 and 54 years), but in their study both sexes were affected equally [6]. Although the circumstances showed variations, but the greatest number of bites had occurred during brawls in drinking establishments, similar to the findings in the EARDEY study [13] in England. In a Nigerian study, marital brawls and quarrels were also among the main causes [6]. The consultation time was relatively short, with 78.6% patients presenting for consultation within 12 h. This readiness for consultation would be related to the functional and esthetic discomfort related to labial lesions [2]. The lesion was most often located on the lower lip (75% lesions). It is the most common site of human bite lip lesions as described by most authors [4,2,6]. The lesions found were almost always severe because they were very deep with tissue loss in 82.13% cases. The depth of the lesions is relative to the length of the anterior teeth (incisors and canines), which are the sharpest and which can bite into the lip on both sides (cutaneous and mucous). When bitten, the opposing teeth of the two arches may come into contact, resulting in a section of the lip being severed entirely [7,15]. However, the tissue loss was not significant compared to the lesions described in the bites inflicted by other animals (dogs and donkeys) [1,16]. In fact, the tissue loss did not exceed one-third of the lip. The lesions were immediately sutured contrary to the classic bite treatment response [5,8,12,[17][18][19]. The anatomical and functional features of the lips help to explain this urgency. Of course, although the main function of lips is to act as a sphincter, they constitute an essential part of the face when it comes to static and dynamic esthetics. It follows that any lesion affecting this anatomical entity could have significant consequences on this dual roles of esthesis and functionality [2,20]. As a result, a wound in this region, with additional tissue loss, cannot be left open to heal for managed healing [6,20]. In addition, immediate repair has the advantage of improving the esthetic and functional results because the anatomical landmarks are always clearly visible. However, when the patient presents late with an already infected wound, this surgery should be postponed until the infection has subsided [6]. Every precaution must be taken to minimize the risks of infectious complications, which are very likely because human saliva generally contains several types of pathogens. In addition to bacteria, there are also some potentially transmissible viruses, such as hepatitis viruses, that should be considered in prophylactic management [4,8]. We opted for broad-spectrum antibiotic therapy against aerobic and anaerobic bacteria to minimize the risk of infectious complications. Conclusion Human bites were relatively infrequent in our context and young women were the most frequently affected. The main causes were brawls and fights, with the lesions primarily affecting the lower lip. Tissue losses were present in the most serious cases, along with a increased risk of infection. The surgical treatment was performed immediately, leading to promising results in all the patients. In emergencies, it is classically recommended never to suture a bite wound. However, in the case of the lip injuries, the wound is gaping and is located in an area where the esthetic and functional prognosis is unique. Therefore, immediate suturing should always be considered with strict asepsis for lip bites, along with prophylactic antibiotic therapy and appropriate prophylactic vaccination. Conflicts of interests: The authors declare that they have no conflicts of interest in relation to this article.	2019-01-31T14:09:26.229Z	2018-12-01T00:00:00.000	{ "year": 2018, "sha1": "0fc93cdbc5b21e4b04e54206889f72544fa015d3", "oa_license": "CCBY", "oa_url": "https://www.jomos.org/articles/mbcb/pdf/2018/04/mbcb170070.pdf", "oa_status": "GOLD", "pdf_src": "Adhoc", "pdf_hash": "5c8532619cc1ff3f98c65878a7b0a51ef404b78c", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
12967450	pes2o/s2orc	v3-fos-license	Usefulness of repetitive intraoperative indocyanine green-based videoangiography to confirm complete obliteration of micro-arteriovenous malformations Background: It is difficult to intraoperatively confirm the total disappearance of arteriovenous (AV) shunts during surgery for microarteriovenous malformations (micro-AVMs), especially when the nidus is extremely small or diffuse on preoperative angiography. Although intraoperative angiography is effective for evaluating residual shunts, procedure-related risks raise important concerns. The purpose of this study was to assess the usefulness of intraoperative indocyanine green-based videoangiography (ICG–VA) to determine complete disappearance of micro-AVMs during surgery. Methods: We retrospectively analyzed eight patients with ruptured micro-AVMs who were treated using craniotomy with ICG–VA at our institution. Results: Two patients underwent emergency partial evacuation of hematoma and external decompression before the diagnostic angiography. While three patients had a nidus smaller than 1 cm, five patients had only early draining veins without an appreciable nidus. The draining veins were superficial in six cases and deep in two cases. The average interval from onset to surgery was 33 days (range, 2–57). ICG–VA was repetitively conducted until disappearance of the AV shunt was confirmed. No residual AV shunt was observed on postoperative radiological examinations. In all cases, the diagnosis of AVM was confirmed from the results of postoperative pathological examination. Conclusions: ICG–VA could detect early draining veins more clearly in situ than diagnostic angiography. Although it is not as effective for visualizing lesions with deep draining veins, repetitive ICG–VA was safe and effective for confirming the disappearance of AV shunts with superficial drainage. INTRODUCTION Since Margolis emphasized small cerebrovascular malformations as an important cause of intracerebral hemorrhage (ICH), [35,36] such small vascular lesions as a cause of cerebral hemorrhage have been the focus of much debate. Small vascular malformations that are not appreciable on angiography have been referred to as "cryptic" [7] or "occult." [5,18,30] Yaşagil differentiated cerebral microarteriovenous malformations (micro-AVMs) from such angiographically occult arteriovenous malformations (AVMs) and defined them as AVMs with a nidus of 1 cm or smaller. [54] However, the clinical characteristics, namely, (i) significant neurological deficits despite the small lesion size, (ii) frequent incidence in young and previously healthy individuals, and (iii) difficulty in diagnosis, [1,2,6,36,44,51] and treatment strategies for angiographically occult AVMs and micro-AVMs are rather similar. With regard to diagnosis, visualization of these small lesions partly depends on the quality of imaging, [51] and the lesions may be transiently invisible under the influence of temporary local conditions around hematomas, such as compression or thrombosis. [11,33,51] These problems can lead to difficulties in intraoperative confirmation of arteriovenous (AV) shunt obliteration. Although intraoperative angiography has been recommended for AVM resection in general, [4,16,20,32,34] to our knowledge, its usefulness for micro-AVMs has not been discussed in the literature. Indocyanine green-based videoangiography (ICG-VA) has been recently used widely as a safe and simple method to intraoperatively evaluate vascular flow in situ, [3,19,31] and its use in AVM surgery has been reported. [14,15,21,25,28,38,40,45,46,55] We have been using intraoperative ICG-VA to identify the early filling vein in the case of micro-AVMs. Our experience led us to believe that ICG-VA will be particularly useful during micro-AVM surgery. In the present study, we present our observations on the use of ICG-VA to confirm AV shunt obliteration during micro-AVM surgery. Patients, diagnosis, and treatment strategies This study was approved by the Saitama Medical Center Institutional Review Board (No. 954). Between January 2011 and April 2014, 19 patients with AVM were treated at the Saitama Medical Center: There were 2 cases of unruptured AVM and 17 of ruptured AVM. In the present study, micro-AVM was angiographically defined as an AVM with a nidus smaller than 1 cm or as an early draining vein without an angiographically appreciable nidus. [1,42] An early draining vein was characterized as a vein identified in the capillary phase or earlier that have connections to larger veins in the drainage route. Micro-AVMs were detected in 10 of the 19 cases mentioned above, and all 10 patients showed hemorrhagic onset. The nidus or feeders were located in the deep white matter, close to the lateral ventricle in two patients, who were treated with stereotactic radiation; therefore, they could not be included in the study. In the present series, no patient with micro-AVM underwent endovascular embolization. The intracerebral hematoma volume was calculated from computed tomographic (CT) scan slices using the formula a × b × c/2. Our policy regarding the treatment of ICH in young patients or atypically located hypertensive ICH is to investigate the pathology of the hemorrhage using angiography as well as 3D-CT angiography and magnetic resonance (MR) imaging. When the patients' condition is critical because of the mass effect of the hematoma, emergency partial evacuation of the hematoma and external decompression are conducted before the angiography. Because a small nidus may not be visible even on angiograms recorded in the acute phase of hemorrhage, we carefully examined digital subtraction angiography images, especially for cases of lobar hemorrhage in young patients with no history of hypertension. If the result of the initial angiography performed immediately after onset is negative, angiography is repeated 4-6 weeks after hematoma onset. To delineate the early draining vein clearly on angiography, we obtain images with a short serial duration (12 shots per second) and a double dose of contrast medium. Patients diagnosed with micro-AVMs were scheduled for craniotomy and resection of superficial micro-AVMs. Surgical technique and intraoperative ICG-VA The patient's head was fixed in a 3-pin Mayfield headholder. The Stealth Navigation system (Medtronic, Surgical Navigation Technologies, Louisville, CO) was set up to confirm the location of the small nidus in the operative field. After standard craniotomy was performed to create ample space for surgery, ICG-VA was routinely conducted, and early draining veins were visualized. To compare early draining veins and normally irrigating veins, it was important to adequately expose the cortex. ICG dye diluted to 2.5 mg/mL was injected intravenously as a bolus (0.1 mg/kg of ICG dye, followed by 20 mL of saline). Vessels were visualized using surgical microscopes integrated with a video camera (OPMI PENTERO with INFRARED 800 microscope [Carl Zeiss, Oberkochen, Germany] or M720 OH5 [Leica Microsystems, Heerbrugg, Switzerland]). ICG-VA was performed to confirm the disappearance of the early draining vein when complete shunt obliteration was suspected. If the targeted early draining vein was still visualized on angiography, exploration of the small nidus remaining on the hematoma walls was continued. The ICG-VA was repeated until all the veins around the hematoma could be visualized in the normal venous phase. Postoperative angiography or MR imaging was performed 1-2 weeks after the surgery to confirm total obliteration of AV shunts. RESULTS The patient characteristics and outcomes are summarized in [ Table 1]. The mean age of the eight patients finally examined was 30.8 years (range, 6-79). The average interval from onset to surgery was 33 days (range, 2-57). The presentations of hemorrhage were ICH in five cases, ICH with intraventricular hemorrhage in two cases, and ICH with subarachnoid hemorrhage, and subdural hemorrhage in one case. The average volume of ICH was 23.4 mL (range, 5.4-59.2). Emergency partial evacuation of hematoma and external decompression before angiography were required in two of the eight cases since these two patients were in a deep coma because of the mass effect of the hematoma and urgently required hematoma evacuation. Per our policy regarding urgent surgery for life-threatening ICH accompanied by a high index of suspicion of AVM, the extent to which the hematoma was evacuated was adequate to reduce the intracranial pressure, and aggressive inspection of the hematoma cavity was avoided. In these two cases, angiography was conducted the day after the urgent partial hematoma evacuation, and definitive treatment for AVMs was administered 1 and 2 months after the first surgery. Only early draining veins without angiographically appreciable nidi were observed in five cases, while nidi of size <1 cm were identified in three cases. The draining veins were superficial in six cases and deep in two cases. No case showed two or more drainers on angiography. The location of the nidus was the frontal lobe in three patients and the parietal, temporal, occipital, temporoparietal lobes, and the cerebellum in one patient each. Intraoperatively, early draining veins could be identified in the initial ICG-VA, and changes in the timing of visualization (e.g. in the capillary or arterial phase) for all six cases of superficial drainers were repeatedly monitored. However, ICG-VA was not useful for evaluating deep drainers in the brain parenchyma. The pathological diagnosis made from the specimens obtained during surgery was compatible with AVM in all cases. Six patients showed complete recovery (modified Rankin Scale [mRS] 0), while two experienced moderate neurological deficits rated as mRS 2 and 3). The deficits were attributed to initial damage due to the hemorrhage and were not considered surgical complication. The average duration of follow-up was 5.2 months (range, 1.2-15.5). The disappearance of the AV shunt was postoperatively confirmed on angiography in seven cases and on MR images in one case. Illustrative case A 43-year-old male with no past medical history woke up one day with a severe headache, nausea, and subjective visual disturbance. On admission, neurological examination showed slight disturbance of consciousness and homonymous hemianopia on the left side. An emergency CT scan showed a 5.3 × 2.2 × 3.8 cm-sized ICH in the right parietal lobe [ Figure 1a]. The patient underwent emergency angiography, which showed no definitive vascular abnormality [ Figure 1b]. Time-of-flight MR angiography, however, showed a small but abnormally dilated vein near the hematoma [ Figure 1c]. The angiography was therefore repeated in 2 weeks, and it clearly showed an early draining vein near the hematoma [ Figure 1d]. On the basis of this finding, the hematoma was evacuated and the micro-AVM was explored. The patient was placed in the prone position, and right occipital craniotomy was performed. After the dura was opened, the first ICG-VA was conducted, and the early draining vein was detected in the cerebral cortex [ Figure 2a and b]. ICG-VA in the late venous phase showed the hematoma cavity as a dark area [ Figure 2c]. A small corticotomy was performed on the surface of this area to access the hematoma with minimal additional injury to the cortex. During hematoma removal, the surface of the brain parenchyma was meticulously examined under high magnification by rubbing the cavity wall with bipolar forceps, a suction tube, and cotton patties, with copious irrigation. Small tangled abnormal vessels were found on the hematoma wall and resected. When the inspection was largely considered complete, ICG-VA was repeated. In this second examination, the timing of early draining vein observation changed from the arterial phase to the normal venous phase [ Figure 2d and e]. However, another vein near the hematoma that had not been recognized as abnormal was clearly identified as an early draining vein, indicating that AV shunts remained in the vicinity (see Supplemental Video, which demonstrates how the delineation of early draining veins varies as the obliteration of AV shunt proceeds, 1 min 37 s). The exploration was therefore continued, and tiny, tangled thread-like vessels were found buried behind this vein. A final ICG-VA confirmed the complete disappearance of the early draining vein in the cortex [ Figure 2f]. Postoperative angiography showed no residual nidus [ Figure 2g]. The pathological diagnosis of abnormal vessels was AVM. The patient showed good recovery with no new neurological deficits and was discharged home. DISCUSSION In the present study, we examined the usefulness and limitations of ICG-VA during surgical treatment for micro-AVMs. We highlighted some diagnostic and operative tips specific for micro-AVMs and not large AVMs. Our technique using ICG-VA seem especially useful for intraoperative confirmation of complete AV shunt obliteration. Problems related to the preoperative diagnosis of micro-AVMs and the importance of in situ evaluation of early venous drainage using ICG-VA Despite their small size, micro-AVM can cause large hematomas, leading to serious neurological impairments. [1,2,6,44,51] In our series as well, the average hematoma volume was high enough to cause significant symptoms. Surgical resection is critical to treat ruptured micro-AVMs. [44] Many investigators have emphasized that the diagnosis of micro-AVMs requires a high index of suspicion; [1,2,6,18,44] yet, various conditions inherent to acute-stage hematomas opacify the AV shunt on imaging, whereby this indicator can be missed. [1,2,22,41] Increased local pressure may cause collapse of the nidus. Further, the partial thrombosis that occurred during hemostasis or posthemorrhagic vasospasm may also lead to obscuration of micro-AVMs. Therefore, a second angiography is important for seemingly common subcortical hemorrhage in order to identify occult vascular malformations undetected in the initial angiography. [22,33,42] Superselective angiography has also been reported to be useful in the detection of micro-AVMs that cannot be discerned on conventional angiography. [1,2,6,42,52] With recent innovations in imaging modalities, time-of-flight MR angiography [1,53] or cone-beam CT angiography [43] has also been reported to have equivalent or even better detection rates of micro-AVMs than conventional angiography. These preoperative radiological examinations and ICG-VA have not been compared previously. However, from our experience, in situ observation using ICG-VA is more sensitive than these imaging modalities to distinguish subtle differences between a normal vein and an early draining vein. This might be partly because of the lowering of local pressure after the dura is opened or increased shunt flow after evacuating the hematoma that results in better visualization. Elhammady et al. recommended elective surgical exploration to prevent re-bleeding in relatively young and healthy patients with unexplained ICH in accessible locations. [11] We believe that intraoperative ICG-VA should be considered for these explorative surgeries because it might be able to better delineate superficial draining veins. Effectiveness of ICG-VA in confirming total obliteration of AV shunts for micro-AVMs or diffuse AVMs Localization of the nidus and drainer is of paramount importance in micro-AVM surgery in order to ensure complete AV shunt obliteration, [6,44] and intraoperative navigation systems are helpful for identifying these lesions. [1] However, considerable brain shift may occur after hematoma evacuation, which would reduce the accuracy of navigation. In addition, the nidus in micro-AVMs or diffuse AVMs is often diffusely distributed in the brain parenchyma, making it difficult for surgeons to be certain that all abnormal vessels have been eliminated. In such cases, normalization of the early draining vein is usually utilized as an important index of total obliteration. We believe that intraoperative ICG-VA is the most sensitive and precise technique available for detecting early draining vein normalization. However, the effectiveness of ICG-VA specifically for micro-AVMs has not been discussed enough even in a recent large series of AVM treated using ICG-VA. [55] In most cases of compact AVMs with a medium-to-large nidus, surgeons are usually able to find the lesion easily and ensure complete obliteration of AV shunts by visual confirmation of changes in the color of the drainer. This empirical judgment can be rationally reinforced by intraoperative angiography, which may also help improve the obliteration rate. [4,19,34] With regard to micro-AVMs, however, the intraoperative visualization of AV shunts can be difficult because of the small size of these lesions. This also makes it difficult to confirm complete obliteration of the targeted AV shunt. To our knowledge, the clinical significance of intraoperative angiography for micro-AVMs has not been discussed previously, but given the preoperative difficulties in lesion delineation, its effectiveness would probably be limited. Identification of residual AV shunt after surgical resection is critical because partial resection of AVMs may increase the risk of hemorrhage. [23,24,26,37] In the present study, we achieved complete obliteration of micro-AVMs in all eight cases without the formation of new neurological deficits, and this outcome seems to support the efficacy and safety of intraoperative ICG-VA. Takagi et al. described the case of a 2-year-old patient with a diffuse AVM. [45] In their experience, repetitive ICG-VA was effective to ensure complete removal of residual nidus. Our results are in agreement with theirs, and we consider ICG-VA to be useful for diffuse AVMs as well as for micro-AVMs. Diffuse AVMs are more common in children than in adults, accounting for 18-25% of all pediatric AVMs. [12,27,32] Further, this high frequency of diffuse AVMs underlies the high recurrence rate of pediatric AVMs. [27,32] We speculate that this is partly related to the difficulty in intraoperatively confirming the complete obliteration of AV shunts in diffuse AVMs. We previously treated a 6-year-old girl with frontoparietal lobar hemorrhage, when ICG-VA was not available [ Figure 3a]. Angiography failed to reveal any abnormal vessels [ Figure 3b]. The hematoma was evacuated, but no apparent abnormal vessels were found during surgery. The patient underwent a second angiography 1 month after surgery, and again, no lesion was found. Eight years after the initial attack, she had recurrent hemorrhage in the same location [ Figure 3c], and angiography at this time finally showed a diffuse AVM [ Figure 3d]. After a thorough discussion of treatment options, she and her parents selected stereotactic radiosurgery. Complete obliteration of AV shunt was confirmed 2 years after radiosurgery. We believe that ICG-VA might be helpful in cases like this one, but its effectiveness to prevent recurrent AVM from micro-AVM and diffuse AVM that are frequently seen among young and pediatric populations should be carefully verified by additional studies with long-term follow-up. If the early draining vein had been detected on ICG-VA in situ during the evacuation surgery, we could have followed the vein and rigorously searched for hidden AV shunts. Zaidi et al. recently reported that the significance of intraoperative ICG-VA alone did not improve the identification of residual nidus and clinical outcome based on the data of 56 patients with AVM treated using ICG. [55] However, their study cohort consisted of mostly patients with larger AVMs, not micro-AVMs. In our experience, ICG-VA provides superior visual intelligence to differentiate an arterialized vein from a normal vein, and this may contribute to the discovery of indistinct AV shunts. We also recommend that ICG-VA be repetitively performed until all the veins around the hematoma appear normalized, as described in the illustrative case in this paper. Although the safety and usefulness of intraoperative angiography in AVM resection has been established, [4,8,9,17,29,39,49] the reported procedure-related complication rate is slightly high in pediatric cases (3.3% [13] and 4.8% [16] ). From this aspect as well, pediatric cases or surgeries conducted with the patient in the prone position would benefit greatly from ICG-VA. Limitations of ICG-VA There are some limitations regarding the use of ICG-VA. Deep-seated lesions are not visualized on ICG-VA. [15,55] In fact, we were unable to identify the early draining veins on ICG-VA in two cases because they were buried in the brain parenchyma. If the lesion has only a deep draining vein, the role of ICG-VA for confirming the disappearance of AV shunt would be minimal. Another limitation is that ICG-VA would find limited use if urgent surgery is required. Some patients with ruptured AVMs certainly require life-saving urgent surgeries before diagnostic angiography. It is recommended that if a vascular abnormality is suspected, the hematoma wall should be explored for ICH to reduce the risk of rebleeding. [10,47,48,50] However, we think that this approach is advocated after the possibilities of other high-risk hemorrhagic lesions are excluded by preoperative imaging studies. Nonspecific exploration without any information on the lesion may lead to bleeding and further injuries to the surrounding brain. Intraoperative ICG-VA may compensate for the limitations of preoperative angiography, but it provides little data on the overall angioarchitecture of vascular lesions. In our opinion, AVM resection during urgent operation should not be attempted based on ICG-VA findings alone, and angiography is essential. This should especially be avoided when the bleeding occurs in eloquent areas. Rebleeding rarely occurs between the initial hematoma evacuation and the subsequent curative resection of micro-AVM. [1] In our series, two patients (25%) required urgent hematoma evacuation and subsequently underwent curative surgery on days 29 and 57 after onset, without suffering from re-bleeding. We agree with Elhammady et al., [10] who recommended explorative surgery be planned 2-4 weeks after onset, if the patients' condition allows it, when the hematoma is semi-liquefied and can be easily evacuated without inducing further bleeding. CONCLUSIONS ICG-VA is a powerful tool for in situ localization of AV shunts of micro-AVMs with superficial drainage. It is also extremely useful for confirming complete obliteration of AV shunts for such micro-AVMs. We believe that repetitive ICG-VA contributes to reducing the chances of residual AV shunts, especially micro-AVMs.	2016-09-14T22:35:13.896Z	2015-05-21T00:00:00.000	{ "year": 2015, "sha1": "231c9db5890162ae32a7ddb7117510fa9329f1b9", "oa_license": "CCBYNCSA", "oa_url": "https://doi.org/10.4103/2152-7806.157445", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "231c9db5890162ae32a7ddb7117510fa9329f1b9", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
207852709	pes2o/s2orc	v3-fos-license	$q$-Plane Zeros of the Potts Partition Function on Diamond Hierarchical Graphs We report exact results concerning the zeros of the partition function of the Potts model in the complex $q$ plane, as a function of a temperature-like Boltzmann variable $v$, for the $m$'th iterate graphs $D_m$ of the Diamond Hierarchical Lattice (DHL), including the limit $m \to \infty$. In this limit we denote the continuous accumulation locus of zeros in the $q$ planes at fixed $v = v_0$ as ${\mathcal B}_q(v_0)$. We apply theorems from complex dynamics to establish properties of ${\mathcal B}_q(v_0)$. For $v=-1$ (the zero-temperature Potts antiferromagnet, or equivalently, chromatic polynomial), we prove that ${\mathcal B}_q(-1)$ crosses the real-$q$ axis at (i) a minimal point $q=0$, (ii) a maximal point $q=3$ (iii) $q=32/27$, (iv) a cubic root that we give, with the value $q = q_1 = 1.6388969..$, and (v) an infinite number of points smaller than $q_1$, converging to $32/27$ from above. Similar results hold for ${\mathcal B}_q(v_0)$ for any $-1<v<0$ (Potts antiferromagnet at nonzero temperature). The locus ${\mathcal B}_q(v_0)$ crosses the real-$q$ axis at only two points for any $v>0$ (Potts ferromagnet). We also provide computer-generated plots of ${\mathcal B}_q(v_0)$ at various values of $v_0$ in both the antiferromagnetic and ferromagnetic regimes and compare them to numerically computed zeros of $Z(D_4,q,v_0)$. Introduction We derive some exact results concerning the zeros in the complex q plane of the partition function, Z(D m , q, v), for the q-state Potts model on Diamond Hierarchical Graphs D m at various fixed values of a temperature-like Boltzmann variable v. We also derive exact results concerning the continuous accumulation set B q (v 0 ) of these zeros on the limiting Diamond Hierarchical Lattice D ∞ , again at various fixed values of v = v 0 , and we present computer-generated images of these loci. The Diamond Hierarchical Graphs D m are defined by starting with a graph D 0 consisting of two vertices (sites) and an edge (bond) joining them. The iterative graphical transformation replaces this single edge by four edges and two additional vertices, as shown in Fig. 1, yielding the next iterate, D 1 (which has the appearance of a diamond, whence the name). Fig. 1 also shows the next iterate, D 2 . We shall use the term Diamond Hierarchical Lattice (DHL) to refer to the (formal) limit lim m→∞ D m ≡ D ∞ . It is a self-similar, fractal object. We recall the procedure for calculating the Hausdorff dimension d H of a hierarchical lattice G ∞ . If the renormalization-group (RG) transformation reduces the length of each edge by a blocking factor of b and gives rise to N copies of the original graph, then N = b d H , so [1,2] d H = ln(N )/ ln(b). In the case of the iteration procedure for DHL, one has b = 2 and N = 4, yielding the well-known result that (1) d H (D ∞ ) = 2 . For this reason, we interpret the Diamond Hierarchical Lattice D ∞ as being two-dimensional. (See also [3,Appendix E.3] for an interpretation of the Diamond Hierarchical Lattice as an anisotropic version of the Z 2 lattice.) The q-state Potts model has been of longstanding interest in the area of phase transitions and critical phenomena. On a graph G, the partition function of this model, denoted Z(G, q, v), is a polynomial in two variables, q and (2) v ≡ y − 1 . In the original statistical physics formulation, q is a positive integer specifying the number of possible values of a classical spin defined at a given site of a lattice, σ(i) ∈ {1, ..., q}, and y is a non-negative temperature-like Boltzmann variable. (Throughout this paper we will primarily use the variable v because certain expressions are simpler in v rather than y.) As is evident from the expression (7) given below for this partition function, it is a polynomial in both q and v, and one can generalize both of these variables to complex values. Indeed, this generalization is necessary when analyzing the zeros of the partition function in the q plane for a given value of v and in the v plane for a given value of q. Part of the interest in the Potts model partition function stems from the fact that it is equivalent to a function of central importance in mathematical graph theory, namely the Tutte polynomial, T (G, x, y) (see Eq. (15) below). For some basic background on graph theory and the Tutte and chromatic polynomials, see, e.g., [4]- [9]. On a family of n-vertex lattice graphs, as n → ∞, an infinite subset of the zeros of Z(G, q, v) merge to form certain continuous loci. In this n → ∞ limit, we denote the continuous accumulation locus of zeros of Z(G, q, v) (i) in the complex q plane, for a given v = v 0 , as B q (v 0 ), and (ii) in the complex v plane, for a given q = q 0 , as B v (q 0 ). In this paper we will primarily be interested in the q-plane loci B q (v 0 ), however it will occasionally useful to relate them to the v-plane loci B v (q 0 ) and to discuss similarities and differences between these loci. Although no exact closed-form expression for Z(G, q, v) with general q and v, or for the corresponding dimensionless reduced free energy has been obtained on (the thermodynamic limit of) any regular lattice graph G of spatial dimension d ≥ 2, it has been possible to characterize the renormalization group (RG) action for the model exactly on certain hierarchical lattice graphs, including the Diamond Hierarchical Graphs. By performing a sum over spins at each iterative step, one can construct an exact RG transformation relating Z(D m+1 , q, v) to Z(D m , q, v ), where v is related to v according to a function v = F q (v) or equivalently, y = r q (y), (see Eqs. (20) and (21) below). This result follows because of the self-similarity of the Diamond Hierarchical Lattice. The properties of this model in the m → ∞ limit are then determined by the properties of the iterated function F q or equivalently r q . The properties of iterated analytic functions have been of considerable importance in mathematics (e.g., [10]- [11] and physics [12]- [13]). As will be clear, there are also interesting connections with complex analysis (see, e.g., [15]). There have been many studies of spin models on hierarchical lattices, including [14]- [39], which primarily analyze the zeros of Z(D m , q, v) in the complex plane of the temperature-like Boltzmann variable v. (A notable exception is [3] where the Lee-Yang (complex magnetic field) and Lee-Yang-Fisher (complex magnetic field and complex v simultaneously) zeros are studied for the Diamond Hierarchical Lattice.) It was natural for these previous works to focus on the v-plane zeros and their continuous accumulation set as m → ∞, B v (q 0 ), for a given q = q 0 , because it is directly related to the iteration of the RG transformation F q 0 (v) at fixed value of the parameter q 0 . Indeed, (in most settings) B v (q 0 ) is the Julia set in the v plane for the mapping F q 0 (v). Considerably less attention has been paid to the zeros of Z(D m , q, v) in the q plane and their continuous accumulation set as m → ∞, B q (v 0 ), at fixed values of v = v 0 . Rather than being the Julia set of a rational mapping, B q (v 0 ) is related to the parameter dependence of the iterates of F q (v) for the fixed choice of initial condition v = v 0 . (This will be elaborated in Section 5.) We have noted above the study of the zeros in the q plane for the Sierpinski gasket [38]. In the case of the Diamond Hierarchical Lattice the locus B q (−1) has been recently studied in [34], [39], and [40]. Wang, Qui, Yin, Qiao, and Gao [34] and Yang and Zeng [39] proved that the bifurcation locus M for the renormalization mapping F q (v) given in Eqn. (19) is connected. In Ref. [40], Chio and Roeder use techniques from complex dynamics to show that M ⊂ B q (−1) for the DHL and, in particular to prove for the DHL that the Hausdorff dimension of B q (−1) is 2. The paper [40] also provides a quantitative description of the limiting behavior of the chromatic zeros in terms of measure theory. In this paper we will study properties of the loci B q (v 0 ) at various choices of v 0 , beyond the case of the chromatic zeros v 0 = −1. Using techniques from complex dynamics similar to those in [40], we will make computer images (see of these loci, which we relate to numerical computations of the 172 zeros of Z(D 4 , q, v 0 ). We will also rigorously determine properties of the intersection between B q (v 0 ) and the real q-axis. The latter results are new even in the case of the chromatic zeros v = −1. We will also make use of results from statistical mechanics to gain further insight into the properties of B q (v 0 ). This paper is organized as follows. In Sections 2-4 we review some relevant background on the Potts model, the family of Diamond Hierarchical Graphs, D m , and the iterative RG transformation F q (v) that relates Z(D m+1 , q, v) to Z(D m , q, v ). In Section 5 we present some necessary background in complex dynamics and use it to relate the locus B q (v 0 ) to the "active parameters" q for the RG transformation. In Sections 6 and 7 we present our results on zeros of the partition function in the q plane for the Potts antiferromagnet at zero and finite temperature, respectively. We also state Theorem 7.1 describing the intersections of B q (v 0 ) with the real q-axis in this regime (−1 ≤ v 0 < 0). In Section 8 presents results on the zeros in the q plane for the Potts ferromagnet, including the statement of Theorem 8.1 describing the intersections of B q (v 0 ) with the real q-axis in this regime (v 0 > 0). Section 9 is devoted to proofs of Theorems 7.1 and 8.1. Section 10 contains our results on partition function zeros in the v plane for various values of q. Our conclusions are summarized in Section 11, and some auxiliary information is given in Appendix A. Background from Graph Theory and Statistical Physics In this section we discuss some relevant background from graph theory and statistical physics. A graph G = (V, E) is defined by its set V of vertices (= sites) and its set E of edges (= bonds). We denote n = n(G) = \|V \| and e(G) = \|E\| as the number of vertices and edges of G. At temperature T , the partition function of the q-state Potts model is given by with the Hamiltonian Here, the sum is taken over all edges e ij of G, with i and j labeling vertices of G; σ : V → {1, ..., q} is an assignment of classical spins to the vertices; β = (k B T ) −1 ; J is the spin-spin interaction constant; and k B is the Boltzmann constant [41]. Further, δ r,s is the Kronecker delta function. We define the notation The signs of J favoring ferromagnetic (FM) and antiferromagnetic (AFM) spin configurations are J > 0 and J < 0, respectively. Hence, the physical ranges of v are v ≥ 0 for the Potts ferromagnet (FM) and −1 ≤ v ≤ 0 for the Potts antiferromagnet (AFM). The partition function for the q-state Potts model can equivalently be written as Thus Z is invariant under a global symmetry that acts on the spins, namely for any permutation π q of {1, . . . , q} we can apply the mapping σ(i) → π q (σ(i)) to the spin at each site i, leaving Z unchanged. At high temperatures, this symmetry is realized explicitly in the physical states, while in the n → ∞ (thermodynamic) limit on a lattice graph with dimensionality greater than a lower critical dimensionality, it can be broken spontaneously with the presence of a nonzero long-range ordering of the spins. This ordering is ferromagnetic or antiferromagnetic, depending on the sign of J. A spanning subgraph of G is G = (V, E ) with E ⊆ E. The number of connected components of G is denoted k(G ). The partition function of the Potts model can equivalently be expressed in a purely graph-theoretic manner as the sum over spanning subgraphs [42] Eq. (7) shows that the partition function Z(G, q, v) is a polynomial in q and v with positive integer coefficients for each nonzero term. As is evident from Eq. (7), Z(G, q, v) has degree n(G) in q and e(G) in v, or equivalently, in y. Furthermore, Eq. (7) allows one to generalize the parameter q beyond the positive integers, Z + . In particular, for the ferromagnetic case v > 0, Eq. (7) allows one to generalize q from the positive integers to the positive real numbers while keeping a Gibbs measure, i.e., keeping Z(G, q, v) > 0. (This is not, in general, possible for the antiferromagnetic case, except when q is an integer, so one can revert to the Hamiltonian formulation in Eq. (4) and (6), since v is negative, so Z(G, q, v) contains terms of both signs.) More generally, Eq. (7) allows one to generalize both q and v from their physical ranges to complex values, as is necessary in order to analyze the zeros of Z(G, q, v) in q for fixed v and the zeros of Z(G, q, v) in v for fixed q. Since the coefficients in Z(G, q, v) are real (actually in Z + , but all we use here is the reality), it follows that for real v, the zeros of Z(G, q, v) in the q plane are invariant under complex conjugation and for real q, the zeros of Z(G, q, v) in the v plane plane are invariant under complex conjugation. Since k(G ) ≥ 1 for all G , it follows that Z(G, q, v) always contains an overall factor of q. We can thus define a reduced partition function which is also a polynomial in q and v Let us denote G ∞ as the formal limit, as n → ∞ on a family of graphs G n (here, G n = D m ). In this limit, the dimensionless, reduced free energy, per vertex, is defined as (The actual free energy is equal to −k B T f .) For the Potts antiferromagnet, T → 0 means K → −∞ and thus v → −1. As is clear from Eq. (6), the only spin configurations that contribute to Z(G, q, v) in this limit are those for which the spins on adjacent vertices are different. Hence, where P (G, q) is the chromatic polynomial, which, for q ∈ Z + , counts the number of ways of assigning q colors to the vertices of G subject to the condition that no two adjacent vertices have the same color (called proper q-colorings of G). The minimum integer q that allows a proper qcoloring of G is the chromatic number, χ(G). Since D m is bipartite, χ(D m ) = 2. Since P (G, q) always contains a factor of q, we also define Besides its intrinsic interest in mathematical graph theory, the chromatic polynomial is important for physics because of its connection with ground-state entropy. For a family G m , in the limit m → ∞ (and hence n → ∞), the configurational degeneracy per site (vertex) of the Potts antiferromagnet is For real q < χ(G m ), P (G m , q) can be negative; in this case, since there is no obvious choice for which of the n roots of (−1) to pick, one can only determine \|W (G m , q)\| [43]. For a set of special q values, {q s }, the limits n → ∞ and q → q s do not, in general, commute for P (G, q) 1/n , so one must specify the order of limits in defining W (G ∞ , q) [43,44]. The set {q s } depends on the family G m but usually includes q = 0 and q = 1. Here we define the order of limits as q → q s first and then n → ∞. For a wide class of G m families, if q is sufficiently large, then the number of proper q-colorings of G m grows exponentially with m and n, so that W (G ∞ , q) > 1 and hence the Potts AFM has nonzero ground-state entropy per vertex on G ∞ , S(G ∞ , q) = k B ln[W (G ∞ , q)] = 0. The Tutte polynomial, denoted T (G, x, y), of a graph G is defined by where, as above, k(G ) denotes the number of connected components of the spanning subgraph G , and c(G ) denotes the number of linearly independent circuits on G , given by c(G ) = e(G ) + k(G ) − n(G ). With y = e K , as defined in Eq. (5) and it follows that Thus, the partition function of the Potts model is equivalent, up to the indicated prefactor, to the Tutte polynomial on a given graph G. Zeros of Z(G, q, v) in q for a given v and in v for a given q are of interest partly because for many families of graphs, such as strips of regular lattices, in the m → ∞ limit, an infinite subset of these respective zeros typically merge to form certain continuous loci. As stated above, for a one-parameter family of graphs G m , we define the locus B q (v 0 ) as the continuous accumulation set of zeros of Z(G m , q, v 0 ) in the complex q plane as m → ∞ for a fixed complex value v = v 0 . (There may also be discrete zeros that do not lie on this locus.) Similarly, we define the locus B v (q 0 ), or equivalently, B y (q 0 ), as the continuous accumulation set of zeros of Z(G m , q 0 , v) in the complex plane of the temperature-dependent variable v, or equivalently, y, as m → ∞ for a fixed complex value q = q 0 [44]. For infinite-length, finite-width strips of regular lattices, and also chain graphs, B q (v 0 ) is generically comprised of real algebraic curves, including possible line segments [43]- [61] (for a review and further references, see [62]). The underlying reason for this is that P (G, q), and more generally, Z(G, q, v), for these classes of graphs consist of a sum of m'th powers of certain algebraic functions, denoted generically as λ j , where m is the length of the strip, and the continuous loci B q (v 0 ) occur at values of q where there are two or more λ j functions that are largest in magnitude and degenerate in magnitude. An early mathematical analysis of this sort of behavior was given in [46,45]. The Tutte polynomials for such strip graphs satisfy certain recursion relations [63]. The loci B q (v 0 ) may be connected, as, e.g., for the m → ∞ limit of the circuit graph and strips of the square, triangular, and honeycomb lattices with periodic or twisted periodic (Möbius) longitudinal boundary conditions, and corresponding strips with toroidal or Klein-bottle boundary conditions [47], [43]- [57]. These generically separate the q plane into different regions. For infinite-length limits of strip graphs and chain graphs with periodic longitudinal boundary conditions, B q (v 0 ) crosses the real q axis at a maximal point denoted q c (G ∞ ), as well as at one or more other points [43]. If q c (G ∞ ) ∈ Z + , this corresponds to the property that for this value of q, B v (q) passes through v = −1, signifying that the Potts antiferromagnet with q = q c (G ∞ ) has a T = 0 critical point on G ∞ . For a wide class of families of graphs, B q (v) also crosses the real q axis on the left at q = 0, but there are self-dual families where this left crossing is shifted to q = 1 [57]. In contrast, for infinite-length limits of lattice strip graphs with free longitudinal and transverse boundary conditions, B q (v 0 ) consists generically as (complex-conjugate pairs of) arcs and possible real line segments that do not separate the q plane into different regions [64]. In all of these cases, the Hausdorff dimension of B q (v 0 ) is 1. The result proved in [40] that the Hausdorff dimension of B q (−1) is 2 for the Diamond Hierarchical Lattice, thus provides an interesting contrast with the behavior of B q (−1) for these other infinite-n limits of families of graphs. The locus B v (q 0 ) is commonly also comprised of curves and possible line segments in the v plane. Although this is common behavior, it is not always true; a counterexample to this was found for the (4 · 8 · 8) Archimedean lattice, where these zeros form areas rather than curves even in the case of isotropic spin-spin interaction constants considered here [65]. These areas reduce to points where B v (q 0 ) crosses the real axis. Note that the property that B q (v 0 ) crosses the real q axis at a point q 0 does not imply that Z(G, q, v 0 ) vanishes at the point q 0 , and similarly, the property that B v (q 0 ) crosses the real v axis at a point v 0 does not imply that Z(G, q 0 , v) vanishes at that point v 0 . Indeed, from Eq. (7), it is evident that Z(G, q, v) is a polynomial in q and v with positive coefficients, and hence for fixed positive v, Z(G, q, v) has no zeros for any real positive q, and for fixed positive q, Z(G, q, v) has no zeros for any real positive v. The property that the continuous accumulation locus of the chromatic zeros B q (−1) crosses the real q axis at a point q 0 also does not imply that P (G, q 0 ) = 0, although here the argument is more subtle, since P (G, q) has terms that alternate in sign with descending powers of q. The precise meaning of the statement that, for a given v 0 , the continuous locus B q (v 0 ) crosses the real q axis at a point q 0 is that in the limit n(G) → ∞, the zeros of Z(G m , q, v 0 ) approach arbitrarily close to q 0 . This type of behavior is familiar from statistical physics. For example, for the q-state Potts model on the square lattice with integral q ≥ 2 one has that B v (q) crosses the real v axis at v c = √ q. This critical point separates the paramagnetic phase with 0 ≤ v ≤ v c with explicit S q symmetry from the ferromagnetically ordered phase with v > v c , in which the S q symmetry is spontaneously broken (e.g., [41]). Diamond Hierarchical Graphs D m and Diamond Hierarchical Lattice D ∞ In this section we discuss further details of Diamond Hierarchical Lattice graphs D m and the limit m → ∞. We have discussed above how one defines D m iteratively, starting with D 0 = T 2 , the tree graph with two vertices. The numbers of vertices and edges on D m are (16) n(D m ) = 2(4 m + 2) 3 and e(D m ) = 4 m . The degree, ∆, of a vertex on a graph G is defined as the number of edges that connect to it. (The word "degree" is thus used in two different ways, here, but this should not cause any confusion.) A ∆-regular graph is a graph all of whose vertices have the same degree. Although D 0 and D 1 are ∆-regular graphs with ∆ = 1 and ∆ = 2, respectively, the D m graphs with m ≥ 2 are not ∆-regular, but instead have vertices with degrees ranging from 2 to 2 m . For an arbitrary graph G, the average (effective) vertex degree is For the Diamond Hierarchical Graphs This limit is approached exponentially rapidly as m gets large. RG Transformations or equivalently, in terms of y = v + 1, The two mappings are conjugate under the change of variables y = v + 1. The iterative transformation (20) (or equivalently, (21)), embodies the action of the real-space renormalization group action here. Although we do not append a subscript to v or y , it is understood that these quantities are transformed at each iteration. We denote F 2 q (v) and r 2 q (y) as the functional composition, i.e., F 2 q (y) ≡ F q (F q (v)) and r 2 q (y) ≡ r q (r q (y)), respectively, and similarly for F m q (v) and r m q (y). Note that the transformation (20) is singular at v = −q/2 (and equivalently, (21) is singular at y = 1 − (q/2)), which is a physical antiferromagnetic value of v if q ∈ (0, 2]. For illustrative purposes, we record the expressions for Z(D m , q, v) for the first two values of m. Note that D 1 = C 4 , the circuit graph with four vertices. Elementary calculations yield (22) Z(D 0 , q, v) = Z(T 2 , q, v) = q(q + v) and In passing, it may be remarked that many expressions are simpler when written in terms of v instead of y. This is the case for the basic Eq. (7), and, for example, Z(D 1 , q, v) consists of 5 terms when written as a polynomial in q and v, but 10 terms when written in terms of q and y. For this reason, we shall generally express our results in terms of q and v. However, some formulas show an interesting structure when written in terms of y; for example, r q (y) is a perfect square. To keep matters simple, we will focus on the v variable throughout this paper and mention the y variable only when it is helpful in an explanation. In the special case v = −1, Eq. (22) and (23) yield the chromatic polynomials (24) P (D 0 , q) = q(q − 1) and The chromatic polynomial P (G, q) of any graph G has a zero at q = 0, and the chromatic polynomial of any graph G with at least one edge has a zero at q = 1. As we shall show, for D m and the limit D ∞ , the zero at q = 1 is isolated, while the zero at q = 0 occurs at one of the points where the continuous locus B q (−1) crosses the real q axis. The explicit expressions for Z(D m , q, v) become cumbersome to work with by hand for m ≥ 2, but computer algebra systems can work with them and solve for the zeros of Z(D m , q, v 0 ) (fixed v 0 ) and of Z(D m , q 0 , v) (fixed q 0 ) for m ≤ 4, at which point they become polynomials of degree 172 and 256, respectively. (Plots of such zeros are shown on the right-hand sides of Figures 3-16). 4.2. Preservation or reversal of the sign of J. It will be convenient to write F q (v) as the product of one factor, F q,1 (v), that is positive-semidefinite for real values of q and v, times another, F q,2 (v) that can have either sign: A basic question that one can ask about the RG transformation F q (v) is whether it keeps the sign of J invariant or reverses it. Recall that the physical ranges of v are v > 0 for the ferromagnetic sign, J > 0, and −1 ≤ v < 0 for the antiferromagnetic sign, J < 0, with v = 0 corresponding to J = 0 or infinite temperature, β = 0. Clearly, this transformation F q (v) maps v = 0 to v = 0. For nonzero v, as is evident from Eq. (20), the sign of v is determined by the sign of the factor F q,2 (v). Now, with q and v real, For physical q values, the inequality (27) holds for all ferromagnetic couplings, i.e. for all v > 0. However, the situation is different for antiferromagnetic couplings, i.e., v ∈ [−1, 0]. As v decreases from 0 to −1, the right-hand side of the inequality in (27) increases from 0 to 3/2. So for q > 3/2, which includes the usual integer values q ≥ 2, F q,2 (v) is positive, so the RG transformation F maps an antiferromagnetic v to a ferromagnetic v > 0. Further iterations of this RG transformation keep the coupling ferromagnetic. It should also be noted that if one considers q values down to, and including, q = 1, then one must take account of the fact that as v decreases through AFM values to −q/2, F q (v) diverges. This divergence occurs in the physical AFM interval v ∈ [−1, 0] if 0 < q ≤ 2, which includes the integer values 1 and 2. In particular, for the q = 2 (Ising) case, F 2 (v) maps the limit T → 0 for the AFM (i.e., v −1) to the T = 0 FM (v = ∞). For q ≥ 2, a small real positive value of v is mapped by F q (v) to a value v that is smaller than v (and positive). Thus, for small positive v, each iteration yields a smaller v and hence a higher temperature, so that the fixed point is v → 0 + . RG Fixed Points. A particularly important set of values of v is the set left invariant by the transformation, i.e., the values that are RG fixed points. Here we will focus on those fixed points occurring for real values of q and v. They are obtained by solving for the variable v in terms of the parameter q. Two fixed points, namely exist for every choice of q. The fixed point at v = 0 corresponds to infinite-temperature (equivalently, zero-coupling), where the spin-spin interaction has no effect. The fixed point v = ∞ can be interpreted as meaning lim v→∞ F q (v) = ∞, and it corresponds to zero temperature (equivalently infinite-coupling) in the ferromagnetic case. Besides the fixed points at v = 0 and v = ∞, solutions to the fixed point equation (28) correspond to values of v satisfying the following cubic equation The nature of the real roots of a cubic equation depend on the sign (or vanishing) of its discriminant ∆ 3 (see, e.g., [67]): Let us consider the possibilities as q increases from negative to positive values: (i) For q < 0 we have ∆ 3 < 0 and (29) has one real root and a complex-conjugate pair of non-real roots. In particular, for these values of q, the RG transformation F q (v) has only one additional fixed point (other than v = 0, ∞). (ii) For q = 0, Eq. (29) has a triple root at v = 0, which therefore coincides with the fixed point of F q (v) already discussed at the beginning of this subsection. (Note that the degree of F q (v) drops from 4 to 2 at this parameter value, producing a dramatic change in the mapping.) (iii) For 0 < q < 32 27 , we have ∆ 3 > 0, and (29) has three real solutions, corresponding to three additional real fixed points of F q (v). (v) For q > 32 27 we again have ∆ 3 < 0, and so the same description as Case (i) applies. We summarize this discussion with Figure 2. The additional fixed points of F q (v) have a special interpretation when q = 32/27. The fixed point at v = 16/9 = 1.77778 is the critical value of v for a Potts ferromagnet with this value of q = 32/27, and the double root at v = −8/9 corresponds formally to a finite-temperature antiferromagnet. We use the word "formally" here, because for non-integral q the partition function of the Potts antiferromagnet does not, in general, define a Gibbs measure and hence a normal statistical physics system. from a paramagnetic (PM) phase with manifest S q symmetry at high-temperatures T > T c,P M −F M , i.e., 0 ≤ v ≤ v c,P M −F M , to a low-temperature phase with ferromagnetic (FM) long-range order (magnetization) and associated spontaneous symmetry breaking of the S q symmetry to S q−1 for T < T c,P M −F M , i.e., v > v c,P M −F M . We can generalize this to a phase transition from a paramagnetic to ferromagnetic phase for q not restricted to integers ≥ 2 but instead taking on any real value q > 32/27, since, as discussed above, Z(G, q, v) defines a Gibbs measure for real positive q and v ≥ 0. We recall that, on a regular lattice (in the thermodynamic limit), a standard Peierls argument can be used to prove that a discrete spin model without frustration, competing interactions, disorder, or dilution has a finite-temperature phase transition if the spatial dimension of the lattice is greater than the lower critical dimensionality, d = 1. Although a fractal lattice is not a regular lattice in the conventional sense, arguments have been given [1,2] that if the Hausdorff dimensionality of the fractal lattice (in the n → ∞ limit) is greater than d , then a discrete model (without frustration, competing interactions, disorder, or dilution) will also have a finite-temperature phase transition on the fractal lattice. This conclusion applies to our present case, since the Hausdorff dimension d H (D ∞ ) = 2 given in Eq. (1) is greater than 1. On a regular lattice, the PM-FM phase transition in the q-state Potts model is second order, with a divergent correlation length, if q ∈ (0, 4] and first-order, with a nonzero latent heat, if q > 4 (e.g., [41] and references therein). As discussed above, in this FM case, one can generalize q here from positive integers to positive real numbers for the ferromagnetic case while retaining a Gibbs measure, and one may assume this generalization here. In the limit q → 0, one must take account of a relevant noncommutativity [44]; one can define a nonvanishing free energy if one takes n → ∞ first, and then q → 0, and with this order of limits, there is again a second-order phase transition. An alternate approach to the q = 0 case is to deal with the reduced partition function, Z r (D m , q, v). For q ≥ 32/27, the physical root of Eq. (29) is given by (the real, positive number) where the discriminant ∆ 3 was given in Eq. (30). The corresponding physical temperature is given by Eq. (31). The expression for v c,P M −F M in Eq. (32) is a monotonically increasing function of q for q ≥ 32/27. This is understandable physically, since the larger q is, the more statistical fluctuations there are, so one must cool the system to a lower temperature, i.e., a larger value of K and hence v, for it to undergo the phase transition to a phase with ferromagnetic order. In addition to the special values q = 0 and q = 32/27 for which Eq. (29) factorizes into three linear factors, there are also values of q for which it factorizes into a linear factor times a quadratic factor in v, so that the expressions for all of the roots simplify considerably. For example, for q = 1, Eq. (29) factorizes ; and so forth for certain larger values of q. We list some illustrative values of v c,P M −F M (q) in Table 1. For our analysis, we also display the other two solutions in v of the cubic equation (29), where the subscripts ± correspond to the ± signs in front of the factor of i. Although we initially chose v c,P M −F M to be the unique root of Eq. (29) for q ≥ 32/27, the formula (32) uniquely determines a fixed point of F q (v) for all q ∈ R, which we will continue to refer to as v c,P M −F M . We remark that, extending to complex values, \|v c,P M −F M \| ∼ \|q\| 2/3 as \|q\| → ∞. We refer the reader to Figure 2 where the three fixed points v − , v + , and v c,P M −F M are labeled for varying q. Remark that they are ordered by for those q where all three exist and are distinct, i.e. for 0 < q < 32 27 . The effect of the RG transformation Eq. (20) on v can thus be explained physically. For the hightemperature region of the ferromagnet, T > T c,P M −F M , i.e., the interval range 0 < v < v c,P M −F M , the RG transformation Eq. (20) maps the initial value of v to a smaller value of v , with the RG fixed point being at T = ∞, i.e., v = 0. This is the standard attractive infinite-temperature fixed point of the real-space renormalization group in statistical physics and reflects the fact that a critical point is a repulsive fixed point of the RG in the temperature direction. In turn, this is a consequence of the fact that the real-space RG blocking transformation with a blocking factor of b reduces the correlation length as ξ → ξ/b, which is reduced finally to ξ = 0 at T = ∞. If, in contrast, the initial (real) value of the temperature is less than T c,P M −F M , i.e., v is greater than v c,P M −F M , then the RG transformation Eq. (20) maps v to a larger value of v , with the RG fixed point in this phase being the zero-temperature fixed point where again the correlation length (defined by the connected spin-spin correlation function) vanishes. We remark that since D m is bipartite, the PM-AFM critical point is given by As q increases above 32/27, the other two roots of Eq. (29), which formed a double root at v = −8/9, bifurcate into a complex-conjugate pair of roots, which move to the upper and lower left away from the real axis. Asymptotically, As noted above, for values of q in the interval 0 ≤ q ≤ 32/27, Eq. (29) has three real roots. As q decreases below the value 32/27, the real root v c,P M −F M decreases below 16/9, and the double root at v = −8/9 bifurcates into two real roots, with v − decreasing and v + increasing as q decreases toward 1. When q reaches 1, Eq. (29) factorizes as ( . Of these solutions, the first, v = −1, corresponds formally to the T = 0 Potts antiferromagnet; the second, v − = −0.6180.., corresponds to a finite-temperature Potts antiferromagnet, and the third, v + = v c,P M −F M = 1.6180.., corresponds to a finite-temperature Potts ferromagnet. Rigorous Results from Complex Dynamics In order to prove rigorous results about the continuous accumulation loci B q (v 0 ) for various values of the temperature-like Boltzmann variable v 0 and to make computer pictures of it, we will need some results from complex dynamics. We will first describe them in a general context and then specialize to the case of the renormalization mapping F q (v). 5.1. Generalities on marked points and the passive/active dichotomy. Let f λ (z) denote a family of rational mappings f λ : C → C of the Riemann Sphere C depending holomorphically on a complex parameter λ ∈ Λ. For our purposes, Λ is an open subset of the complex plane C. Let a(λ) be a choice of initial condition for the iterates of f λ that depends holomorphically on the parameter λ. It is called a "marked point", and historically [72,73] this theory was used for marked points that are critical points of the rational mapping, in order to understand bifurcations of the mapping itself. We remark that none of the results presented in this subsection are new, and the proofs presented below are adaptations of those from the classical papers to our current context and notations. Following the terminology of McMullen [74], the marked point a(λ) is "passive" at parameter λ 0 if there is an open neighborhood U of λ 0 on which the the sequence of functions λ → f m λ (a(λ)) forms a normal family, in the sense of Montel's Theorem (see, e.g., [15] and references therein). As before, the notation f λ (a(λ)) denotes functional composition, f 2 λ (a(λ)) = f λ (f λ (a(λ))), and so forth for higher m. The set of all passive parameters is open and is called the passive locus for the marked point a(λ). If the marked point a(λ) is passive at the parameter λ = λ 0 , then the behavior of the initial condition a(λ) does not change much as λ varies in a small neighborhood of λ 0 . A parameter value λ 0 is defined as "active" if it is not passive. At these active parameter values, the initial condition a(λ) undergoes quite different dynamical behavior as λ is varied. Roughly speaking, this is analogous to the notion of bifurcation in the theory of dynamical systems, but the global behavior of the mapping f λ need not change at λ 0 , even if the marked point a(λ) is active at λ = λ 0 . One can also think of the passive locus as a "parameter space analog of the Fatou set" that is associated to the marked point a(λ) and the active locus as a "parameter space analog of the Julia set" that is associated to the marked point a(λ) [76]. Let us describe a simple way for a parameter λ 0 to be active. Suppose that z • is a repelling fixed point for f λ 0 . Then z • can be holomorphically continued to be a repelling fixed point z • (λ) of f λ for all λ in some neighborhood U of λ 0 . One says that "f n 0 maps the marked point a(λ) non-persistently onto the repelling fixed point z • (λ) at parameter λ 0 " if In this case, it is easy to show that λ 0 is an active parameter for the marked point a(λ) under f λ . (The same holds if f n 0 maps the marked point a(λ) "non-persistently" onto a point from a repelling periodic cycle at parameter λ 0 .) Lemma 5.1. The set of active parameters of the marked point a(λ) under the mapping f λ contains no isolated points. Proof. Suppose that λ 0 is an active parameter and U is any neighborhood of λ 0 . Since repelling periodic points are dense in the Julia set J(f λ 0 ) we can choose a repelling periodic cycle z 1 , z 2 , . . . , z k of period k ≥ 3 for f λ 0 that is disjoint from a(λ 0 ). Restricting U to a smaller neighborhood of z 0 , if necessary, we can suppose that this periodic cycle varies holomorphically as forming a repelling cycle of period k of f λ for all λ ∈ U . Since λ 0 is active, Montel's Theorem implies that there is some parameter λ 1 ∈ U and some iterate n 0 such that f n 0 λ 1 (a(λ 1 )) = z j (λ 1 ) for some 1 ≤ j ≤ k. Since a(λ 0 ) = z (λ 0 ) for all 1 ≤ ≤ k we see that f n 0 maps the marked point a(λ) non-persistently onto the repelling periodic cycle from Eq. (36) at parameter λ 1 . Therefore, λ 1 ∈ U is another active parameter. Recall that a point b ∈ C is "exceptional" for a rational map f : The key statement we need from holomorphic dynamics is the following simple lemma: Suppose f λ is a holomorphically varying family of rational maps, that a(λ) and b(λ) are marked points, and that b(λ) is not persistently exceptional for f λ . Then, if λ 0 is an active parameter for a(λ) under f λ we have The proof is classical, but we include it here for the convenience of the reader, closely following [75, Proposition 3.5]. Proof. Suppose for contradiction that a(λ) is active under f λ at parameter λ 0 and that there is some neighborhood U of λ 0 such that f m λ (a(λ)) = b(λ) for all λ ∈ U and all n ≥ 0. (37) Since b(λ) is not persistently exceptional, there is some ≥ 1 such that f − λ (b(λ)) contains k ≥ 3 points for all but finitely many possible values of λ, for which it has fewer than k points. If λ 0 is not one of those values, we can work in sufficiently small neighborhood U ⊂ U of λ 0 over which f − λ (b(λ)) consists of disjoint graphs of holomorphic functions of λ. Together with Assumption (37) and Montel's Theorem, this implies that λ → f m λ (a(λ)) forms a normal family on U , contrary to the hypothesis that λ 0 ∈ U is active. If it happened that λ 0 is one of the finitely many parameters for which f − λ (b(λ)) has fewer than k points, then we use the fact that active parameters are not isolated (Lemma 5.1) to replace λ 0 with another active parameter λ 1 for which f − λ 1 (b(λ 1 )) has the maximal number of preimages k. We then apply the reasoning from the previous paragraph to the new active parameter λ 1 . The following general classification of the types of behavior of a marked point a(λ) for the case in which λ is in the passive locus will be helpful for our discussion: Dujardin-Favre Classification of Passivity Locus [71,Theorem 4]. Let f : Λ × P 1 → P 1 be a holomorphic family and let a(λ) be a marked point. Assume U ⊂ Λ is a connected open subset where a(λ) is passive. Then exactly one of the following cases holds: (i) a(λ) is never preperiodic in U . In this case the closure of the orbit of a(λ) can be followed by a holomorphic motion. (ii) a(λ) is persistently preperiodic in U . (iii) There exists a persistently attracting (possibly superattracting) cycle attracting a(λ) throughout U and there is a closed subvariety U U such that the set of parameters λ ∈ U \ U for which a(λ) is preperiodic is a proper closed subvariety in U \ U . (iv) There exists a persistently irrationally neutral periodic point such that a(λ) lies in the interior of its linearization domain throughout U and the set of parameters λ ∈ U for which a(λ) is preperiodic is a proper closed subvariety in U . We do not include a proof of this rather difficult theorem. 5.2. Application to F q (v) and the q-plane zeros B q (v 0 ). We will now explain how to use the techniques from the previous subsection to study the q-plane zeros for the DHL. (The reader may wish to compare this discussion with that from [40], but note that in that paper the variable y = v + 1 is used instead of the equivalent variable v.) Recall that q = 0 is always a zero for Z(D m , q, v). The renormalization procedure from Eqs. (19) and (20) implies that for any v 0 ∈ C and any q = 0 we have (20). Therefore, for our purposes, the marked point a(q) will correspond to the desired choice of q-plane and will be constant: a(q) ≡ v 0 . (For example a(q) ≡ −1 will correspond to the case of chromatic zeros.) Meanwhile, the other marked point will be b(q) = −q. We remark that the degree of F q (y) drops from 4 to 2 when q = 0 due to the appearance of a common factor of v in the numerator and denominator. This is the only parameter where such a drop in degree occurs, and therefore, F q (v) is a holomorphic family of rational maps with parameter space Λ = C \ {0}. In fact, the entire discussion in the remainder of this section will only pertain to q ∈ C \ {0}. For any v 0 ∈ C let us denote the active locus of the marked point a(q) ≡ v 0 by A q (v 0 ). , so that b(q) = −q is not persistently exceptional for F q (v). Therefore, Lemma 5.2 implies that any point of A q (v 0 ) is in the accumulation set of solutions to F m q (v 0 ) = −q and hence the accumulation locus of zeros for Z(D m , q, v 0 ), both considered as m → ∞. By Lemma 5.1, A q (v 0 ) contains no isolated points, so we conclude that A q (v 0 ) ⊂ B q (v 0 ). We will now make some observations that will help when drawing computer pictures of A q (v 0 ) and that will also help us to further relate A q (v 0 ) to B q (v 0 ). For every complex parameter q = 0 we have that: (i) The point v = 0 is a superattracting fixed point for F q (i.e., F q (0) = 0 and F q (0) = 0). (ii) The point v = ∞ is a superattracting fixed point for F q (i.e., F q (∞) = ∞ and F q (∞) = 0, with the derivative computed in suitable local coordinates centered at ∞.) Therefore, for every q = 0 there is an open neighborhood W s q (0) consisting of initial conditions v 0 whose orbit under F m q converges to 0 and similarly an open neighborhood W s q (∞) consisting of initial conditions v 0 whose orbit under F m q converges to ∞. These neighborhoods depend continuously on the parameter q. Let Each is a subset of the passive parameters for the marked point a(q) ≡ v 0 and each is an open set. Proof. The proof is similar to that of Lemma 5.1. Suppose for contradiction that q 0 ∈ A q (v 0 ) and suppose there were an > 0 such that the disc D (q 0 ) were disjoint from P 0 q (v 0 ). Then, there exist values v = v • and v = v * close to v = 0 such that v • , v * ∈ W s q (0) for all q ∈ D (q 0 ). In this case the family of holomorphic functions Montel's theorem would then imply that this is a normal family on D (q 0 ), contradicting that q 0 is active. Therefore . If q 0 were a passive parameter, then there is an > 0 such that the sequence of functions given in Eq. (38) forms a normal family on D (q 0 ). However, D (q 0 ) contains points of P 0 q (v 0 ), so the identity theorem for holomorphic functions implies that the sequence of functions given in Eq. (38) converges to 0 on all of D (q 0 ). Therefore, q 0 is in the interior of P 0 q (v 0 ) contradicting the hypothesis that q 0 ∈ ∂P 0 q (v 0 ). We conclude that q 0 ∈ A q (v 0 ) and hence that ∂P 0 is the same, so we omit it. If the marked point a(q) ≡ v 0 were to be preperiodic on some connected component of the passive locus, then it would be preperiodic to the same periodic cycle for all q ∈ C \ {0}, by the identity theorem for holomorphic functions. However, there are parameters q 0 and q ∞ for which a(q) ≡ v 0 maps onto the two different fixed points 0 and ∞, so this is impossible. The conditions that q ∈ P 0 q (v 0 ) or that q ∈ P ∞ q (v 0 ) are easy to check numerically on the computer, so Lemma 5.4 makes visualization of the active locus A q (v 0 ) easy. See, for example, Figures 3-12. Proof. Over any compact K ⊂ P 0 (Here we used that q = 0.) This implies that B q (v 0 ) is disjoint from P 0 q (v 0 ). The same holds for P ∞ q (v 0 ) using identical reasoning. The following proposition summarizes Lemmas 5.3 -5.6: Proposition 5.7. Within C\{0} the continuous accumulation locus B q (v 0 ) of zeros for Z(D m , q, v 0 ) as m → ∞ contains all parameters in the boundary of P 0 q (v 0 ) and all parameters in the boundary of P ∞ q (v 0 ) (these boundaries are equal). Moreover, On the possibility that . For a given choice of v 0 it may be possible to have A q (v 0 ) B q (v 0 ) as a result of there being components of the passive locus for a(q) ≡ v 0 other than those in P 0 q (v 0 ) ∪ P ∞ q (v 0 ). In our computer-generated pictures (Figures 3-12) for many values of v 0 one can see subsets of points colored black, corresponding to the condition that F m q (v 0 ) → 0 and F m q (v 0 ) → ∞. Any connected component of the interior of this black subset will be a passive component for the marked point a(q). We will see that some of these components correspond to the marked point having orbit attracted to attracting periodic orbits for F q other than v = 0 and v = ∞. For such components, one can use similar reasoning to the proof of Proposition 5.7 to rule out points of B q (v 0 ). However, the other behaviors (i) and (iv) described in the Dujardin-Favre classification of the passive locus could lead to points of B q (v 0 ) that are within these black regions. (Behavior (ii) is ruled out by Lemma 5.5.) This is similar to the situation for the continuous accumulation locus of zeros B v (q 0 ) in the v-plane for a fixed q 0 . So long as v = −q 0 is not an exceptional point of F q 0 it follows from Montel's theorem the Julia set of F q 0 satisfies J(q 0 ) ⊂ B v (q 0 ). However, it can be possible to have J(q 0 ) B v (q 0 ). For example, this will happen if: (i) F q 0 has a Siegel Disc D in its Fatou set (i.e. a component of the Fatou set on which F q is conjugate to an irrational rotation), (ii) v = −q 0 ∈ D, and (iii) v = −q 0 is not equal to the unique fixed point of F q that is in D. Then the closure of its forward orbit of v will form a simple closed curve γ ⊂ D, because the dynamics of F q is conjugate to an irrational rotation on D. This leads to some of the iterated preimages of v under F q 0 accumulating on every point of γ, implying that γ ⊂ B v (q 0 ) \ J(q 0 ). For the v-plane zeros, this issue can be handled by considering only the locus B v (q 0 ) where a positive proportion of the zeros of Z(D m , q 0 , v) accumulate. So long as −q 0 is not an exceptional point for F q 0 (v) it is a consequence of the Lyubich and Friere-Lopes-Mañe Theorems [77,78,79] that B v (q 0 ) = J(q 0 ). In other words, any possible zeros of Z(D m , q 0 , v) occurring in the Fatou set of F q 0 do so with arbitrarily small proportion, in the limit m → ∞. This quantitative approach can also be taken in the q-planes. Consider the locus B q (v 0 ) ⊂ B q (v 0 ) where a positive proportion of the zeros from Z(D m , q, v 0 ) accumulate. For rational v 0 (and even any algebraic number v 0 ) it is a consequence of Theorem C' from [40] that A q (v 0 ) = B q (v 0 ). In other words, any possible zeros of Z(D m , q, v 0 ) occurring in the passive locus of a(q) ≡ v 0 for F q do so with arbitrarily small proportion, in the limit m → ∞. 5.4. On thinking in C 2 . It is very helpful to think about the continuous accumulation loci of zeros for Z(D m , q, v), m → ∞, as being a single object in C 2 , with the loci B q (v 0 ) being horizontal slices and B v (q 0 ) being vertical slices of the same object. This allows one to gain insight about B q (v 0 ) by looking at B v 0 (q) near q = q 0 and vice-versa. While this is very good intuition, additional care must be taken to ensure the results are rigorous, as the following delicate example shows. It is very natural to expect that However, there are some delicate situations where this potentially might not hold. Suppose that there is a component P of the passive locus for the marked point a(q) ≡ v 0 under F q (v) for which a(q) ∈ J(q) (Julia set for F q ) for all q ∈ P. (This would correspond to Case (i) of the Dujardin-Favre classification, since we've ruled out Case (ii) by Lemma 5.5.) Then, v 0 ∈ B v (q) for all q ∈ P because J(q) ⊂ B v (q). If there are some parameters q ∈ P for which b(q) = −q is in the basin of attraction of an attracting cycle of F q , then such parameters would form an open subset P 0 of P. For any q ∈ P 0 we cannot have F m q (a(q)) = b(q) because the Julia set and Fatou set are complementary and invariant under F q . We would therefore have Z(D m , q, v 0 ) = 0 on the open set P 0 for all m ≥ 0. In particular, such q are not in B q (v 0 ) and Eq. (39) would fail. We do not have an explicit example of this problematic behavior for the RG mapping F q (v). However, to avoid such delicate issues, we will work with the techniques from Section 5.2 involving the active parameters A q (v 0 ). (See also Appendix A for discussion of an additional problem that happens when q = 0.) 6. Zeros in the q Plane for v = −1: Chromatic Zeros In this section we study the chromatic polynomial P (D m , q) = Z(D m .q, −1) and its zeros in the complex q plane, i.e., the chromatic zeros of D m and their continuous accumulation set B q (−1) for m → ∞. The left side of Figure 3 shows a computer-generated image of the regions Any point that is not in one of these two sets is colored black. Because the sets P 0 q (−1) and P ∞ q (−1) are open, the set of points colored black is closed. According to Proposition 5.7, B q (−1) contains any point of the boundary between the white, blue, and black sets. It may contain additional points, but they are necessarily in the interior of the black set. The right side of Figure 3 shows a plot of the 171 zeros of Z r (D m .q, −1) computed numerically in Mathematica. (We have omitted the zero at q = 0, hence the subscript r indicating that we consider the reduced partition function.) According to Lemma 5.4 the interior of the black set is contained in the passive locus for the marked point a(q) ≡ −1 under F q . Let us consider what happens on the connected components R 1 , R 2 , and R 3 of the interior of the black set that are labeled on Figures 3 and 4. We begin with the component labeled R 1 , containing the point q = 0.5. It intersects the real q axis in the interval (0, 32/27). For q ∈ (0, 32/27), the fixed point v − (q) given in Eq. (34) is an attracting fixed point for F q that differs from 0 and ∞. Furthermore, this fixed point v − (q) can be analytically continued for all q ∈ R 1 and it remains an attracting fixed point of F q at these q values. For all q ∈ R 1 the orbit of initial condition a(q) ≡ −1 converges to v − (q), which is an example of passive behavior for this marked point. Now consider component R 2 . We have performed numerical computer studies that indicate that for every q ∈ R 2 , the RG mapping F q has a periodic orbit of period 3 that is attracting. For values of q ∈ R 2 , the orbit of a(q) ≡ −1 under F q converges to this attracting cycle. Finally, consider component R 3 , which is the cardioid of the "baby Mandelbrot set" in the upper right corner of the left half of Fig. 3 that is also shown in a magnified view in Fig. 4. Numerical experiments show that for all q ∈ R 3 , the RG mapping F q has an attracting periodic cycle of period 2, and that the marked point a(q) ≡ −1 has orbit converging to it. The complexity of this region diagram is evident, even with the finite resolution of Fig. 3. One reason for the complexity is the appearance of baby Mandelbrot sets. We show one of them in Fig. 4, corresponding to a magnified view of the region enclosed by a red box in Fig. 3. However, there are infinitely many baby Mandelbrot sets in Fig. 3. Their existence can be explained using complex dynamics renormalization theory: Theorem 6.1 (McMullen 1997 [74]). Suppose f λ (z) is a holomorphic family of rational maps and c(λ) is a marked critical point for f λ (z). If there is at least one parameter λ 0 so that the marked critical point c(λ) is active under f λ (z), then the active locus for the marked critical point c(λ) contains quasiconformal copies of the the Mandelbrot set (or possibly of the degree d > 2 generalization thereof ). In particular, it follows from the work of Shishikura [80] that the active locus of c(λ) has Haudorff dimension equal to 2. With regard to the application of McMullen's Theorem to our F q (v) in Eq. (20) with marked point a(q) ≡ −1, one must note that a(q) = −1 is not a critical point for F q . However, c(q) = −1+ √ 1 − q is a marked critical point for F q , and one can check that (40) F q (c(q)) = F q −1 + 1 − q ≡ −1 ≡ a(q) , so that a(q) and c(q) have the same orbit under F m q , which, in turn, implies that their active and passive loci are the same. Therefore, McMullen's theorem implies that the limiting locus of chromatic zeros for the DHL has Hausdorff dimension equal to 2 and that it contains small copies of the Mandelbrot set (as seen in the figures). (This was first observed by Chio and Roeder [40,Theorem B].) An important remark here is that it is essential that the marked point be a critical point for F q . So, the McMullen theorem does not apply to any of the other slices with constant v 0 = −1 that we are considering. In particular, it does not imply that the limiting locus of zeros, B q (v 0 ), for these other slices have Hausdorff dimension 2. This also explains the absence of small Mandelbrot sets in Figures 5 -12 for v = −1. In fact, one can prove that the activity locus A q (−1) for the marked point a(q) ≡ −1 under F q (v) coincides with the bifurcation locus M of the mapping. This is done in Proposition 7.2 from [40] for the mapping r q (y) and marked point y = a(q) = 0, which correspond under conjugacy to the situation here. Therefore, the work of Wang, Qui, Yin, Qiao, and Gao [34, Theorem 1.1] and Yang and Zeng [39,Theorem 1.2] implies that the boundary between the white, blue, and black regions shown in Figure 3 is connected (see Lemma 5.4). Furthermore, because A q (−1) = M, parameters where the dynamics of F q (v) bifurcates will be in B q (−1). Again, this is special to the case v = −1. We comment on several general properties. For this and the other region diagrams studied here, with both v 0 ∈ [−1, 0) (antiferromagnetic range) and v 0 > 0 (ferromagnetic range), the outer part of the diagram, extending infinitely far from the origin, is characterized by the property that lim m→∞ F m q (v 0 ) = 0, as indicated by the white color. For real v 0 , this can be understood as follows. In both the antiferromagnetic and ferromagnetic Potts model, an increase in q introduces more fluctuations in the system, since the spin at each site can take on values in a larger set. Hence, for a given temperature and hence a given value of v 0 , the system will be in the disordered phase with no long-range spin-spin ordering. The RG transformation will thus move the system toward the infinite-temperature fixed point at v = 0. The inner part of the diagram is comprised of blue, black, and also white regions (the white regions being separated from the outer white region by parts of B q (v 0 ). An outer boundary separates the outer white region from this complex inner set of regions. Let us now discuss some properties of how B q (−1) intersects the real q axis. We refer the reader to Figure 3 throughout the discussion. We claim that: (i) The left-most real point where B q (−1) intersects the real q axis is q = 0, (ii) The right-most real point where B q (−1) intersects the real q axis is q = 3, and (iii) There is an infinite sequence of points q n where B q (−1) intersects the real q axis converging to q ∞ = 32 27 ∈ B q (−1). We will first give a physical description and interpretation of these properties. They will later be proved as part of Theorem 7.1. Let us denote by q c (D ∞ ) = 3 the right most place where B q (−1) intersects the real q axis. It can be explained by the fact that there is a qualitative change in the locus B v (q) at q = 3, namely the appearance of an antiferromagnetic transition at T = 0 [21]. Therefore, the intuitive approach that (q 0 , v 0 ) ∈ B q (v 0 ) if and only if (q 0 , v 0 ) ∈ B v (q 0 ) indicates that this should lead to q = 3 ∈ B q (−1). This expresses the property that the q = 3 Potts antiferromagnet has a zero-temperature critical point on D ∞ . Note that although B q (−1) crosses the real q axis at this point, the chromatic polynomial P (D m , 3) has the nonzero value (44) at q = 3. We remark that q c (D ∞ ) is equal to the value q c (S ∞ ) = 3 that was inferred in a similar manner for the Sierpinski gasket fractal S ∞ in [38] and is also the same as the value q c (sq) = 3 for the (infinite) square lattice [68]. Interestingly, it is also the same as the value of q c that was derived for infinite-length self-dual strips of the square lattice [57]. The nature of the Julia set J(q) and hence of B v (q) changes qualitatively depending on whether the discriminant (30) is positive, negative or zero, and the demarcation point between positive and negative values occurs at the special value q = 32/27. Connected with this, we infer that B q crosses the real q axis at q = 32/27 and, furthermore, that this is the minimal positive value of q where such a crossing occurs. Note that the point q = 32/27 itself is not a chromatic zero of D m for any m. More generally, it has been proved that for an arbitrary graph, the real interval (1, 32/27] is free of chromatic zeros [69] (see also [70]). Let us now start at the right-most crossing at q = 3 and move to the left through the adjacent inner blue region. This blue region extends down to a point that is the unique real root of the cubic equation (41) q 3 − 5q 2 + 11q − 9 = 0 , q 0 = 3 q 1 = 1.6388969195 q 2 = 1.4097005138 q 3 = 1.3232009243 q 4 = 1.2798668287 q 5 = 1.2546493642 q 6 = 1.2385319865 q 7 = 1.2275429153 q 8 = 1.2196860382 q 9 = 1.2138598416 Table 2. Approximate values for the 10 largest intersections between B q (−1) and the real q axis. namely, where B q (−1) crosses the real q axis. Indeed, one finds that if q = q 1 or q = 3, and The former implies that the marked point v = −1 is active for q = q 1 and for q = 3 so that Lemma 5.3 gives that they are in B q (−1). The latter implies that (q 1 , 3) ∈ P ∞ q (−1) (blue region); see Lemma 9.4(ii). Eqn. (42) was obtained by solving whose solutions correspond to all values of q for which v = −1 is either a fixed point or mapped to a fixed point. Eqn. (43) has four real solutions q = 1, q = 3 2 , q = q 1 1.6388, and q = 3. (The first two solutions correspond to passive behaviors for the marked point v = −1, with it being a superattracting fixed point when q = 1 and it being mapped by F q to the superattracting fixed point at v = 0 when q = 3 2 .) Fig. 3 also shows a succession of regions and associated crossings q n of B q (−1) with the real q axis, extending to the left of q 1 and converging on q ∞ = 32/27 from above. Between these crossings are an infinite set of regions, alternating between blue and white, also decreasing to q ∞ = 32/27 from above. However, the figure, calculated and presented to finite resolution, can only show a finite subset of these. We have numerically computed the ten largest crossing points of B q (−1) with the real q axis and we present them in Table 2. They were computed using a method similar to the computation of q 1 , described above. More specifically, for 1 ≤ k ≤ 5 we numerically solved the equation It is of interest to compare the exact results discussed above and depicted on the left hand side of in Fig. 3 with the chromatic zeros of D m calculated for finite m shown on the right hand side of Fig. 3. Extensive experience with chromatic zeros of sections of regular lattices has shown that a subset of these approach the locus B q (−1) as the number of vertices gets large (e.g., [43], [49]- [56]). Here we observe a similar behavior, although for the diamond hierarchical lattice B q (−1) is obviously much more complicated than the real algebraic curves and possible line segments comprising the loci B q (−1) for the n → ∞ limits of sections of regular lattices and related chain graphs. In particular, one can see (complex-conjugate pairs of) zeros near q = 0 and q = 3, as well as a clustering of chromatic zeros forming a wedge-shaped pattern, with the apex of the wedge facing left and located on the real q axis at q 1.2, close to q ∞ . The zeros that we have calculated for D m graphs show considerable scatter, and in this respect they differ from the chromatic zeros that were calculated in [38] for Sierpinski graphs. From a comparison of the zeros for 1 ≤ m ≤ 4, we find that the left-most complex-conjugate pair of zeros move toward the point q = 0 as m increases, in agreement with the property deduced from the analysis leading to Fig. 3, that q = 0 is a crossing of B q (−1). This is consistent with the fact that in other cases (e.g. [48,54,44,59,58] where this behavior (of complex-conjugate pairs of zeros in the vicinity of q = 0 moving toward the latter point as n increases) is observed, and one has exact results for B q (−1), it is associated with the property that for n → ∞, the locus B q (−1) passes through q = 0 and separates the complex q plane into different regions. In analyzing these chromatic zeros and their limiting behavior for m → ∞, it is useful to recall some rigorous results on zero-free regions on the real q axis. Since the signs of descending powers of q in P (G, q) alternate, an elementary property is that P (G, q) has no zeros in the interval (−∞, 0). For an arbitrary graph G, there are also no chromatic zeros in the interval (0,1) and, as mentioned above, in the interval (1, 32/27]; see [69], [70], [9]. Thus, although B q (−1) crosses the real q axis at the point q = 32/27, this point itself is not a chromatic zero. Since P (G, q) always has a factor of q, it always vanishes at q = 0, and if, as is the case here, G has at least one edge, then P (G, q) also vanishes at q = 1. For 1 ≤ m ≤ 4, we find that the only real zeros of P (D m , q) are q = 0, 1. It is interesting to note that although P (D m , 2) = 2, this polynomial can be quite small for part of the interval 1 ≤ q ≤ 2. For example, as q increases from 1 to 2, P (D 2 , q) reaches a local maximum of 0.041 at q = 1.1, then decreases to a local minimum of 0.0080 at q = 1.36, and finally increases to 2 as q → 2. In the same interval, P (D 3 , q) reaches a maximum of 0.0090 at q = 1.02, decreases to a minimum of approximately 1.7 × 10 −9 at q = 1.37, and then increases to 2 at q = 2. Let us remark on another property of P (D m , q). Because D m is bipartite, P (D m , 2) = 2. By explicit iterative calculation, we obtain (44) P (D m , 3) = 2 · 3 n(Dm)/2 . Consequently, so that the 3-state Potts AFM has ground-state entropy on the D ∞ fractal given by Interestingly, these results are the same as for the infinite-length square-lattice ladder graph (with any longitudinal BC) [43]), for which (47) W (sq, 2 × ∞, q) = q 2 − 3q + 3 , and hence W (sq, 2 × ∞, 3) = √ 3. The cyclic and Möbius strips of the square lattice with width L y = 2 vertices are ∆-regular graphs with vertex degree ∆ = 3, and the free L y = 2 square-lattice Im(q) Im(q) strip also has ∆ ef f = 3 in the m → ∞ limit. These values of ∆ and ∆ ef f are the same as the value (18). Zeros in the q Plane for the Potts Antiferromagnet at Nonzero Temperature We next consider the zeros of Z(D m , q, v 0 ) for the Potts antiferromagnet with temperature T 0 > 0 which corresponds to the range −1 < v 0 ≤ 0. On the left sides of Figures 5 -7 we present computergenerated images of the regions for v 0 = −0.8, −0.5, and −0.2. Any point that is not in P 0 q (v 0 ) or P ∞ q (v 0 ) is colored black. According to Proposition 5.7, B q (v 0 ) contains any point of the boundary between the white, blue, and black sets. It may contain additional points, but they are necessarily in the interior of the set of black points. For comparison, on the right sides of Figures 5 -7 we present the numerically computed zeros of the reduced partition function Z r (D 4 , q, v 0 ) at these values of v 0 . (As usual, the zero at q = 0 is omitted.) First, we observe that in the case of infinite temperature, or equivalently, zero spin-spin coupling, v 0 = 0 we have that Z(G, q, 0) = q n(G) for any graph G, so that all of the zeros of Z(G, q, 0) occur at q = 0. Previous studies (e.g., [44,58]) showed that generically, as v 0 approaches 0, the zeros of Z(G, q, v 0 ) in the q plane progressively move in toward the origin, q = 0. We see that behavior here for the diamond hierarchical graphs D m , as evident from the decreasing scale of As v increases in the interval v 0 ∈ (−1, 0], there is an evident simplification in the different regions on the left side of Figures 5-7, as compared with the v 0 = −1 case. As was found for regular lattice graphs (e.g., [44]), as v 0 increases in the interval (−1, 0], the maximal point, q c (v 0 ), at which Although it is not completely evident from Figures 5 -7, for any −1 ≤ v 0 < 0 the locus B q (v 0 ) continues to intersect the real q axis in infinitely many points, just like in the case v 0 = −1. One can observe this by using the computer to zoom in when investigating the region diagrams. However, we will prove it rigorously in the theorem below. The following theorem summarizes the antiferromagnetic case: Theorem 7.1. For any −1 ≤ v 0 < 0 the locus B q (v 0 ) intersects the axis at 0 and at The locus B q (v 0 ) does not intersect the real q axis at any point outside of the interval [0, q c (v 0 )]. Furthermore, there is an infinite sequence of points q k (v 0 ) where B q (v 0 ) intersects the real q axis converging to which is also in B q (v 0 ). We refer the reader to Figure 8 Zeros in the q Plane for the Ferromagnetic Potts Model We now present some results on the zeros of Z(D m , q, v) and the locus B q (v 0 ) in the complex q plane for the Potts ferromagnet, corresponding to v 0 ≥ 0. On the left sides of Figures 9 -12 we present computer-generated images of the regions Im(q) Figure 9. Left: Region diagram for D ∞ in the complex q plane for v 0 = 1. Right: Zeros of the reduced partition function Z r (D 4 , q, 1) (171 zeros). Both left and right figures depict −3 < Re(q) < 1 and −3 < Im(q) < 3. for v 0 = 1, 2, 4, and 99. As in previous figures, any point that is not in P 0 q (v 0 ) or P ∞ q (v 0 ) is colored black. According to Proposition 5.7, B q (v 0 ) contains any point of the boundary between the white, blue, and black sets. It may contain additional points, but they are necessarily in the interior of the set of black points. For comparison, on the right sides of Figures 9 -12 we present the numerically computed zeros of the partition function Z(D 4 , q, v 0 ) at these values of v 0 . (As usual, the zero at q = 0 is omitted.) In general, for all four of these ferromagnetic values of v, especially for the largest two values, the region diagram appears considerably simpler than for the antiferromagnetic values. As in the antiferromagnetic case, the region diagrams have an outer white area extending infinitely far from the origin, separated from an inner blue and black portion by part of B q (v 0 ). For v 0 = 1 and v 0 = 2 there are still some black regions, but they have become rather small, and for v 0 = 4 and v 0 = 99, at the resolution of the figures, one sees only an inner blue region and an outer white region, separated by B q (v 0 ). Furthermore, they appears to become smoother, approaching a nearly circular form for large v 0 , as will be discussed further below. The computer images also indicate that the Hausdorff dimension of B q (v 0 ) decreases as v 0 increases sufficiently. Another difference from the antiferromagnetic values of −1 ≤ v 0 < 0 is that for the ferromagnetic values v 0 > 0 the locus B q (v 0 ) intersects the real q axis in only two points: Theorem 8.1. For any v 0 > 0 the locus B q (v 0 ) intersects the real q-axis only at the two points q − (v 0 ) < q + (v 0 ) given by Im(q) Figure 10. Left: Region diagram for D ∞ in the complex q plane for v 0 = 2. Right: Zeros of the reduced partition function Z r (D 4 , q, 2) (171 zeros). Both left and right figures depict −6 < Re(q) < 3 and −6 < Im(q) < 6. Im(q) Re(q) Re(q) Im(q) We can give a statistical physics explanation of the crossing of B q (v 0 ) on the positive real q axis; the Potts ferromagnet has a phase transition at this value of q = q c for the given value of the temperature variable v 0 , namely T c,P M −F M as given in Eq. (31). For larger q, the system is more disordered; for integral q, this can be understood from the fact that each spin on the lattice can take values in a larger set, {1, ..., q}. In this case, the infinite iteration of the RG transformation maps the temperature to T = ∞, or equivalently, decreases v to 0, hence the white color. For 0 < q < q c the system is more ordered, so the infinite iteration of the RG transformation maps the temperature variable to T = 0, i.e., v = ∞, hence the blue color. In contrast with the scattered pattern of zeros in the q plane for the T = 0 AFM case v 0 = −1, the zeros in the q plane for the FM case illustrated by these four values of v 0 tend to cluster along a curve. This curve encircles the origin. As v 0 gets large, this curve assumes an oval-like shape. In [82] (see also the related [81]), we showed that, in the n → ∞ limit of a recursive family of graphs, as \|v 0 \| increases to values 1, the accumulation set of zeros in the q plane, B q (v 0 ), forms a closed oval curve that encircles the origin and crosses the positive q axis at (49) q \|v\| ∆ ef f /2 , where ∆ ef f was defined in Eq. (17). In this m → ∞ limit, B q (v 0 ) forms a closed oval curve approaching a circle, crossing the positive and negative real q axes at a value of \|q\| that behaves asymptotically as (50) \|q\| ∼ \|v 0 \| 3/2 . Our calculations of zeros of Z(D m , q, v 0 ) for finite m are in agreement with this result, as is illustrated in Fig. 12. Note that for m = 4 we have (51) ∆ ef f = 3 1 + 2 −7 = 2.98 for D 4 , which is quite close to the limiting value lim m→∞ ∆ ef f = 3 in Eq. (18). 9. Proofs of Theorems 7.1 and 8.1. We will consider the dynamics of the RG mapping F q (v) for v ∈ R and q ∈ R \ {0}. 9.1. Lemmas and setup. Let us briefly summarize some properties of F q (v) that will be used throughout this section. For any q ∈ R \ {0} the mapping F q (v) has a superattracting fixed point at v = 0 and another fixed point v c,PM−FM (q) > 0. Moreover, v c,PM−FM (q) is the unique positive fixed point of F q (v) for any q ∈ R \ {0}, a property that will play an important role in several of the proofs. The mapping F q (v) has a pole at The critical points of F q (v) are: Note that for any q. Proof. One can check that the expression for F q (v) + 1 is a perfect square. Lemma 9.2. For any real q = 0 we have that v c,PM−FM (q) is a repelling fixed point for F q (v) satisfying F q (v c,PM−FM (q)) > 1. Proof. We remark that for q = 0 the fixed point v c,PM−FM (q) is a solution to the equation F q (v) = v that occurs with multiplicity one so that we cannot have F q (v c,PM−FM (q)) = 1. Case 1: q > 0. The pole and all of the critical points of F q (v) lie in (−∞, 0]. A simple calculation shows that F q (v) > 0 for v > 0. Since F q (0) = 0 and v c,PM−FM (q) is the unique fixed point of F q (v) occurring at positive v we conclude that F q (v) < v for 0 < v < v c,PM−FM (q) and that F q (v) > v for v > v c,PM−FM (q). Since F q (v c,PM−FM (q)) = 1 we must have F q (v c,PM−FM (q)) > 1, as desired. Case 2: q < 0. The pole at −q/2 and two of the critical points −q and −1 + √ 1 − q occur at positive v and they do so with the following order: Intermediate Value Theorem implies that F q has a fixed point between −1 + √ 1 − q and −q/2. Since v c,PM−FM (q) is the unique positive fixed point of F q we conclude that with v chosen sufficiently close to v c,PM−FM (q), one has F q (v) < v. Since lim v→−q/2 F q (v) = +∞ the Intermediate Value Theorem would imply that there is an additional fixed point v • of F q with v c,PM−FM (q) < v • < −q/2. This contradicts that v c,PM−FM (q) is the unique positive fixed point of F q (v). Since F q (v c,PM−FM (q)) = 1 we must therefore have F q (v c,PM−FM (q)) > 1. Lemma 9.3. We have: Proof. Claim (i): When q < 0 the pole and the critical points −q and −1 + √ 1 − q occur at positive v. The critical point −1 − √ 1 − q occurs for v < −1. The only real fixed points for F q (v) are 0 and v c,PM−FM (q) > 0. A calculation shows that F q (−1) > −1 and hence that F q (v) > v for all −1 ≤ v < 0. Another calculation shows that F q (v) < 0 for all −1 ≤ v < 0. Therefore, the sequence F m q (v 0 ) is increasing sequence that is bounded above by 0. It must converge to some limit, which will be a fixed point for F q (v). Therefore, F m q (v 0 ) → 0. Claim (ii): Suppose 0 < q < 1 and that q is sufficiently small that −1 ≤ v 0 < v − (q). In this case, we have The critical point −1 − √ 1 − q occurs at v < −1 and the other two critical points and the pole of F q are ordered as q-PLANE ZEROS OF THE POTTS PARTITION FUNCTION ON DIAMOND HIERARCHICAL GRAPHS 31 We have that This implies that for any Lemma 9.4. For any real q = 0 we have We remark that in Claim (i) when q < 0 the orbit of v 0 may pass into the interval [−1, 0). Proof. We split the proof of Claim (i) into two cases: Case 1: q > 0. As in the first paragraph of the proof of Lemma 9.2, we have If 0 ≤ v 0 < v c,PM−FM (q) then the orbit F m q (v 0 ) forms a decreasing sequence, which is bounded below by 0, and hence converges. The limit must be a fixed point of F q (v) and, since v c,PM−FM (q) is the only positive fixed point of F q (v), we see that F m q (v 0 ) → 0. Case 2: q < 0. As in the second paragraph of the proof of Lemma 9.2, the critical points, pole, and fixed points of F q (v) occur with the following order: For v 0 ∈ I 1 it follows from Lemma 9.3(i) that F m q (v 0 ) → 0. The zeros of F q (v) are 0, −2 ± −2 q + 4. The orbit F m q (v 0 ) cannot remain in I 3 because, if it did, the orbit would form a decreasing sequence that is bounded below by −2 + √ −2 q + 4. It would therefore converge to some fixed point v , which is impossible because v c,PM−FM (q) is the unique positive fixed point of F q (v). Therefore, for any v 0 ∈ I 3 there is some iterate m 0 for which F m 0 q (v 0 ) ∈ I 2 , at which point the reasoning in the previous paragraph implies that F m q (v 0 ) → 0. We will now prove Claim (ii). Note that pole v = −q/2 cannot occur on [0, v c,PM−FM (q)]; see Eqn. (52) for the case that q < 0. Therefore, since F q (0) = 0 and F q (v c,PM−FM (q)) > 1, by Lemma 9.2, we have that F q (v) > v for all v > v c,PM−FM (q). (When q < 0 we allow for the possibility that v = −q/2 is the pole of F q (v).) This implies that if v 0 > v c,PM−FM (q) the sequence of iterates F m q (v 0 ) is increasing. It must converge to infinity as there is no fixed point of F q (v) that is larger than v c,PM−FM (q). We are now ready to prove Theorems 7.1 and 8.1. We will begin with Theorem 8.1 as the proof is somewhat easier. 9.2. Proof of Theorem 8.1. Recall from Section 4.3 that the fixed points of F q (v) other than 0 and ∞ are roots of the equation Solving for the values of q for which F q (v) has a fixed point at v 0 yields the two solutions Since we consider v 0 > 0, the resulting fixed point will necessarily be v c,PM−FM (q), as it is the only positive fixed point of F q (v). If q < q − (v 0 ) or q > q + (v 0 ) then 0 < v 0 < v c,PM−FM (q) and F m q (v 0 ) → 0 by Lemma 9.4. Therefore, q (v 0 ) and Proposition 5.7 implies that B q (v 0 ) does not intersect such points. (Such points are colored white in When q = q ± (v 0 ) the marked point a(q) = v 0 hits the fixed point v c,PM−FM (q), which is repelling by Lemma 9.2. Since there are values of q for which a(q) = v c,PM−FM (q) this corresponds to the marked point a(v 0 ) being mapped by F 0 q (v) non-persistently to the repelling fixed point v c,PM−FM (q). Therefore, the parameters q ± (v 0 ) are active parameters, and hence and F m q (v 0 ) → ∞ by Lemma 9.4. Therefore, (q − (v 0 ), q + (v 0 )) ∈ P ∞ q (v 0 ) and Proposition 5.7 implies that B q (v 0 ) does not intersect such points. (Such points are colored blue in Figures 9 -12.) 9.3. Proof of Theorem 7.1. Let −1 ≤ v 0 < 0. Lemma 9.3(i) gives for any q < 0 that F m q (v 0 ) → 0 and hence (−∞, 0) ∈ P 0 q (v 0 ). Proposition 5.7 implies that B q (v 0 ) does not intersect such points. We will now show that q = 0 is in B q (v 0 ). This requires some care because the degree of F q (v) drops from 4 to 2 at q = 0. (It is why we omitted q = 0 from the parameter space in the discussion from Section 5.) For this reason, Lemma 5.3 does not immediately apply to q = 0. Instead, we will show for any > 0 that there is a parameter q = 0 with \|q\| < that is active for a(q) ≡ v 0 under F m q . Such a point is in B q (v 0 ) by Lemma 5.3, and, since > 0 is arbitrary, this will prove that 0 ∈ B q (v 0 ). As seen in the previous paragraph, any real point q < 0 is in P 0 q (v 0 ). Note that as q 0 the fixed point v − (q) increases to 0; see Figure 2. Therefore, we can choose q > 0 sufficiently small so that −1 ≤ v 0 < v − (q). It then follows from Lemma 9.3(ii) that F m q (v 0 ) → v − (q). We conclude that on any arbitrarily small annulus 0 < \|q\| < there are two different passive behaviors for the marked point a(q) ≡ v 0 : (i) convergence to 0 and (ii) convergence to v − (q) = 0. Therefore, there must be active parameters in this annulus. We will now prove that q c (v 0 ) = (−2 − √ −v 0 ) v 0 is the largest point where B q (v 0 ) hits the real q-axis. One finds that which is negative for q > −2v 0 . We have Since v c,PM−FM (q) is positive and increasing for q > 0 the Intermediate Value Theorem gives that for any −1 ≤ v 0 < 0 there is a unique q c (v 0 ) > −2v 0 for which In particular, when q = q c (v 0 ) the marked point a(q) ≡ v 0 lands non-persistently on the repelling fixed point v c,PM−FM (q) when mapped by F q (v). This implies that q c (v 0 ) is an active parameter for a(q) ≡ v 0 and hence that q c (v 0 ) ∈ B q (v 0 ), by Lemma 5.3. . Lemma 9.4(i) then implies that such q are in P 0 q (v 0 ) and Proposition 5.6 then implies that they are not in B q (v 0 ). for q and selecting the branch of the solutions that is larger than the pole q = −2v 0 . It remains to show that there is a sequence of real parameters q k (v 0 ) ∈ B q (v 0 ) that converge to q ∞ (v 0 ); See Eq. (48). We claim that for any k ≥ 2 the function q → F k q (v 0 ) has at least one pole in the interval 32 27 , 3 . To see this, note that if F k−1 q (v 0 ) has a pole q • ∈ 32 27 , 3 , then q • is also a pole of F k q (v 0 ) since F q (∞) = ∞ for any q. Otherwise, if F k−1 q (v 0 ) has no pole in the interval 32 27 , 3 then so that the Intermediate Value Theorem implies that there is some 32 27 , 3 . We have with the fact that the limit is +∞ (not −∞) a consequence of Lemma 9.1. For any q ∈ 32 27 , 3 the repelling fixed point v c,PM−FM (q) remains in [0, 3]. Therefore, the Intermediate Value Theorem implies that there exists q k (v 0 ) ∈ 32 27 ). As this does not hold identically for all q ∈ 32 27 , q ∞ k (v 0 ) , we conclude that the marked point a(q) ≡ v 0 is mapped by F k q non-persistently at q k (v 0 ). It follows from Lemma 5.3 that q k (v 0 ) ∈ B q (v 0 ). For any q 0 > 32 27 there is a definite constant C(q 0 ) such that F q (v) > v + C(q 0 ) for all q > q 0 and all −1 ≤ v < − q 2 . Meanwhile, F q (v) > 0 for any v > − q 2 . Therefore, there is a definite K(q 0 ) such that for any is invariant under F q for q > 0 and the pole − q 2 is negative, this implies that for any k > K(q) we have 32 27 Since q 0 > 32 27 was arbitrary, we conclude that q k (v 0 ) → 32 27 as k → ∞. Case 2: − 8 9 < v 0 < 0: A straightforward modification of the proof from Case 1 applies, after replacing q = 32 27 with the unique value of q = −1 − Zeros in the v Plane In addition to the zeros of Z(D m , q, v) in the q plane for fixed v = v 0 and their accumulation locus B q (v 0 ) in the limit m → ∞, it is also of interest to investigate the zeros of Z(D m , q, v) in the v plane for fixed q = q 0 and their accumulation locus B v (q 0 ) in this limit m → ∞. We present some new results on these in this section. Hence, Z(D m+1 , q, v 0 ) = 0 if v 0 = v 0 . This connects the set of points v 0 for which Z(D m , q, v 0 ) = 0 to the set of values of v that are left invariant by the transformation F q (v). Since a deviation of a given v from this invariant set will generically be multiplied by successive iterations of F q (v), this yields the identification of B v with the Julia set of F q (v). Previous studies have presented zeros in the y plane for low values of q. Here we extend this study of zeros of Z(D m , q, v) in v with new results for large positive and negative values of q. Although negative values of q do not have direct significance for the physical q-state Potts model, they are of interest in investigating the mathematical properties of the full Z(D m , q, v) polynomial. In [82] we showed that B v crosses the positive v axis at (54) v \|q\| 2/∆ ef f . Since, as noted above, ∆ ef f is close to 3 for D 4 , it follows that the zeros to cross the positive v axis at v q 2/3 for \|q\| 1. In agreement with this, we show our calculations of zeros of Z(D 4 , q, v) in the y plane for the large positive values q = 10 2 , 10 3 , and the large negative value q = −10 2 in Figs. 14-16. (The number of zeros displayed in each of these figures, namely 256, is sufficiently great that the plotting program yields what appears to be a curve, although it is really a discrete set of zeros.) There is a striking change in the structure of the zeros as q grows to values that are 1 [16,21,39]. In particular, the Hausdorff dimensionality of the Julia set approaches 1 from above asymptotically as q → ∞ [21,35]. Related to this, the global pattern of zeros becomes a single Jordan curve rather than the more complicated patterns observed for small q. The fact that this Jordan curve is not too different from a circle is similar to the pattern of complex-temperature zeros for the Potts model on regular lattices in the limit of large q [81,82]. Conclusions In this paper we have derived several exact results on the continuous accumulation sets of zeros B q (v 0 ) of the Potts model on the Diamond Hierarchical Lattice, D ∞ , in the complex q-plane at various fixed values of a temperature-like Boltzmann variable v = v 0 . We have applied methods from complex dynamics to determine the region diagram in which the infinite iteration of the renormalization group transformation exhibits different behavior in the various regions and related them to the locus B q (v 0 ). We have also used these techniques to prove rigorous results (Theorems 7.1 and 8.1) about the intersection of B q (v 0 ) with the real q-axis. For the chromatic zeros (i.e., partition function zeros of the zero-temperature Potts antiferromagnet), we have shown that the locus B q (v 0 ) crosses the real q axis at q = 0, q = 3, q = 32/27, and an infinite set of points between 32/27 and the value q = q 1 1.639 given analytically in Eq. (42). A similar behavior occurs for any finitetemperature Potts antiferromagnet on the DHL. For the finite-temperature Potts ferromagnet on the DHL, the locus B q (v 0 ) crosses the real q axis at only two points. We have also studied region diagrams and the structure of B q (v 0 ) for the finite-temperature Potts antiferromagnet and for the Potts ferromagnet on the DHL and compared them with the patterns of zeros calculated for D m with m = 4. Another result of our present work is that as \|v\| → ∞, B q approaches a circular form with \|q\| ∼ \|v\| 3/2 . Finally, we determine properties of the locus B v for q → ±∞. Appendix A. On the Case q = 0 Besides the very delicate potential issues about relating B q (v 0 ) with B v (q 0 ) that we discussed in Section 5.4, there is an additional problem that occurs when q = 0, which we explore it in this appendix. It is related to the more general fact that in defining the free energy from the partition function, it is necessary to take into account a noncommutativity in the limits n → ∞ and q → q s at special values q = q s (which may include q s = 0, 1..., χ(G), where χ(G m ) denotes the chromatic number of G m ). This was discussed in [43] for the chromatic polynomial and in [44] for the full free energy. The noncommutativity is (55) lim As in the text, we denote G ∞ as the formal limit of an iterative family of n-vertex graphs G m as m → ∞ and hence n → ∞. Because of this noncommutativity, the definitions of both the free energy in (9) and B v (q) require that the order of limits be specified. A simple example will illustrate this. Consider the Potts model on the n-vertex circuit graph, C n . An elementary calculation yields, for the partition function, the result (56) Z(C n , q, v) = (q + v) n + (q − 1)v n . As a special case of Eq. (7), Z(C n , q, v) has q as a factor, so Z(C n , 0, v) = 0. As an explicit example of the noncommutativity (55) with q s = 0, The loci B v (q) and B q (v) are given by the solution of the condition of equality of dominant terms in Z, namely (60) \|q + v\| = \|v\| . In the q plane, the solution locus, B q (v), is a circle with radius \|v\|. If v is real, this circle is centered at q = −v and crosses the real q axis at the two points q = −2v and q = 0. Thus, the maximum (finite) point at which this locus B q (v) crosses the real q axis is For example, in the case v = −1 where Z(C n , q, −1) = P (C n , q) is the chromatic polynomial and B q (−1) is the continuous accumulation set of the chromatic zeros, the locus B q (−1) is the unit circle \|q − 1\| = 1, which crosses the real q axis at q = 2 and q = 0. To show the connection with B v (q) or equivalently, B y (q), with v = y − 1, we first note that the solution of Eq. (60) in the complex v plane is the infinite vertical line crossing the real v axis at v = −q/2, i.e., (62) B v (q) : v = − q 2 + iλ, λ ∈ R. It is convenient to change variables to a variable in terms of which the locus is compact. We choose η = y −1 = e −K ; in the complex plane of the variable η, if q = 2, then the locus B η (q) is a circle: (64) B η (q) : η = 1 q − 2 (−1 + e iω ), 0 ≤ ω < 2π. As is evident from Eq. (64), this circle crosses the real η axis at η = 0 and η = −2/(q − 2). If q = 2, then B y (2) is the line y = iλ with λ ∈ R, i.e., the imaginary y axis. This locus is invariant under y → 1/y, so the locus B η (q) is also the imaginary axis in the η plane. Here we observe that the point q = 0 is in the locus B q (v) for all v, in particular, for v = −1, i.e., y = 0, but y = 0 is not in the locus B y (q) for q = 0. Explicitly, if q = 0, then B y (q = 0) is the vertical line in the y plane crossing the real y axis at y = 1; equivalently, B η (q = 0) is the circle \|η − (1/2)\| = 1/2 crossing the real η axis at η = 0 and η = 1.	2019-11-11T00:04:14.000Z	2019-11-11T00:00:00.000	{ "year": 2019, "sha1": "7d04ad9dac211404b02f61aaeb982b957b27fd39", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1911.04012", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "7d04ad9dac211404b02f61aaeb982b957b27fd39", "s2fieldsofstudy": [ "Mathematics", "Physics" ], "extfieldsofstudy": [ "Physics", "Mathematics" ] }
32135165	pes2o/s2orc	v3-fos-license	Forming Plants in Words and Images Forming Plants in Words and In The Rhetoric of Science rhetorical concepts can the persuasive work of scientific arguments Communicating Science the history of the scientific article as a genre, showing how it evolved in length, style through the nineteenth centuries Insight, with Harmon, communication and argues for the salience of visual modes of persuasion in scientifi illustrate Gross’s mastery of different scholarly methodologies, from the theoretically and visuals, to the comparison of tactics across several works, to the compilation of large databases statistically sampled. Altogether Alan Gross’s body of work, including seminal articles and significant anthologies, has established the field of the rhetoric of science and given it methods and a trajectory. No one after him has had to take this ground. update ancient herbals, assimilate newly discovered plants, and begin to systematize the accumulated body of botanical knowledge. In Germany, the verbal practices underwriting these achievements were rationalized in the discourse arts and combined with new methods of visual representation and reproduction to create the new scholarly herbals appearing from the 1530s on. When accounting for the complex influences that brought about the new "sciences" of the sixteenth century (astronomy, anatomy and botany), historians of science certainly acknowledge the influence of the humanist-reformed discourse arts (see Serjeantson, 2006, for an overview). The evolving discourse of botany alone has received significant attention from Kristian Jensen (2001), Ian Maclean (2005), Brian Ogilvie (2006), and Sachiko Kusukawa (1997Kusukawa ( , 2012, as discussed below. But while the importance of rhetoric and dialectic is recognized (see Ogilvie,[118][119], it is not often used in the analysis of early modern texts. Historians tend to ignore the linguistic dimension of this influence, preferring philosophical accounts of argument procedures over discoursebased accounts. And though some scholarship on the influence of the discourse arts on the new sciences exists, no attention at all has been paid to the possible reverse influence, from the new sciences to the discourse arts. The case study offered here suggests how intimate and reciprocal the association was between the language arts and the developing science of botany in the sixteenth century. It can even be argued that the methods of medical botany influenced the methods of definition and description recommended in treatises on the discourse arts, at least in Protestant Germany. And once these expanded methods were taught to succeeding generations, they became the default practices of textual description and, arguably, of the observation that led to that description. Making a case for this influence requires, first, a reminder of the "formal" qualities of the early modern discourse arts and second, an account of the special processes of descriptive defining that were licensed in contemporary dialectical treatises. Next, plant descriptions from contemporary herbals are reviewed to see how they match the dialectical standards for descriptive definition for both known and newly discovered species. The following section then reviews defining through images, based on well-known exemplars produced in the early sixteenth century. The conclusion suggests that the practices in these texts of forming plants in words and images yielded long-term scientific results, enabling systematic botany in the eighteenth century and, arguably, even evolutionary botany in the nineteenth. Poroi 10,2 (December 2014) Training in the Discourse Arts Every natural philosopher writing in the sixteenth century -from Copernicus and Vesalius in the 1540s to Brahe and Kepler in the 1580s and 90s -had a very detailed and very formal training, in Latin, in the allied discourse arts of grammar, rhetoric and dialectic. These arts were transformed between 1480 and 1520 by Agricola, Erasmus, and their humanist followers across Europe. Any university-trained scholar's understanding of language and argumentation was not merely influenced by, it was determined by this humanist construction of Latin into a common instrument of thought and expression. Nor was knowledge of language and of argumentative procedures disjunct. In fact throughout the rhetorical tradition, but peaking in the early modern period, scholars understood forms as content-laden in themselves. To appreciate the influence of an early modern education in the discourse arts requires understanding the role of linguistic forms in such training since these forms constitute what the linguist Edward Sapir once called the "grooves" of thought (Sapir, 1921, 14-15;217). Students learning to compose in Latin followed prepared forms at all discourse levels, from the word to the sentence to the passage. These forms could be purely linguistic, or they could specify a function to be fulfilled, or they could even require the production of a certain kind of content. The content-less linguistic forms as extractable, repeatable patterns are ubiquitous in grammar, rhetoric and dialectic, from metaplasms of word formation, to syntactic schemes for sentences, to templates for syllogisms in their various figures. These linguistic forms are "empty"; they require a language about language to come into view. Advice about form could also specify functional slots to be filled, as in the epicheireme calling for certain kinds of support or amplification (Cicero,(101)(102)(103)(104)(105), or the chreia, one of the early progymnasmatic exercises, where templates call for inserting comparisons, examples, and so on in a certain order (Kennedy, 2003, 15;76;97;139;193). Such functional forms require a language about argumentation to come into view differentiating what counts as a claim, what as support and what kind of support. A third kind of form, the least appreciated, specifies the nature of the content required to fill it. The forms in question here include the patterns presented in rhetorical and dialectical treatises for certain topoi. The a fortiori topics A minore and A maiore, for example, where differences of more or less of a phenomenon have to be found, present their users with templates demanding content to follow the phrases multo minus or multo magis (Melanchthon 1846(Melanchthon [1547, 695). Similarly, the topic of proportion required formulaic phrasing to be filled in with correct numerical analogies (Melanchthon, 1846(Melanchthon, [1547, 696; una hydria ad duas metretas...sex hydrae ad duodecim metretas). Among these content-defined forms are the "formulae," as Melanchthon called them, for descriptive definitions, which are explained below. These contentspecifying forms require a subject matter language to come into view, and this subject matter language can then become the framework for a body of learning. There may be, in other words, a non-trivial relationship between such forms and the generation of knowledge. The new science of botany in the sixteenth century offers a particular case of synergy between a form specified in the discourse arts, its embodiment in a visual representation, and the management and even generation of knowledge. Forming Botanical Descriptions The initial stimulus for the renewal of botany, as for so many other fields of learning in the early modern period, was the recovery and correction of classical works, in this case by Theophrastus, Dioscorides, Pliny, Galen and others, all appearing in new editions and translations by the first decades of the sixteenth century. But the printing between 1460 and 1530 of the Greek texts and Latin translations of medical herbals like Dioscorides' De Materia Medica, none with illustrations, created nomenclature problems, especially for those interested in the medical uses of plants 2 (Hoeniger, 1985, 146;Reeds, 1976, 526). How did the names and descriptions in these classical works correspond to living plants with local vernacular names? And how did the plants described by these Mediterranean authors compare to those growing in northern climates? Indeed Brian Ogilvie has described the work of the first generation of sixteenth-century botanists as largely a matter of collation between ancient texts and living plants (Ogilvie, 2006, 34;127;134). There were many uncertainties, and the new botanical treatises published across the century often disagreed with each other over these identifications. Arguments over whether a plant described in a classical text matched a living plant involved definitions, and the rules and tactics for defining belonged to the disciplines of rhetoric and 2 The famous illustrated codices of Dioscorides, now in Vienna and Naples, are not the source of these reprints. There were many other extant ms copies of Dioscorides, some illustrated, though without the riveting naturalism of the images in these most famous codices. Poroi 10,2 (December 2014) dialectic. All university-trained physicians, including the early cadre of pharmacological botanists, had been trained in these arts. Through the frame of Porphyry's gloss on Aristotle's Categories (relevant to the Topics as well), they knew the five "predicables" (genus, species, differentia, property, accident), and they knew that the definition of a species required the predication of a genus followed by the distinguishing differentia. 3 Furthermore, in an Aristotelian understanding, a true definition had to express the essence or substance of the thing defined. Among the other predicables, a "property" had to be a feature unique to its species, and though it could be used to construct the differentia of a definition, it did not necessarily capture the essence of a subject. Nevertheless, the test for both a cogent genus/difference definition or a unique property was the same stylistic manipulation of reciprocal predication: Could the claim sustain a conversion in the form of the figure antimetabole? A genuine definition and a genuine property could, as in Porphyry's example of the latter, "If a horse, then hinnability, and if hinnability, then a horse" (Spade, 1994, 10). These potential conversions, or reciprocal predications, are still important in all sciences where unknowns have to be identified. So for example the definition constructed from a genus term and a unique property -Gallium is an element with an atomic weight of 69.72 a.m.u.converts to An element with an atomic weight of 69.72 a.m.u. is gallium, and offers an identifying test. However predicating just a genus (Copper sulfate is a compound) or an accident (Copper sulfate is a blue crystal) does not meet the test of conversion in A compound is copper sulfate or A blue crystal is copper sulfate. There are many other compounds and many of these form blue crystals. Here the use of the formal test of reciprocal predication worked in synergy with the systematic development of subject area knowledge. It takes external knowledge, after all, to know that there are other blue compounds in the world. Early modern botanists wanted to know the unique properties of plant species in order to be able to identify plants with certainty, but such knowledge was for the most part inaccessible to them. More accessible were the "accidents," the qualities taken in by the senses, making up the fifth predicable. Porphyry dissociated these into the separable versus inseparable: "Accident is what comes and goes without the destruction of the substrate. It is divided into two kinds. One kind of accident is separable and the other inseparable. Thus sleeping is a separable accident, whereas being black is an inseparable accident of the crow and the Ethiopian" (Spade, 1994, 11). The difference between the two is simple: an inseparable accident is always present in its subject and a separable accident is not. Copper sulfate is blue predicates an inseparable accident, but Copper sulfate is on the third shelf in the storeroom states an easily changed separable accident, and such separable accidents, useful as they may be for identifying things in particular contexts, cannot form definitions, a rule with the authority of Aristotle who clarifies in the sixth book of the Topics that "the differentia of a thing cannot both belong and not belong to it" (Aristotle, 1984, I, 243). It is easy to dismiss separable accidents like "on the third shelf" from the work of definition. But inseparable accidents always belong to a species. How then are inseparable accidents any different from properties which are also always present in a species? Why are they also ruled out for conversion-sustaining definition? The answer is that an inseparable accident, like having a blue color, is not unique to its subject the way a true property is (like an atomic weight or a spectrographic signature to use contemporary examples). In the case of defining plants, this problem of the non-uniqueness of inseparable accidents is acute because any particular leaf shape and size, any flower color, and any root or stem type is likely to be shared with another plant. So no inseparable accident, on its own, can serve as the differentia in the definition of a plant, or of anything else. But a solution to the problem of defining and identifying through inseparable accidents is possible. Though no single inseparable accident may be uniquely and hence convertibly predicated of its subject, a collection of them might be so that their assemblage could add up to the differentia creating a convertible definition. Furthermore, what is defined in dialectic as an inseparable accident is typically a quality salient to the senses such as color, shape, texture and taste. So anyone listing multiple inseparable accidents of an object will in effect describe it in the act of defining it -or define it in the act of describing it. And, according to Porphyry, extending Aristotle, accidents can include differences in degree, namely comparisons about more or less of a feature (Spade, 1994, 18). This resource will become a mainstay in descriptions of allied species when observers note that one plant is taller or has hairier leaves than another etc. The solution that salvages inseparable accidents for arguments from definition has been attributed to Boethius in De Topicis Differentiis in a passage discussing the variety of topics concerned with definitions: Definition differs from description because a definition contains genus and differentiae; a description comprises understanding of the subject, either by means of certain accidents producing one property or by means of the differentiae of the substance (substantialibus differentiis) brought together apart from the appropriate genus (Stump, 1978, 49-50). When Boethius equates a collection of accidents with a property -"certain accidents producing one property" -he is saying that such a collection meets the standard of reverse predication. When he talks about such a collection producing a differentia apart from a genus, he means that a description can simply list features without placing an item in a genus. 4 Overall, Boethius licenses two standards in definition, one with Aristotelian genus/difference rigor and the other laxer and more functional, compiled from several inseparable accidents and amounting to a description. 5 A clearer characterization of pragmatic defining by a collection of enduring accidents, and one that takes this distinction into the early modern period, is found in Rudolph Agricola's influential late fifteenth-century De Inventione Dialectica under the topos of definition. (Peter Mack has drawn attention to Agricola's comments in this key passage [Mack, 1993, 151-56]). After Agricola acknowledges the standard method of defining with genus and differentia, he points to the lack of true differentiae, making it 4 Further passages in Boethius legitimize definition from accidents as description: "If the argument is taken from the things themselves, it must be taken from their substance, from the things that follow from the substance, from the things that are inseparable accidents-those that adhere and cannot be or generally are not separated or disjoined from their substance. Those which are drawn from their substance consist in description, definition, or in addition, explanation of the name" (Stump 1978, 60; see also 73, 74 [in the diagram of Themistius' topics]). 5 This distinction between definition and description is probably not, however, attributable solely to Boethius. A distinction between definitions based on substance and other discursive forms was worked out by the rhetorician Victorinus several decades earlier in his De Definitionibus. This work lists fifteen methods of definition, only the first concerning the substance of the thing defined (Stangl, 1888, 33). necessary to construct definitions that at least approach the truth. Such approximations, he says, require circumlocution in speaking [loquendi circuiti], collecting many details which the thing is known to possess and which, when joined, create a certain property. Agricola offers as a clarifying example the definition of an ass as an animal "solid-footed, large-eared and fecund." He points out that none of these features belongs to the ass alone: The mule and hare are long-eared and all animals are fecund. But "solid feet" excludes all animals except the horse and mule; "long-eared" then excludes the horse and "fecund" the mule. "At last," he says, "as with steps that which is defined is reached" (Agricola, 1563(Agricola, [1485, 41). Agricola does not use forms of the term accidens in this passage, but any feature not unique to a species would be understood as belonging to that category. Agricola is also aware that these definitions are in effect descriptions, but he wants to distinguish these defining descriptions from other kinds: "For the description which poets and orators sometimes use expresses a thing more verbosely, nor is it used to express what a thing is but what kind it is, as it places it before the eyes for inspecting." (Agricola, 1563(Agricola, [1485, 42). In this difference between what and what kind Agricola is referencing the "essence" requirement for definitions as well as the stasis distinction in rhetoric between the definitional and qualitative stases. 6 In formal linguistic terms, however, the descriptions themselves arguing for what or what kind cannot really be distinguished, though Agricola suggests that evocative poetic descriptions tend to be longer. The term descriptio used in these passages from dialectical treatises by Boethius and Agricola was of course also a term found in rhetorical treatises as a label for a functional form, a form widely noted in manuals and figure lists under various names from antiquity through the early modern period: hypotyposis, demonstratio, descriptio, tractatio (Sonnino, 1968). In the progymnasmata also, description, or ekphrasis, was a distinct compositional exercise (Kennedy, 2003, 45;86;117;166;218), and though plants were rarely specified as subjects for such exercises, Libanius did produce partial descriptions of the date palm and apple tree embedded in his sample encomia (Gibson, 2008, 261-6 Porphyry, following Aristotle in the Categories, distinguished between predicables that concern what a thing is versus what kind or what manner of a thing it is. "For to the question what manner of thing a man is, we say 'rational.' And to the question what manner of thing a crow is, we say 'black.' (Rational is a difference, and black an [inseparable] accident.) But when we are asked what a man is, we answer 'animal.' (The genus of man was animal.)" (Spade 1994, 3). Poroi 10,2 (December 2014) (Erasmus, 1963(Erasmus, [1514(Erasmus, /1534, 50). There are then converging or overlapping accounts in rhetoric and dialectic of functional descriptions, even descriptions of plants, serving arguments in the definitional and qualitative stases. Early modern students would both practice the textual routines of description in their compositional exercises and learn their philosophical uses in their study of dialectic. 266). Erasmus also listed Pliny's many descriptions of living things as examples of such visualizations in Book II of De Copia Melanchthon recorded his great admiration for Agricola in an encomium on his life, derived in part from conversations he had in Tübingen with elderly scholars who had known Agricola. He even credited Agricola with botanizing in Italy while he was a student of Theodore of Gaza, the translator of Pliny (Reeds, 1976, 527). But Agricola's Dialectica has very few examples from the domain of natural history and those it has, like defining an ass, are traditional. And while Agricola separated oratorical from dialectical definitions on the basis of function, Philip Melanchthon, well known for fusing rhetoric and dialectic, combined them. Indeed in his final and fullest rhetoric text, Elementorum Rhetorices (1542), he claimed that "Definition" as a means of amplification "has here [in rhetoric] the same meaning altogether as in dialectic" (LaFontaine, 1968, 279: "Definitio prorsus hic significant idem quod dialecticis [sic]," translation modified; see also LaFontaine, 1968, 82). Most important for the case at hand, Melanchthon specified a distinct method of definition from combined inseparable accidents and he applied it to botanical description, in effect both reflecting and licensing the usage of this technique in herbals. Melanchthon's comments and application are found in the four dialectical treatises that he wrote spanning his long career at the University of Wittenberg. The first appeared in 1520 just two years after he came to Wittenberg as a professor of Greek and at a time when he was also required to lecture on Pliny's Natural History. In the first book of this brief dialectic, Melanchthon distinguishes definitions of terms from definitions of things, and gives four forms of the latter: from essentials or properties, from causes, from division into parts, and finally, The fourth type of defining [is] definition based on accidents, when many added forms describe a thing in some manner ... Among these we frequently use genus with accidents, as the Halcyon is a bird a little larger than a sparrow, with a predominantly blue color, with purples and whites mixed in the wings, with a slender and elongated head and throat. Larch wood is honey colored, imperishable, with no easily split cracks. Of animate beings, of trees, of gems, of plants, and of similar things, definitions are sought from this type, and we use this type of defining more frequently because the human mode of cognition generally assembles a certain image of a substance from many accidents. Substances themselves are otherwise unknown. For you will not distinguish between Cato and Julius unless you collect distinctions from the figure, body type, face, voice, stature, walk and age. You have the formulas [formulae] of definitions which I urge you to practice diligently and accurately, because a method of definition is necessary in the highest degree, not only for judging axioms but also for inventing arguments (Melanchthon 1521, Biiv; the reference to a mental image here will be discussed below). 7 In his second dialectical treatise written in 1528, the connection to botanical description is stronger, and as examples of definition from combined accidents he cites " …the definitions of plants in Dioscorides and Pliny, such as the Narcissus: It is a flower, similar to a crocus, three-fourths stem, flowering in the month of August, coming after all other flowers" (Melanchthon 1529, E1v). And again for Melanchthon, this type of defining is preferable to a search for elusive substances: "Moreover we should use this type of definition often because accidents show us substances otherwise unknown" (Melanchthon, 1529, E1v). 8 In these first two treatises, Melanchthon does not refer to such definitions as formae but as formulae, adding the diminutive ul to create a cognate term that suggests a brief set form or phrase, a template for content (Melanchthon, 1521, Biii;1529, D3v). The phrase "Formulae Definitionum" [Formulas of Definitions] also appears as a heading in a third dialectical treatise under Melanchthon's name, published in Wittenberg in 1541 with the assistance of his colleague Paul Ebers (Dannenfeldt, 1972, 227). This treatise, though perhaps augmented from material in Melanchthon's lectures, repeats the content in the first two on the tactics for defining things, including definition form a collection of accidents (Melanchthon, 1541, 75-76). But when reviewing the kinds of qualities that count as accidents (color, taste, and sense to the touch), this 1541 version, and this version alone, includes a long list of color terms that would be useful to anyone trying to describe plants (Melanchthon, 1541, 36). Melanchthon produced his final and most extensive dialectical textbook in 1547; this work is addressed not to students, like the first two, but to colleagues and fellow teachers of dialectic. In the intervening years, the Wittenberg medical curriculum had been expanded to include pharmacological botany, and new botanical treatises began to appear that are now famous, both for their naturalistic images of plants and for their increasing inclusion of new species (see below). The physician-botanists producing these and later works had more than a passing connection to Melanchthon and his circle at Wittenberg as Karl Dannenfeldt's research has revealed (Dannenfeldt, 1972). Leonhart Fuchs was his correspondent through the 1530s and 1540s and Charles L'Ecluse, or Clusius, was his student (Ogilvie, 2006, 64;284). Camerarius the Younger was the son of his best friend, and the gifted Valerius Cordus, who is considered by historians to be the most scientifically 8 Quintum genus est, definicio [sic] ex genere et accidentibus collecta, ut sunt herbarum definitiones apud Dioscoridem et Plinium, ut Narcissus, est flos similis Croco, caule dodrantali, florens mense Augusto, postremus omnium florum. Hoc genere definitionum utimur in describendis personis, ut apud Homerum Thersites describitur, strabis oculis, gibbolsus, garrulitate scurrili. Est autem nobis ideo saepius hac forma definiendi utendum, quia accidentia ostendunt nobis substantias alioqui ignotas, per haec Malvam a Marrubio, Cygnam a Corvo, Pompeium a Iulio, in summa res inter se omnes discernimus (Melanchthon, 1529, E1v). inclined of this generation, was his colleague at Wittenberg in the early 1540s. Given these connections among German academics and botanizing physicians, it is perhaps not surprising that the 1547 treatise has many references to medicinal plants (see for example 1846 [1547], 520, 522, 612, 621, 653, 710, in addition to those cited below). In fact the changes and new emphases in this text are arguably a response to the new botany. For example, when discussing differentia and properties in Book I, Melanchthon repeats that everyone knows that differentia are properties, but that when unique properties are unknown, distinctions of things can come from "an accumulation of accidents, as in the descriptions of plants" (Melanchthon, 1521, 522: "ab accidentium coacervatione, ut in descriptionibus plantarum;" this section is followed by a discussion of the difference between separable and inseparable accidents). In the section on "What is called a definition collected from accidents?" Melanchthon uses a more detailed example than he used in 1521 or 1529, one closer to those found in contemporary herbals: A definition collected from an aggregation of accidents is a text [oratio] attaching to the genus or species either the proper accident of the thing [i.e. a unique property] or such a collection of common accidents so that it distinguishes that thing from others ... Many, moreover, use this form in descriptions of plants. Most well known to all is the shrub which bears a fruit, which in our language we call Heimboten. For this fruit the name is Cynosbatos. Its definition is such pertaining to this fourth form: Cynosbatos is a shrub larger than a bramble, in the form of a tree, with thorns around hardy twigs, bearing a fruit with a nut like that of olives, but tinged with red when it ripens, by nature resembling wool on the inside, etc. 9 (Melanchthon, 1846(Melanchthon, [1547, 567-568). 10 9 Cynosbatos is an alternate for the capparis or caper (see the translation of Dioscorides by Ruel, 1516, 56). However, Melanchthon's definition only matches Dioscorides description of the capparis in the detail of the fruit first resembling an olive. Cynosbatos is also mentioned in Pliny (Book 16, Ch. 71) as an alternate name for the dog-rose. It is unclear which plant Melanchthon has in mind or whether he wrote this description himself. As an even more important sign of influence from the new botany, in the fourth book of this treatise Melanchthon adds two new loci that did not appear in his two earlier textbooks on dialectics. The first of these, "Adjacents or Adjuncts," has a label also found in classical manuals like Cicero's Topica, where, however, it is defined for forensic uses as the circumstances surrounding an event (Melanchthon, 1846(Melanchthon, [1547, 418-419). In Melanchthon's 1547 Erotemata Dialectices it is identified as the source for arguments derived from inseparable or perpetual accidents. Again, the prototypical example of this locus is the defining description of plants: The next locus prescribes considering the perpetual accidents of a species, and it is the locus close to Property, because the accumulation of perpetual accidents is in fact the property of a species, just as we distinguish glykyrrisan [licorice] from other roots by its figure and sweetness. The rule handed down by Aristotle is useful. Given a species, the perpetual accidents should be sought next, as writers do in the description of plants … Given a species, the adjacents and adjuncts are posited at the same time, and in turn, destroy the species and at the same time the adjacents and adjuncts are destroyed (Melanchthon, 1846(Melanchthon, [1547, 691-92). 11 10 Definitio collecta ex coacervatione accidentium est oratio generi vel speciei adiungens vel propria accidentia rei, vel talem coacervationem communium accidentium, quae discernat eam rem ab aliis, ut: Ignis est elementum calidum et siccum, et levitate omnibus antecellens. Multum autem hac forma utimur in descriptionibus plantarum, item in individuorum, ut personarum discriminibus indicandis, ut: Mitionem et Demeam discernunt accidentia. Notissimus omnibus est frutex, qui fert fructum, quem nostra lingua vocamus heimboten. Huic frutici nomen est Cynosbatos. Definitio talis est, pertinens ad hanc quartam formam: Cynosbatos est frutex maior rubo, arboris instar, aculeis circa virgas robustis, fructum ferens similem nucleis olivarum, sed rubentem cum maturescit, lanea intus natura, etc. (Melanchthon, 1846(Melanchthon, [1547, 567-568). 11 Sequens locus iubet perpetua accidentia specierum considerare, estque locus hic vicinus proprio, quia perpetuorum accidentium coacervatio est re ipsa speciei proprietas, ut: γλυκυρριςαν discernimus ab aliis radicibus, figura et dulcedine. Utilis est regula ab Aristotele tradita: Posita specie mox perpetua eius accidentia quaerenda esse, ut faciunt scriptores in plantarum descriptione. . . .Posita specie simul ponuntur adiacentia et adiuncta, et rursus: Destructa specie simul destrunntur The next new locus, "Communiter Accidentia" or "Accidents in Common," is devoted to separable accidents which usually though not inevitably occur and so may support an inference. According to Melanchthon, these produce weaker arguments than inseparable accidents do, but their appearance can be used in support of certain probable causal claims such as the prediction that indigestion will, usually if not always, impede the "heat" in the ventricle (Melanchthon, 1846(Melanchthon, [1547, 693). These two new loci offer evidence of the alteration of the classical topical tradition to accommodate the argument practices of sixteenth-century natural philosophers. Melanchthon's distinctions among types of definition and loci, and the botanical examples he uses to illustrate them, provide one source of evidence for the reciprocal connections between natural philosophy and the discourse arts in the sixteenth century. His dialectical textbooks, especially the 1547 version, circulated widely in Protestant schools and universities in the second half of the sixteenth century (Green and Murphy, 2006, 299-301). But Melanchthon's familiarity with the natural philosophy of his day, extensive as it was, was primarily scholarly. Another author of a dialectical textbook at the time was a practicing physician as well as a humanist professor at the University of Frankfurt on the Oder. This humanist physician, Jodocus Willich, also wrote an Erotematum Dialectices Libri Tres that includes a division of accidents into the separable and inseparable as well as the same four tactics of defining things, culminating in the fourth type, the definition from a collection of accidents: "ex genere & congeries accidentium constatur" (Willich, 1540, 19;76). As an example of this fourth type, Willich offers a paraphrase of the full description of glycyrhizza [licorice] found in Dioscorides. Willich's textbook of 1540, reprinted several times, may have been in Melanchthon's mind when he briefly referred to the definition of the glycyrhizza in his new locus of "Adjacents and Adjuncts" (see quotation above). Willich does not expand the loci but his text does contain a further refinement in listing a second type of definition from accidents that clarifies not what but what kind a thing is, putting the "poetic" definition distinguished by Agricola within the domain of dialectic. 12 adiacentia et adiuncta. Sic contra, posita coacervatione adiacentium ponitur species, ut fit in descriptione plantarum. (Melanchthon 1846(Melanchthon [1547 In contrast to the dialectical textbooks from Melanchthon and Willich, both (Eck, 1517). At the opposite end of the spectrum is the very expansive dialectical treatise by Johannes Sturm, which does include a novel third book on Euclidean demonstration, but which also lacks attention to inseparable accidents among definition tactics and loci, as well as botanical examples (Sturm, 1566). The popular dialectical textbook of Johann Caesarius, first published in 1525, does distinguish inseparable from separable accidents but does not link them to definitions and goes so far as to say that descriptions are fine for rhetoricians but not for philosophers (Caesarius, 1525, 156-157). It seems likely then that the changes to the art seen in Melanchthon and Willich stem from their exposure to the revived natural history, and especially botany, of their day. Descriptive Definitions in Early Modern Herbals Recovering strategies from Boethius and Agricola in his first two dialectical treatises, Melanchthon licenses definitions based on perpetual, inseparable accidents and their use as defining descriptions of plants. He does not see himself as innovating in this characterization but as restoring and reaffirming the practice of descriptive definition canonized by its use in Pliny and Dioscorides. Melanchthon's dialectical textbooks from the 1520s precede the new sixteenth-century herbals by Brunfels (1530) and Fuchs (1542) that are seen as milestones in the history of botany. His dialectical in Dioscorides" (Willich, 1540, 79-80). Thus Dioscorides' practice exemplifies two methods of definition. These plants are however, not listed in Dioscorides under those names. Peter Ramus, whose pared down dialectic was first published in French in 1555, does not distinguish types of accidents or expand the loci, but he does distinguish true definitions from the lengthier and celebratory descriptions typical of poets. And after a long example of such a description from the Aeneid, he acknowledges briefly that "telles sont souvent les description des plantes, arbres & animaux es philosophes" (Ramus, 1555, 60). Ramus's dialectic was also in substantial circulation in the Protestant north, especially in England and even New England, throughout the seventeenth century. textbook of 1547, which adds the new loci, comes after those works but before an outpouring of similar ones. Did the verbal descriptions in these herbals match the tactics that Melanchthon defended and that he found in classical sources? What follows is a selection of passages describing a plant known as prassion in Greek, marrubium in Latin and horehound in English; it appears consistently in the herbals, and is a plant that Melanchthon mentions in passing (Melanchthon, 1541, 33). The passages quoted illustrate the actual practice of descriptive defining in sixteenthcentury herbals and furthermore show the persistence of a formula based on inseparable accidents in the definitions of new varieties discovered across the sixteenth century. The first of the new herbals is Otto Brunfels' Herbarum Vivae Eicones of 1530 (Living Images of Plants or Images of Living Plants). It contains striking naturalistic illustrations (see below), but while the images are new, the text itself consists of excerpts from classical treatises, always beginning with Dioscorides' verbal description of the plant, inevitably also a definition complied from a collection of inseparable accidents as approved by Melanchthon. Here is Dioscorides' description of the marrubium from a widely used 1516 Latin translation of the Materia Medica. It is left in the original Latin to facilitate comparison. Brunfels changes the wording a bit but has nothing to add to this description: Marrubii descritio secundum Dioscoridem lib. 3 Frutex est ab una parate ramosus, hirsutus, albescens, quadrangulis caulibus, foliis humano pollici similibus, paulominus que rotundis, hirsutis, rugosis, sapore amaro. Semen per intervalla in caulibus illi est. Flores asperi, verticillorum figura. Nascitur circa areas domorum, & inter rudera (Brunfels, 1530, 160). … according to Dioscorides Bk 3. It is a shrub branching from one source, bristly, light-colored, with quadrangular stalks, with leaves like human thumbs [a comparison for size], smaller and round, bristly, wrinkled, with a bitter taste. The seed appears in intervals along the stem. The flowers sharply pointed, in the figure of little whirls. It grows around the sites of houses and among rough places. Twelve years later, Leonhart Fuchs, in perhaps the most ambitious herbal of the century (discussed below), essentially paraphrases this definition, repeating terms such as verticillatim, "like little whirls," which become part of the formidable special lexicon of botanical Latin (e.g., verticillasters, meaning "aggregates of little flowers"). Dioscorides and other ancient authors knew of two species labeled marrubium (the marrubium proper, sometimes called a white marrubium, and a black marrubium, also called ballot or now ballota nigra [Ruel, 1516, 79]). But newly observant botanizing physicians in the sixteenth century accumulated more varieties over the decades that had not been mentioned in classical sources. Melanchthon's former pupil Charles L'Ecluse, for example, traveled through Austria and Hungary deliberately looking for new plants, and in an herbal published in 1583 he recorded another type of marrubium that he was sure no one had described before, one with narrower leaves and bushier stalks branching from the roots. On the Pannonian Marrubium The common marrubium is rarer in Austria and Pannonia, and is found springing up spontaneously in no more than a few places. But a certain other type is most numerous which has been described by no one that I remember. Many stems spring from the same root, intermittent, quadrangular, sturdy, abundant, divided into many wings: the leaves arising from single nodes, two by two, in opposite positions, two inches long, one wide, serrated on the edge, almost similar to the common marrubium, but narrower, and exactly circular growing out a little in breadth from the pedicle, and gradually spun into a fine point, altogether gray, with a pleasant odor and a warm and bitter taste: flowers white, similar to the marrubium, emerging from a soft and gray calyx; the image of small whirls [verticillasters], they circle the highest stalk and highest branches: the root is similar to the common [marrubium] from which each year new and more numerous stolens [transverse roots] spring (L'Ecluse, 1583, 588-590). Also in the later sixteenth century, Rembert Dodoens notes the two types of marrubium known from antiquity, but he is also aware of still other varieties, and he writes a description in fresh language that nevertheless follows the same content-ordered form as the previous ones. The marrubium properly called, grows several quadrangular little stalks, low-footed, and with thin, whitish down: leaves at intervals in pairs, almost round, rough, with a serrated edge, and themselves with whitish down in maturity. Little flowers in some number with pointed almost white receptacles circle the stalk in little whirls. The seed almost round, bitter tasting, the root in darkish fibers. The entire plant has a sweet and pleasing odor and emits a scent not unlike musk, especially in winter months (Dodoens, 1583, 88). The stylistic form, or formula, of all these descriptive definitions includes keywords (root, stem, leaves, flowers) heavily modified from the lexicon of qualities (words for color, form, shape, texture taste), occasional comparisons, and a stereotyped order, all features approximated in Melanchthon's 1547 textbook description of cynosbatos, quoted above. Usually a "genus" comes first according to Theophrastus' broad categories (tree, shrub, bush, plant). Next the stem, leaves, flower, and seeds are described in that order. Overall the descriptive definition takes the form of an incrementum, a series ordered according to growth, though the roots could be mentioned either first or last. These accounts might seem to be based on a listing of parts, but a definition constructed by enumerating parts had to be exhaustive and all the parts are seldom mentioned in any one definition (Melanchthon, 1846(Melanchthon, [1547, 672). The repetitious language in these definitions is also understandable given that each individual perpetual accident is understood to appear in other species as well; it is only the unique combination of characters, as they came to be called, that distinguishes the individual species. Most important, the stereotyped formulas of these descriptive definitions had heuristic value. They clearly taught the botanist what to observe and record when encountering a new plant. Images as Botanical Arguments If the goal of definition from inseparable accidents is the accurate identification of a plant through a formulaic description, it seems obvious that the visible plant would far surpass any description in words. And indeed Melanchthon's 1547 Dialectic acknowledges the superiority of a first hand encounter with the immediate thing. In a section distinguishing definitions of words from definitions of things, he writes It is the definition of a word when you interpret a word from a foreign language with a more familiar word from our language and you name the genus, as when you say: Centaurium is a plant which we call "tausent gülden" or Aurin; you hear the genus and name as less strange, and yet it can happen that the thing itself is unknown. But if the plant is brought forward and placed under the eyes so that you can consider it, now you have a clear definition. For the ancients said, and it is worth remembering: All intuitive knowledge is definition (Melanchthon, 1846(Melanchthon, [1547 There are three ways to understand this perspective on visualization as definition in the 1540s. First, it is impossible to read phrases like "placed under the eyes" and not recall the rhetorical goal, specified from Aristotle to Erasmus, of a verbal description that creates a visual impression. With rhetorical skill, a plant can be placed before the eyes in words. A second sense, however, seems closer to Melanchthon's meaning in this passage: What is placed under the eyes for definitive inspection is the physical plant itself. That standard was met in the early sixteenth century thanks to the addition of botany to the medical curriculum; physicians in training at Wittenberg and elsewhere went on what we would now call field trips to collect plants, or they inspected them in university gardens or as dried specimens in herbaria (Ogilvie, 2006, 149-152). But a possible third sense also deserves consideration, for a plant can be placed under the eyes for consideration when it is drawn in a life-lie image that supports its identification. The rhetorical goal of "placing before the eyes" then shifts to the available means for producing and reproducing a lifelike image. Manuscript herbals were often illustrated; over twenty illustrated mss. of Dioscorides alone survive from late antiquity through the middle ages (Touwaide, 2008, 55). But a justification for using images at all was needed in the sixteenth century, at least for scholarly publications, because of an often-cited passage in Pliny that pointed out the problems with images of plants and their accurate transmission. … the subject [of medical botany] has been treated by Greek writers, whom we have mentioned in their proper places; of these, Crateaus, Dionysius, and Metrodorus adopted a most attractive method, though one which makes clear little else except the difficulty of employing it. For they painted likenesses of plants and then wrote under them their properties. But not only is a picture misleading when the colours are so many, particularly as the aim is to copy Nature, but besides this, much imperfection arises from the manifold hazards in the accuracy of copyists. In addition, it is not enough for each plant to be painted at one period only of its life, since it alters its appearance with the fourfold changes of the year. (Pliny, 1966, 141-43) This passage lists the impediments to visualization in a manuscript culture. Just as scribes inevitably introduce copying errors into written texts, those copying images commit similar distortions by making elements of a visual more symmetrical, by smoothing out or embellishing features, and by unconsciously adopting changing conventions of representation. Adding color to an image compounds these problems. And a further complication occurs in the case of plants because static images can never fully represent the seasonal stages of growth and decay. The advent of printed images, largely from woodblocks in the sixteenth century, would answer some of these objections, and indeed William Ivins argued over fifty years ago for the importance of "exactly repeatable pictorial statements" in communicating and expanding scientific and technological knowledge (Ivins, 1953, 2). But new modes of image reproduction did not bring immediate changes, and indeed the first printed herbals of the late fifteenth century often merely repeated the schematic images found in most manuscript herbals. Below is the illustration of the plantago minor from the Hortus Sanitatis of 1497 (Fig. 1) which shows the triumph of the copyist's imposed symmetry. Figure. 1 The plantago minor What was needed was b representation and an exponential improvement in the art of cutting woodblocks to create the subtleties of inseparable accidents taken from observations. It is difficult to date a beginning for representation of plants and animals. many isolated exemplars across the centuries satisfying western standards of naturalism in thei The Vienna and Naples codices of Dioscorides dating from the fifth century CE are triumphs of Byzantine realism; illuminated medieval manuscripts often contain accurate depictions of plants; a manuscript herbal created in Carrara like images (Smith popular German herbal the making" for its informative images of plants which its author claims to have drawn from the life, though the actual images are rather crudely cut (Ivins, 1969, new standard of naturalistic depiction did take hold and was combined with improvements "exactly repeatable" true innovations can be plausibly traced to Dürer (Hind, 1963(Hind, [1935, 379; 21 Poroi 10,2 (December 2014) plantago minor from the Hortus Sanitatis, 1497 (v What was needed was both a new standard in naturalistic and an exponential improvement in the art of cutting woodblocks to create realistic images capable of conveying of inseparable accidents taken from de novo It is difficult to date a beginning for "naturalism" in the ion of plants and animals. Art historians have found many isolated exemplars across the centuries satisfying western standards of naturalism in their depiction (see Blunt, 1950, 18 The Vienna and Naples codices of Dioscorides dating from the fifth ntury CE are triumphs of Byzantine realism; illuminated medieval manuscripts often contain accurate depictions of plants; a manuscript herbal created in Carrara, ca. 1400 has stunningly life like images (Smith, 2008, 15-17); William Ivins identifies the pular German herbal the Gart der Gesundheit of 1485 as "epoch for its informative images of plants which its author claims to have drawn from the life, though the actual images are rather t (Ivins, 1969, 34-36). But by the late fifteenth century a new standard of naturalistic depiction did take hold and was improvements in the art of woodcutting to yield "exactly repeatable" true-to-life images. The combination of these innovations can be plausibly traced to the practice of Albrecht (Hind, 1963(Hind, [1935, 379; 386-7). Supervising the woodblock 10,2 (December 2014) viiiv). and an exponential improvement in the art of images capable of conveying Art historians have found many isolated exemplars across the centuries satisfying western 18-30). The Vienna and Naples codices of Dioscorides dating from the fifth medieval manuscripts often contain accurate depictions of plants; a 1400 has stunningly life-17); William Ivins identifies the epoch for its informative images of plants which its author claims to have drawn from the life, though the actual images are rather century a in the art of woodcutting to yield life images. The combination of these Albrecht Supervising the woodblock Poroi 10,2 (December 2014) cutters in his workshop, Dürer enhanced both the fineness and the density of line and hence the detail that woodcuts could convey, and his own drafting practices conformed to the new naturalistic standard in the depiction of objects in nature, on display in his famous watercolor studies of grasses and animals. Skilled woodcuts of naturalistic drawings represented a stunning shift in a technology of visual reproduction that could both achieve true-tolife representation and be easily combined with printed text. Virtually every text on the history of botany, of herbals, and of book illustration tells the same story of the dramatic difference in botanical illustration achieved by German printers in the 1530s and 40s (Blunt 1950, 45-56;Arber, 1912, 47-65). The naturalistic woodcut was brought to the herbal in 1530 by Otto Brunfels' illustrator, Hans Weiditz, who followed Dürer's practice and may have been his pupil (Landau and Parshall, 1994, 237-250;255). Drawn from nature, Weiditz's images depict the often less than pristine state of the specimen in the hands of the artist (see for example Brunfels, 1530, 63). In dialectical terms, these images include the separable accidents unique to the individual specimen. These details enhance the fidelity of the images to observation and therefore work as evidence for the contemporary existence of the plants depicted. But these separable accidents also compromise the effectiveness of the images in support of identifying definitions. Both the methods of descriptive definition and the visualization practices in Brunfels's herbal had detractors. Sachika Kusukawa has detailed the pamphlet war conducted from 1530 to 1540 between Leonhart Fuchs of Tübingen, producer of the next realistically illustrated herbal in 1542, and Sebastian Monteaux, a professor of medicine at the University of Paris over how plants should be defined and imaged (Kusukawa, 1997(Kusukawa, , 2012. In response to a 1530 work by Fuchs correcting errors in plant identification, Monteaux accused Fuchs of basing his definitions of plants on accidents rather than on essences according to the Aristotelian standard. Fuchs defended both his verbal practice as well as the use of images by connecting both modes of description with the usefulness of inseparable accidents: … [Monteaux] childishly contends that arguments derived from the descriptions of medicinal simples, transmitted by Dioscorides and other ancients, are useless because they do not depend on the genus and differentia of a substance but on genus and accidents which can be present or absent before the corruption of the subject. However, Montuus ignores that not all accidents may be present or absent, but there are certain ones that typically adhere in substances so that they are never separated from the thing itself, which in dialectic are called innate, because they are by nature in the subject, as heat in fire, hardness in stone, forms of root, stem, leaves and flowers in plants. And because accidents very obviously reveal substances otherwise unknown to us (for through these we distinguish all things among themselves, like the mallow from the asphodel, the swan from the crow) the form of defining from genus and collected accidents will certainly have the most important uses. For in fact it is a method of human cognition generally to collect one certain image of a substance from many accidents, in the way that Homer did when he described Thersites to have slanted eyes, a pointed head, a humped shoulder, and a clownish garrulity. In short, why would arguments derived by definition from genus and accidents be inefficacious when in general and everywhere we talk periphrastically [circumloquamur] about the differences of things through accidents, nor do many proper names of differentiae exist (Fuchs, 1533, 9v-10;cited in Kusukawa, 1997, 419-420). Here, in 1533, Fuchs repeats the same defense of forming mental images from accidents that Melanchthon used in 1521, while Fuchs notes the similarity to Homer's practice that Melanchthon in turn repeated in 1547. 14 In the magnificent herbal he produced in 1542, De stirpium historia, Fuchs corrected the over-particularity of the Weiditz's 14 Kusukawa first detailed the debate between Fuchs and Monteaux in an article published in 1997. She identifies Fuchs's persuasive use of pictures to confirm an identification of a plant depicted in his herbal with a plant characterized by Dioscorides; she also cites passages from his defense of defining through inseparable accidents and of using pictures despite the criticisms of Pliny and Galen. She notes Melanchthon's account of definitions based on assembled accidents, but not their source or his new loci or his interest in images. She reprises her 1997 article in a recent book (2012), truncating the references to Melanchthon, citing Ian Maclean's discussions of arguments based on separable versus inseparable accidents and adding the influence of Agricola (Kusukawa, 2012;Maclean, 2005). Jensen disagrees somewhat with Kusukawa's first account of Fuchs's enterprise, but she also emphasizes the importance of the "philosophical" distinction between separable and inseparable accidents" (Jensen, 2001). images in Brunfels's to paint the plant from life, one to transfer the image to a woodblock, and one to incise it in relief composite images that featured different phases of the plant's life cycle in one specimen, creating the convention of multi depiction that still survives. The critical difference between Brunfels' and Fuchs's images is a matter of whether the image, as in Brunfels' case, included separable accidents in the depicted details of the individual specimen's actual included only criterial inseparable accidents, as in Fuchs's case. Fig. 2. below, comparing their images of the Geel Violaten in German call a variety of violet is now commonly called a stock Trueblood images in Brunfels's herbal. He used a team of three artiststo paint the plant from life, one to transfer the image to a and one to incise it in relief --to produce idealized, composite images that featured different phases of the plant's life cycle in one specimen, creating the convention of multi-season depiction that still survives. The critical difference between Brunfels' and Fuchs's images is a matter of whether the image, as in Brunfels' case, included separable accidents in the depicted details of the individual specimen's actual condition, or avoided these and included only criterial inseparable accidents, as in Fuchs's case. 312). This plant What Brunfels and Fuchs Matthiola (Meyer, Melanchthon would have supported Fuchs's practice since he believed that images as formed in the mind were generalizations created from observed particulars. His discussion of this issue, amounting to a key justification for the argumentative use of images, appears in Book I of his 1547 dialectic in the section on the Predicables addressing the enduring Platonic question, "Are there universal things beyond the understanding as certain Ideas have been conceived to be universal things beyond understanding?" Whatever thing is truly and positively in nature is something beyond comprehension; it is singular in itself. Things are beyond comprehension: this servant, this man, this horse. But that common image of the stag, which is called the species, is not something beyond comprehension, nor is it, as the Greeks say, hyphistamenon or hypostasis [substance or underlying reality]. But painting that image in the mind is already an act of understanding, which is therefore called general, because it can be applied to many individuals, and carrying around in the mind an image of a stag, it recognizes stags wherever encountered, comparing the figure to the image in the mind (Melanchthon, 1846(Melanchthon, [1547, 520). 15 The generalized image then counts as a representation of a species, and it is worth remembering that in Latin, the root of species (from specere to look at) suggests the thing seen. The image in the mind can be transformed into the image on a page and serve as a source of identification for new encounters. Meeting the new standard of naturalism in the hands of master draftsmen and woodcutters, it can justly support the definition arguments in the new herbals. The definitions through verbal descriptions of the marrubium, quoted above from Brunfels and below from Fuchs, are reprised in the accompanying images. 15 Quaecunque res in natura vere et positive est quiddam extra intellectionem, est singularis per sese. Res sunt extra intellectionem, hic servus, hic homo, hic equus. Sed communis illa imago cervi, quae vocatur species, non est quiddam extra intellectionem, nec est, ut Graeci loquuntur υϕισταµενον, seu hypostasis. Sed est revera actus intelligendi, pingens illam imaginem in mente, quae ideo dicitur communis, quia applicari ad multa individua potest, ut circumferens in mente imaginem cervi, agnoscit cervos ubicunque oblatos, figuram ad imaginem in mente conferens (Melanchthon 1846(Melanchthon [1547, 520). Connecting Words and Images The point bears repeating: knowledge, the images in with the verbal descriptions, were intended as evidence for the definitive identification of a particular living plant with a classical description/definition Fuchs provide an image not only with Dioscorides' verbal description but also with the "living plant" that served as its model. typically connect his text to the images he provides. these two modes of "placing before the eyes" operate in disconnected tandem. Fuchs, on the contrary, very deliberately refers to the accompanying images between the ancient descriptive definition depicted. Here is his text on the marrubium which can be compared to the excerpts provided above. De Prasio/Marrubium: Form It is a shrub branching from one root, somewhat hairy, whitish, with quadrangular stems. The leaf is equal to a thumb, somewhat round, hairy, full of wrinkles, with a bitter taste. The seed [appears] in intervals along the stalk. The flowers [are] sharp-pointed in the figure of little whirls. This picture [referring to a woodcut appearing on the opposite page, see Figure 3] corresponds accurately to what is called marrubium today; in that it sprouts out from a single root with little branches at an angle, full of down, with leaves almost circular in shape, the size of a thumb, very gray, hairy, bitter, with a milk-white flower in little whirls encircling the stem grasped in intervals, from which grows up a tapering and bitter seed (Fuchs, 1542, 589). In effect Fuchs writes the same description twice, the first time as a paraphrase from Dioscorides and the second time as an account of the picture on the facing page. By matching these, Fuchs in effect claims that the plant identified verbally by Dioscorides is the same plant he had illustrated. The check on the circularity of his argument is presumably the naturalistic image itself which stands in for the living plant to support the verbal account. Oddly, Brunfels' practice of not referring to the accompanying image is more typical in sixteenth-century herbals. And, according to Brian Ogilvie, images eventually fell out of favor in serious botanical works by the early seventeenth century (Ogilvie, 2003, 142). The more scientifically ambitious botanical works by Cesalpino and Bauhin are not illustrated, nor are most of the works of the great systematists of the eighteenth century like Tournefort and Linnaeus. Images of plants, however, did not fall out of favor. Woodcuts could not be improved in the sixteenth century beyond the excellence achieved by Weiditz in 1530, because of the roughness of the available paper and the crudeness of the available techniques of inking (Ivins, 1969, 47-49). Engraving became the preferred technology for reproducing naturalistic images, and more accurate and compelling illustrations became possible. But these engravings were also more expensive to produce, so they were more common in special "high end" publications celebrating flowering plants [florilegia] or individual gardens, rather than in herbals or species compendia that would require hundreds and eventually thousands of images. The Scientific Legacy of Formulaic Definitions In the influential dialectical treatises of Agricola, Melanchthon and others, there is ample authorization and some incentive for the verbal and visual practices of the sixteenth-century physicians and natural philosophers reviving the study of botany. But the connection goes both ways as Melanchthon's expansion of the loci in his 1547 treatise shows how the discourse arts could also accommodate a new empiricism evident in the expanding plant catalogs published in the sixteenth and into the seventeenth centuries. While the descriptive definitions in these works, all following a prescribed form with ordered content and repetitive terms, may seem uncreative, there were in fact important pay-offs from this formulaic method. First, the model definition of a plant from a selection of its unchanging accidents made it possible to assimilate the flood of new plants not described in ancient treatises but discovered through expanding trade and travel in northern Europe, the New World and the near and far East (Saunders, 1995, 65). For example, one of the new species recorded by Fuchs in 1542 was sorghum, probably introduced from Africa and at the time called Serpentaria because of the shape of its roots that were thought to have medicinal value. Fuchs knew of no account of this plant among the ancient Greek and Latin physicians, so without a precedent, he was on his own in defining it descriptively: Form: The small stem stands firm, thin and rush-like, bare, with many leaves on the bottommost part, with a spiked flower on the vertex, in which a more dilute color of purple inheres: the leaves like sorrel cutting from the earth, from above [are] green and grassy: as for the root, the male somewhat in the manner of a dragon involuted and contorted; the female truly oblong, larger and hairy, outside blackish, inside reddish (Fuchs, 1542, 772-775). This verbal definition is clearly based on distinctive visible accidents and not on an enumeration of parts. Just enough detail is given for purposes of contrast to distinguish this plant from others like it. Two images are provided of the two varieties, called male and female simply for the sake of differentiation, but there are no references to these images in the text, perhaps because there is no argument to make about the identity of this living species with a classical precedent. A second benefit from formal definitions occurs when the varieties within a species become obvious in juxtaposed texts, as differences in size or color or leaf shape or other visible aspects are highlighted. When several similar species accumulate, the gradations in differences between them become especially obvious. Compiling species lists in the early seventeenth century when thousands, not hundreds of species were known altogether, Caspar Bauhin produces shorter descriptions of two species of just the white marrubium: White hairy marrubium: grows small stalks a foot long, woody, circular, white, hairy, divided in branches: leaves are almost round, small, rough, notched through the edge, below thick white hairs placed further down, like Gnaphalium, above pliant, wrinkled and blackish: little flowers, as in the common, circle the stalk in whirls. Curly white marrubium: this with small stalks is woody, circular, white, but the leaves smaller, curly, serrated, each side; in part ash-colored, with no thick rough hairs: little flowers almost white in pointed receptacles circle the stalk in little whirls (Bauhin, 1620, 110, from Latin Prodromos Theatri Botanici Frankfurt). Bauhin is but a step away from Linnaeus in the Species Plantarum of 1753. This later work organizes some 10,000 species into classes according to the number of stamens and pistils, but below this later-discredited level, it groups species according to genus and adds a distinctive name, eventually establishing binomial nomenclature. In 1753, the format for each entry does not repeat everything that is common to the genus but only includes a brief description of each species featuring the perpetual accidents that constitute the differentia distinguishing that species from similar ones. Here for example are entries for individual species of Leucadendron; the entries are left in the original Latin for purposes of comparison. LEUCADENDRON foliis calloso-tricuspidatis, calycibus turbinatis nudis. (Linnaeus, 1753, 91-93; each of Linnaeus's descriptions is followed by brief descriptions from other published plant catalogs, not quoted here.) For Linnaeus, leaf shape creates the first distinction (foliis = with leaves). When species share the same leaf-shape, the description continues until it finds a distinctive difference in the flowers (floribus), their disposition (corollarum = of wreathes), or in the stem (caule) or calyx (calycibus = with calyxes). The prominence given to leaves, breaking the incrementum order of full definitions, occurs because of their consistency through the seasons when flowers or seeds may be absent. The overall effect of this catalog with its parallel, pared-down descriptions is to show the slight branching differences in allied species, an achievement made entirely verbally since the Species Plantarum required no illustrations. The effect of this standardization into the eighteenth and nineteenth centuries is an often-told story in the history of botany. A systematics based on visible affinities is one benefit. An awareness of environmental influences on plant forms is another, as differences occur in plants growing in different conditions of climate or soil. And eventually, when ever-larger catalogs of allied species, varieties and sub-varieties were amassed in the nineteenth century, a vision far beyond that imagined by sixteenth-century botanists emerged. Poring over the huge compendia of species available in his day, noting the unsymmetrical distribution of species and varieties in various genera, Darwin began to reason that selection pressures had created these slight changes in leaf shape or stem type, and that therefore there is an evolutionary connection from variety to variety and eventually from species to species (Darwin, 1996(Darwin, [1859, 43-50).	2018-01-18T18:47:32.259Z	2014-12-30T00:00:00.000	{ "year": 2014, "sha1": "f92489c1cbf8406d2b0476c129415e0acecad880", "oa_license": "CCBY", "oa_url": "https://pubs.lib.uiowa.edu/poroi/article/id/3425/download/pdf/", "oa_status": "GREEN", "pdf_src": "ScienceParsePlus", "pdf_hash": "3ea3ffcd1fd598f78d18bf6a53626061ec86be02", "s2fieldsofstudy": [ "Philosophy" ], "extfieldsofstudy": [ "Sociology" ] }
234233876	pes2o/s2orc	v3-fos-license	Genetic variability and character association among the quality traits in rice One hundred rice genotypes released for different ecologies such as upland, irrigated, lowland and saline were evaluated for 12 quality characters like Hulling(%), Milling(%), Head rice recovery(%), Kernel length(cm), Kernel breadth(cm),Length breadth ratio, Volume expansion ratio, Elongation ratio and Amylose content(%) etc. Analysis of variance showed high significant difference among the genotypes. Milling(%) is very good in all the genotypes and ranged from 74.5(PR113) to 81.3 (Improved Lalat). HRR (%) is highest in Bhanja (72.0) and lowest in Konark(42.5). Intermediate amylase content are in most of the genotypes except GR 103(27.1),IR 64 MAS(27.86) WGL32100(27.3), WGL32183(27.5)and Purnendu (27.9). Water uptake ranged from 77.5(WGL 32100) to 342.5(PR116) indicating very wide variation. Elongation ratio is > 2.0 in Sudhir, Sashi, Bindli. PCV is always greater than GCV indicating less influence of environment. Hulling (%) is positively significantly correlated with Milling (%) and kernel breadth and negatively significantly correlated with L/B ratio, volume expansion ratio. Introduction Rice occupies a pivotal role in Indian food and livelihood security system. India is the second most populous nation and stand first in areas and second in production followed by China. It is grown in all agro-climatic zones. This wide adaptation leads to evolution thousands of varieties having diverse cooking and eating characters. Before 2000AD, there is demand for increase the production and productivity to meet the food requirement of the growing population. After that, India became self-sufficient and surplus country so far as rice is concerned. People became more concerned about quality than quantity. Previously breed varieties are mostly bold grain, which people do not like. Grain quality in rice is determined by grain appearance nutritional value, cooking and eating quality (Juliano et al.., 1990) [10] . Good grain quality fetches high market price. Demand for better grain quality is increasing day by day in developing and developed countries. Now quality is an important breeding objective in all rice breeding programme. Subudhi et al. (2012) [19] evaluated 42 released varieties of Odisha for their quality characters. It is evident that there is no systematic study of grain quality characters for released varieties of India except some sporadic reports. Now attempts have been made to evaluate the released varieties for their quality traits to find out better donors for hybridisation, popularisation and development of database. In this study, 100 genotypes of different states and for different ecologies were evaluated to find out better donors for hybridisation and popularisation among the farmers. Materials and methods: One hundred released rice varieties of different states were transplanted in randomised block design with two replications. 25 days old seedlings were transplanted with spacing 15x20cm. All the agronomic practices were followed with N:P:K 80:40;40 to raise good crop. The samples are cleaned and analysed in 12-14% moisture content. Results and discussions The analysis of variance showed highly significant difference among the genotypes for all the 12 characters. (table-1). The rice millers prefer varieties with high milling (%) and Head rice recovery (HRR%). But consumers prefer good cooking and eating quality (Merca and Juliano, 1981) [12] . The hulling (%) is very important for the miller and it ranged from74.5 (PR113) to 81.3 (Improved Lalat). High HRR (%) provide more profit to the consumers and millers and It depends on the varieties, grain type, cultural practices and drying conditions (Asish, 2006) [1] . HRR (%) is heritable and very easy to improve (Jenning et al.., 1979) [8] and depend on environmental factors and post-harvest handling (Fan et al.., 2000) [4] . It ranged from 42.5 (Konark) to 72.0 (Bhanja). Grain shape and size are classified according to IRRI (1996) [7] . Consumer preference depend on grain length and thickness. Now most consumers prefer medium slender grains. Kernel length varied from 4.8(Sarasa) to 7.2 (Bhanja). Similarly low value of kernel breadth is preferred by consumers. Bold grains are not preferred by elite class. Kernel breadth varied from 1.75 (Krishnahamsa) to 2.76 (Golak). More water uptake require more energy to cook. So less water uptake will be preferred and it ranged from 77.5 (WGL 32100) to342.5(PR116). High kernel length after cooking (KLAC) looks good in appearance. It ranged from 8.45 PCV is always greater than GCV indicating less influence of environment. Therefore selection on the basis of phenotype alone can be effective for improvement of these traits. High PCV and GCV are recorded in water uptake. Heritability is classified low (<30%), medium (30-60) and high (>60%). Heritability is lowest in hulling (49%) and highest in VER (152%). So there is less influence of environment. The breeders can make selection on the basis phenotypic expressions and high heritability indicate scope for selection of genetic improvement. Similar result was reported by Panwar et al. (1997) [13] , Sarawagi et al. (2000) [17] , Gannamani (2001) [5] , Sao (2002) [15] . Heritability along with genetic advance would give more reliable selection index value. High genetic advance with high genetic advance as mean are controlled by additive genes and can be improved by simple progeny selection. Correlation studies Hulling (%) is positively significantly correlated with Milling (%) and kernel breadth is negatively significantly correlated with L/B ratio, volume expansion ratio. Similar results were reported by Sarkar et al. (1994) [16] , Chauhan et al. (1995) [3] , Subudhi et al. (2009) [18] . Milling (%) is positively significantly correlated with kernel breadth, alkali spreading value but negatively significantly correlated with kernel length, L/B ratio, kernel length after cooking, elongation ratio and amylase content. Head rice recovery is positively significantly correlated with milling %. Kernel length is negatively correlated with Kernel breadth but positive significantly correlated with L/B ratio, alkali spreading value, water uptake and kernel length after cooking. Kernel breadth is negatively correlated with L/Bratio, amylase content, volume expansion ratio, kernel length after cooking but positively correlated with hulling% and milling %. L/B ratio is positively correlated with kernel length, hulling (%) milling (%), kernel length after cooking and amylase content and negatively correlated with kernel breadth Path analysis As simple correlation is not sufficient to provide accurate contribution of characters to Head rice recovery(HRR%). The correlation is partitioned in to direct and indirect effect through path coefficient analysis. The detailed contribution is presented in table-4.The highest positive direct contribution is made by milling (%)(0.474),followed by amylose content (0.122),KLAC (0.090),VER (0.057). The negative direct effect was from kernel breadth(-0.153),elongation ratio(-0.121),water uptake(-0.067),and L/B ratio(-0.047). Milling and amylase content should be given emphasis for selection criteria as these characters are positively correlated to HRR(%).	2021-05-11T00:07:06.553Z	2021-01-01T00:00:00.000	{ "year": 2021, "sha1": "2750577305e6d9b37b365daa77d6040e6a2e280b", "oa_license": null, "oa_url": "https://www.chemijournal.com/archives/2021/vol9issue1/PartAW/9-1-508-613.pdf", "oa_status": "GOLD", "pdf_src": "Adhoc", "pdf_hash": "879ddf3132174459e8f7d0dfef894fe7286dc429", "s2fieldsofstudy": [ "Agricultural And Food Sciences" ], "extfieldsofstudy": [ "Biology" ] }
259719638	pes2o/s2orc	v3-fos-license	Smart City Policy: Strategy and Implementation to Realize Smart Urban Governance in Indonesia : This paper elaborates on policies, strategies, and implementation of Smart Cities in Indonesia. Due to regulatory, strategy, and implementation problems for Smart City, there are still differences in understanding and strategy in its implementation. Smart urban governance is part of a smart city, so one strategy to implement smart urban governance is a smart city policy in Indonesia. There is a fragmentation of inconsistent approaches, policies, implementations, and strategies in practice. This research will develop arguments, analyze published articles and regulations, and map them out. The research method used was a qualitative approach. The data sources used were articles from Scopus which found 213 articles using the keywords smart city policies and smart urban governance and regulations, which were then processed using VOSviewer and Nvivo 12 Plus software. This study indicated that to implement smart urban governance in Indonesia, strategies such as the first need to be supported by the central government to make a policy regarding smart cities, including smart urban governance, followed by smart city policies in the towns in Indonesia. Second, there are technological developments that support the implementation of smart cities. Third, there are institutional and human resource factors, but these are still obstacles to implementing smart cities in Indonesia. Meanwhile, the urban issues developed in Indonesia are mega-urban in the big island, urban management and financing INTRODUCTION A smart city concept was originally aimed at improving the quality of an area/city by making the area/city become "smart," balanced with improvements in Information Communication and Information Technology Technology (ICT) (Arafah & Winarso, 2020). The development of the Smart City increased when the European Union created the Smart City in 2010, aimed at creating smart solutions to be able to face the challenges of urbanization, globalization, and climate change (Cocchia, 2014;Jucevičius, R., Patašienė, I., & Patašius, 2014) because urbanization, globalization, and climate change are complex problems. Moreover, the issue of urbanization that urban areas must face is due to the movement of people from villages to cities. So that urban areas are increasingly dense with a population that is not matched by the expansion of urban land and other problems arise. World urbanization data estimated that in 2050 there will be approximately 66% of the world's population living in urban areas. The percentage of urban in Indonesia will continue to increase every year. Even in 2025, it is estimated that 67.66% of the population will live in urban areas. In 2045, it is estimated that around 82.37% of the population will live in urban areas due to urbanization. While Woldometers noted urbanization in Indonesia, in 2019, approximately 55.8%, equivalent to 150.9 million people of Indonesia's population, lived in urban areas. This figure increased by 0.7% in 2018, 55.1% or equivalent to 147.6 million people (Khairunnisa, Purnomo, & Salsabila, 2020). Some cities in Indonesia have implemented the Smart City, including Jakarta, Bandung, Makassar, Surabaya, Semarang, Yogjakarta, and Denpasar, followed by other cities in Indonesia. Thus, cities must provide necessities such as healthy food, clean water, sufficient energy, and certainty about economic, social, and environmental stability. With Smart City, the city government will hopefully solve various problems quickly and precisely in serving public needs. The government can also use this information to create comfort, security, order, and a better life. The government has the duty and function to serve the people. Therefore, the key to the success of good governance is to look at or look in the mirror at public services that are no longer adequate to overcome problems. An integrated approach is needed to overcome all difficulties in the city so that a Smart City concept is born (Insani, 2017). Smart urban is one part or dimension of a smart city. A smart city strategy and implementation are needed to realize smart urban governance, including smart environment, mobility, government, economy, people, living, and disaster management. Smart city strategies are needed for city services that refer to smart urban governance. The government makes regulations to control to ensure that the community gets a healthy environment and guarantees business actors to invest as part of a smart city (Susena, B., & Widowaty, 2018). Cities that successfully implement the smart city concept have breakthroughs in solving problems and improving city performance. One of the most critical dimensions of a smart city is to provide services using the latest technology and build smart infrastructure to deliver effective services to everyone living in the town (Insani, 2017;Sudaryono, 2014). Figure 1. Smart City Framework Source: (Chourabi et al., 2012) Based on Figure 1, the smart city framework created by (Chourabi et al., 2012) explains that external and internal factors can affect the success of a smart city. External factors that can influence the success of a smart city are people, communities, governance, natural environment, built infrastructure, and economy. At the same time, the internal factors that can affect the success of a smart city are technology, organization, and policy. The internal factor of the success of this smart city greatly affects whether or not the policy runs to build a smart city. It must be supported by digital and integrated city development with support for physical infrastructure, smart technology, high mobility devices, and computer networks with adequate (technology dimensions). In addition, creativity, knowledge, education, and learning are needed as the main drivers of the formation of a smart city, where manual problems are transformed with knowledge into digital system models through creativity and presented in learning that fully needs to be consistent to be implemented (human resource dimension) and also needed support from the government and policies for governance as the basis for smart city design and implementation. Policies support and play a role where relationships are formed between government institutions and non-government parties, and other sectors in building an integrated administrative environment (institutional dimension) (Hasibuan & Sulaiman, 2019). The smart city concept will succeed if the government understands and contributes to the community's needs and is supported by the highest level of government (Insani, 2017). According to Cohen, "Smart cities use information and communication technologies (ICT) to be more intelligent and efficient in the use of resources, resulting in cost and energy savings, improved service delivery and quality of life, and reduced environmental footprint-all supporting innovation and the low-carbon economy." A smart city can measure or see how far the implementation is in a town or area. Table 1. Smart City Implementation Level Implementation Status 0 a. The initial level of implementing the Smart City b. At this level, the situation is still an ordinary city, but there is potential to become a Smart City 1 a. The starting level of a city or area becomes a Smart City b. This level is characterized by the availability of the internet throughout the city area 2 a. Continuation stage of the first level b. This level is marked by a city that has started to connect with networks in other cities or has implemented a Metropolitan Area Network (MAN) 3 a. Already at the level of open information b. This level is marked when the city has openness with other cities to share data and information online 4 a. A level that already has the process of processing data and information using good security b. Every data accessed is maintained the value of importance contained in the data and information 5 a. This level is characterized by being well integrated within the city and between cities as a combination of levels 2,3 and 4 Source: (Hasibuan & Sulaiman, 2019) Smart cities are needed because many people live in urban areas/areas. This high level of urbanization will bring up various problems that can damage strategic development plans and suck up urban facilities (Khairunnisa et al., 2020). To be able to realize a smart city concept in these cities, several steps are needed, such as encouraging and developing new patterns with leadership and governance structures, collaborating with all parties involved by building and using "smart" infrastructure and preparing models financing capable of responding to future challenges and opportunities (Khairunnisa et al., 2020). However, it is not easy to realize a smart city because it requires strategic and creative thinking. Strategic, in the sense that it must be done in a planned manner for the future (futuristic), starting from problem identification, grouping problems, abstraction processes, determining ways and solutions to problems, as well as planning for implementation (Nugraha, 2014) (Insani, 2017). At the same time, creative thinking is an ability that can produce valuable and new works, both at the individual and organizational levels (Insani, 2017;Lazuardi, 2015). In the 1999 United National Habitat campaign, urban governance needed to achieve the development of human settlements that adhere to the principles of sustainability, subsidiarity, equality, justice, efficiency, transparency, responsibility, civil order, citizenship, and security which are interdependent and mutually reinforcing (Muslim, Ardila, 2020;UN Habita, 2002). Urban governance focuses on the stakeholders' capacity because achieving the expected results comes Tri Sulistyaningsih 1 , Rika Ayu Purnama 2 , Umi Kalsum 3 \| 108 from dialogue and negotiations of actors and strong strategic direction and leadership in an institution (Muslim, Ardila, 2020). Urban governance aims to respond to various problems related to urban development effectively and efficiently, held accountable by the government and the community (Latifa, 2013). Urban governance examines how local, regional, and central governments and stakeholders determine and decide on an urban area's planning, financing, and management (Safaruddin, 2018). According to (Slack, E., & Côté 2014), there are four points in urban governance, namely to play an essential role in shaping the physical and social character of urban areas; affect the quantity and quality of local services and delivery efficiency; determine cost-sharing and distribution of resources between different groups; affect the ability of citizens to access local government and be involved in decision making, affect local government accountability and be responsive to citizen demands (Safaruddin, 2018). There is fragmentation in implementing smart cities in several cities in Indonesia because big cities in Indonesia have many similarities and differences. Applying Smart City in each of these big cities has a different background. Implementing a smart city in Jakarta and Surabaya is different. Likewise, the smart city developed in Bandung and the smart city in Makassar also have differences at the smart city level. The difference in the potential of natural and human resources impacts how a smart city will be built. RESEARCH METHOD This study employed qualitative research techniques in conjunction with a literature review strategy. It is a complete summary of the research on a certain topic to show the reader what is already known about the issue and identify gaps in knowledge and also used to justify decisions based on past research (Denney, Andrew S., 2013). The research design using this literature study focused on the results of the paper that followed this research, elaborating on smart city policies, strategies, and implementation in realizing smart urban governance in Indonesia. The research design will be described in Chart 1. Chart 1. Research Design Looking for the main problem and background This data collection technique was obtained through various stages, (1) Journals obtained from Scopus were the result of using the keywords "smart city policies" and "smart urban governance" and obtained 314 journals. (2) After that, the authors filtered again by year, articles published starting from 2015-2021 with the subject area of social science and environmental scientist. So from the filtering, 213 articles appeared. (3) In the next step, the authors obtained various articles/journals based on those that appear on Scopus and downloads via google scholar. (4) After downloading the articles/journals, the authors enter the articles into Mendeley for later processing using Nvivo 12 plus and Vosviewer software. The data analysis method used the Nvivo 12 plus and Vosviewer software. Using Nvivo software, 314 articles were grouped into 213 journals by creating a themes mapping. In the next step, the authors analyzed the 213 articles from Scopus and related them to the study topic of smart city policies and urban governance. Finally, the authors tried to make the results of the concept map of the study from the results of the literature review. The data analysis technique steps are as follows (1) entering the literature review that has been obtained from Scopus as many as 213 articles and downloaded via Google Scholar to Nvivo 12 Plus and Vosviewer, (2) conducting the coding process based on the theme, (3) reviewing results and discussion based on linkages and dominant themes in smart city and smart urban governance studies and clusters and theme relationships in smart city and smart urban governance studies, (4) collecting all data obtained from Nvivo 12 Plus and Vosviewer and analyzes data explanations, and (5) making report the conclusion of the data analysis results. Figure 4 shows several themes related to and still connected with the concept of urban governance in Indonesia. Smart urban governance is part of a smart city, so one strategy to implement smart urban governance is through a smart city in Indonesia. Smart urban governance is a step that must be taken by the government, especially in big cities that have a high population density and complex problems. Given the problematic issues, especially in big cities in Indonesia, smart urban governance is needed to manage and develop settlements prioritizing sustainability and using technology as part of a smart city. Smart Urban Governance in Indonesia It can be seen in Figure 4 that smart urban governance has a close relationship with urban, policy, city, public, change, climate cities, and sustainability, proved by the appearance of words printed in bold on the results of the Nvivo 12 Plus (Figure 4). Smart urban governance certainly has a close relationship with "urban" "city" because the problems and policies of smart urban governance are suitable to be applied in cities with dense populations with various problems brought about by an imbalance between the population and the available land. The smart urban governance step is closely related to climate change and sustainability. Since climate change can occur because of the habitual pattern of the population that does not reflect urban sustainability, smart urban governance will show an urban area that prioritizes sustainability. Table 3 explains that the strategy and implementation of smart urban governance in Indonesia start with urban, public, city, social, and policies. These five things are interrelated and strengthened. To realize smart urban governance in an urban area is a facility provided for the public and the public's interests, welfare, and convenience. So, to make it happen, a policy made by the government is needed because the government has the authority to make these policies and implement them together with other stakeholders and the community. In Figure 5 are the concepts related to realizing smart urban governance in Indonesia. The related concept is the compact city, which is closely associated with implementing urban governance. In addition, it is closely associated with the urban form because a structured and sustainable urban arrangement is needed to realize smart urban governance. So structured urban planning is required to implement smart urban governance's initial objectives and manage cities with complex problems. Technological urbanism is also necessary for smart urban governance because the system depends on technology. It will create an urban area where all public services are interconnected/integrated using technology. In 2006 studies on urban governance began to be studied frequently, especially on urban form, urban sprawl, compact cities, urban containment, neo-traditional development, design, and sustainability theory. Furthermore, in 2008 the focus of studies often carried out was urban planning and eco-city. Studies on commuting were constantly reviewed in 2010, which are still related to smart urban governance. In 2012, the study focused more on planning related to urban governance. Moreover, the direction of the survey in 2014 was more on technological urbanism, which is closely related to implementing smart urban governance that relies on technological advances. In 2016, the study focused more on regenerative cities and ecological wisdom, which also support the implementation of urban governance in an urban area. Figure 6. Overlay visualization of urban governance Source: processed using VOSviewer Table 4 explains that urban governance has 5 cluster concepts. The first cluster consists of eco-city, planning, urban containment, neotraditional, development, design, compact city, urban form, technological, urbanism, and techno-city. At the same time, the second cluster consists of the concepts of planning, urban form, compact city, eco-city, urban containment, neo-traditional, development, design, regenerative cities, ecological wisdom, resilience, transportation, and history. The third cluster comprises urban form, compact city, sustainability theory, urban sprawl, urban planning, eco-city, urban containment, neotraditional, development, design, and planning. The fourth cluster comprises urban planning, human biometeorology, climate, index, sustainability theory, urban sprawl, urban form, and compact city. Finally, the fifth cluster comprises urban sprawl, urban planning, sustainability theory, compact city, urban form, commuting, traffic, and congestion tools. Based on Figure 7, to implement smart urban governance in Indonesia, one must start with several things, such as national government, urban processes, cultural processes, one-way process, democratic process, criminal justice pro, easy process, design process, research process, passing process, transformative process, ecological processes, evolutionary process, similar processes, transformation process, local policy process, dynamic process, hydrologic processes, genetic processes, and biological processes. Figure 7. A Priority of Urban Governance Source: processed using Nvivo 12 Plus Discussion This study's findings indicate that to implement smart urban governance in Indonesia, an effort or strategy is needed, such as. The central government must first support making a policy regarding smart cities, including smart urban governance, followed by smart city policies in the towns in Indonesia because a policy is an initial foundation for implementing a smart city. The policies contained in the smart city must be clear about the mechanism for the implementation process, how the budget must be issued, and the approaches must be integrated and consistent with each other. Because if a policy content or mechanism is contradictory and inconsistent, it will hinder the implementation of a smart city. As with the implementation of smart cities in Indonesia, there is still no consistency in its policies, and it is still not integrated between them. All procedures must be integrated to implement a smart city that covers all aspects, including urban matters, because a smart city is a system that will make all public services systemized and integrated. Second, the strategy that must be carried out to implement smart urban governance in Indonesia is the development of technology that supports the implementation of smart cities so that all urban services are online-based using an integrated internet site and cover all urban services (Allwinkle, Sam & Cruickshank, 2011). There are still no fully online-based urban services in Indonesia because procedures must go to government agencies to manage or upload files. Furthermore, in addition to technological developments, institutional and human resource factors must support smart cities (Nam, T., & Pardo, 2011). Whereas what happened in Indonesia, smart cities were constrained by two factors or one of these factors. This human resource factor includes employees in all government agencies/agencies related to urban and community services. Because these factors must be balanced, when the local government implements a smart city in its area, then the human resources in government institutions/agencies must be technology literate, and the community as the target user must also be technology literate so that when a smart city is implemented, it does not become a wasted system. In addition to the policies of the central and local governments, the development of internet technology, institutional factors, and human resource factors, several things support the implementation of smart urban governance in Indonesia, as shown in Figure 8. Figure 8. Urban Development in Indonesia Source: (BAPPENAS, 2015) Five urban issues must be developed in urban governance in Indonesia. The first concerns mega-urban in big islands, carried out by standard urban services for large areas. The second is urban management and finance by paying attention to three aspects: city financing, the development of smart cities, and city management. The third is housing for lower-income-the fourth is related to sustainable urban infrastructure. And the fifth is about resilient disaster mitigation and management, which consists of three aspects: early warning system, green urban governance, and standards and procedures. An urban that will transform into a new urban plan must focus on policies and strategies to achieve effective results. There are vital elements to achieve this: national urban policies, regulation in urban governance, and urban economy. Table 5 shows the central government's roadmap regarding national urban development from 2015 to 2045. In 2015 it was the basis or the beginning of urban development. Then in 2025, it is scheduled to fulfill the Urban Service Standards (SPP) and livable cities. The year 2035 is planned so that all cities in Indonesia have reached 100% of the fulfillment of green cities. And by 2045, it is scheduled that all cities in Indonesia will have reached smart and competitive cities and have reached 100%. All cities achieved 100 percent compliance with indicators for green cities 2045 100 percent of smart and competitive city indicators were attained for all cities. Source: (BAPPENAS, 2015) Figure 9 explains the central elements of achieving a sustainable urban in Indonesia. The first element is a national policy regarding sustainable urban development and planning. The second element is sustainable urban governance, a local fiscal system; the third element can implement sustainable urban service standards. The fourth element provides indicators of sustainable urban performance to see every action or plan towards a sustainable urban. Table 6 shows several national programs for urban development in Indonesia. The first program concerns technical assistance conducted by local government and urban development partners. This program is carried out in several areas in Indonesia, such as Banda Aceh, Balikpapan, Denpasar, Palembang, Tangerang, Surakarta, Yogyakarta, Semarang, Palu, Probolinggo, Kupang, and Surabaya. This program has main activities such as technical assistance to prepare technical documents for the city infrastructure program and to link to alternative finance. CONCLUSION In Indonesia, smart cities have no consistency in their policies, and policies are still not integrated because there are no fully online-based urban services. There are still procedures that must go to government agencies to manage or upload files. To succeed in smart cities and smart urban governance, human resources in government institutions/agencies must be technology literate, and the community as the target user must also be technology literate so that when a smart city is implemented, it does not become a problem useless system. The seriousness of the government in implementing smart urban governance is shown by the roadmap that has been made by the central government regarding national urban development starting from 2015 to 2045. Five urban issues developed in urban governance in Indonesia are regarding (1) mega urban in big island, (2) urban management and finance, (3) housing for lower-income, (4) sustainable urban infrastructure, and (5) early warning system, green urban management, and standards and procedures. Implementation of smart urban governance in Indonesia must start with several things, such as national government, urban processes, cultural processes, one-way process, democratic process, criminal justice pro, easy process, design process, research process, passing process, transformative process, ecological processes, evolutionary processes, similar processes, transformation processes, local policy processes, dynamic processes, hydrologic processes, genetic processes, and biological processes. The strategy to achieve smart urban governance in Indonesia requires an effort, the first of which must be supported by the central government to make a policy regarding smart cities, including smart urban governance, followed by smart city policies in the towns in Indonesia. Second, the strategy that must be carried out to implement smart urban governance in Indonesia is the development of technology that supports the implementation of smart cities so that all urban services are based online using an integrated internet site and includes all urban services.	2023-07-12T06:00:47.538Z	2023-06-16T00:00:00.000	{ "year": 2023, "sha1": "6425de5f271c3c3f3b70c560a7009c933e6f1a43", "oa_license": "CCBYNC", "oa_url": "https://journal.umy.ac.id/index.php/GPP/article/download/13840/8429", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "9948dc7b08a351abb9b34bea77f8ab41d74d787b", "s2fieldsofstudy": [ "Economics" ], "extfieldsofstudy": [] }
49339081	pes2o/s2orc	v3-fos-license	Free Cooling of a Granular Gas in Three Dimensions Granular gases as dilute ensembles of particles in random motion are not only at the basis of elementary structure-forming processes in the universe and involved in many industrial and natural phenomena, but also excellent models to study fundamental statistical dynamics. A vast number of theoretical and numerical investigations have dealt with this apparently simple non-equilibrium system. The essential difference to molecular gases is the energy dissipation in particle collisions, a subtle distinction with immense impact on their global dynamics. Its most striking manifestation is the so-called granular cooling, the gradual loss of mechanical energy in absence of external excitation. We report an experimental study of homogeneous cooling of three-dimensional (3D) granular gases in microgravity. Surprisingly, the asymptotic scaling E(t) ∝ t−2 obtained by Haff’s minimal model [J. Fluid Mech. 134 401 (1983)] proves to be robust, despite the violation of several of its central assumptions. The shape anisotropy of the grains influences the characteristic time of energy loss quantitatively, but not qualitatively. We compare kinetic energies in the individual degrees of freedom, and find a slight predominance of the translational motions. In addition, we detect a certain preference of the grains to align with their long axis in flight direction, a feature known from active matter or animal flocks, and the onset of clustering. Most prominent is the permanent loss of kinetic energy in absence of external forcing, called granular cooling [29]: Starting from an initially excited state with spatially homogeneous statistical properties, the ensemble enters an initial period of homogeneous energy loss. At longer time scales, the grains can spontaneously cluster. Such granular clustering is a key ingredient for the formation of planetesimals and larger objects in solar systems [30][31][32]. Preparation of a freely cooling granular gas in 3D is challenging, it is practically impossible under normal gravity. Sounding rockets, satellites and drop towers offer appropriate conditions [41]: excellent microgravity (µg) quality, down to residual accelerations of 10 −5 m/s 2 . In a pioneering µg experiment, dynamical clustering of monodisperse spheres was reported by Falcon et al. [2], but a quantitative analysis at the grainlevel was not possible for technical limitations. Rodshaped grains offer experimental advantages over spheres: a much shorter mean free path at comparable filling fractions [24,42], a more random energy injection by vibrating container walls [43,44], and an efficient energy redistribution among all degrees of freedom (DoF) in collisions [36]. The latter two features can reduce spatial inhomogeneities. Translations and rotations can be followed in 3D [45]. We present results of the first experimental investigation of a homogeneously cooling 3D granular gas. II. EXPERIMENT Ensembles of 374 rods of = 10 mm length and d = 1.35 mm diameter are studied in a container of 11.2 cm × 8.0 cm × 8.0 cm (Fig. 1 a,b) during ≈ 9 s of micro-gravity realized in the ZARM Drop Tower, Bremen. The corresponding volume fraction of grains is φ = 0.75 %. The mean free path estimated from the filling fraction and the rod dimensions is λ ≈ √ 2d 2 /[φ( + 7.55d + 2.02d 2 / )] ≈ 1.65 cm, well below the Knudsen regime (λ > container size). The rods are custom-made from insulated copper wire, their mass is m = 37.5 mg, moments of inertia for rotations around the rod axis and perpendicular to it are J = 4.6 pN m and J ⊥ = 315 pN m, respectively. a. Steady excitation state: During the initial 2 seconds of microgravity, the grain ensemble is excited mechanically. In the initial, driven state, one finds an excess of translational energy in the direction of excitation x (normal to the vibrating walls), E x (0) ≈ 135 nJ per grain. Rotations and translations in the directions y and z are only weakly excited by the vibrating walls [44], they are driven via rod-rod collisions. The indirectly excited spatial directions have equal average energies, initially E y + E z /2 ≈ 90 nJ, and E rot /2 ≈ 64 nJ for rotations about the short rod axes (these two rotations cannot be distinguished in our experiment, we can only determine their sum E rot ). The violation of equipartition in the driven state has been described earlier [24,45]. After the excitation is stopped (t 0 = 0), videos are recorded and a particle-based statistical analysis (see Methods) is performed to evaluate the evolution of the ensemble dynamics. Of primary interest is the energy partition and energy loss. The total kinetic energy decays by almost 3 orders of magnitude during the observation period of about 7 s, as seen in Fig. 2. b. Spatial homogeneity during cooling: Throughout the cooling process we found only a marginal tendency of clustering. This is difficult to evaluate quantitatively but visualized best by overlaying image sequences from a single experiment. Figure 1c shows two such typical overlays. The upper one was recorded immediately after t 0 , eight subsequent frames of the top camera video were overlayed, the field of view is more or less uniformly covered by rods. The bottom view is a superposition of eight frames from the final phase of the experiment. Since the mean velocity is slower by a factor of ≈ 15, every 15th frame was chosen for the overlay. One can recognize some inhomogeneities in the particle distribution, but the ef-fect is marginal. c. Partition of the kinetic energy: After the excitation stopped, the kinetic energy is gradually redistributed by collisions. Thereby, the partition among the DoF changes drastically, as shown in Fig. 2, top. Within statistical fluctuations, a steady distribution of the kinetic energy among all DoF is reached after ≈ 2 . . . 2.5 s. The initial excess of E x has vanished. However, the kinetic energies of rotations around the short rod axes remain slightly smaller than those of the translational DoF, as seen in Fig. 2. These results are in qualitative agreement with simulations of frictionless ellipsoids [35] and rods [36] that predicted a small excess of translational over rotational energies per DoF. The third rotational degree of freedom, rotations around the rod symmetry axis, is only excited by frictional contacts of particles in collisions. The ratio of the moments of inertia is J ⊥ /J ≈ 70, therefore such rotations need to be almost one order of magnitude faster to reach equipartition. For an estimate, we marked a few rods with dots to track their axial rotations. They were evaluated during the cooling process, although with much poorer statistics than for the other DoF. The related mean kinetic energy turns out to be about one order of magnitude lower than those of the other DoF [46]. d. Cooling: Haff [29] predicted that the mean energy of a freely cooling granular gas of frictionless spheres obeys the scaling yielding dissipation rate ∂ ∂t Here, ρ is the mass density of the system, s is the mean grain separation (in our dilute system to be replaced by the mean free path λ), and ε is the normal restitution coefficient,v is the mean absolute velocity. The factor ξ depends upon the dimensionality of the system, it basically accounts for the fact that only the relative velocities of colliding particles (in our system also relative rotations) are relevant for the energy loss, which is then redistributed among all DoF. Haff makes the assumption that the distance between the grains is small compared to their diameter. He sets the quantity v 2 , which is related to the average kinetic energy, equal tov 2 , the square of the mean absolute velocity, that determines the collision ratev/s. The Haff time describing cooling from a given instant t i is given Villemot and Talbot [35] have extended this model to ellipsoidal particles, and Rubio-Largo et al. [36] considered ellipsoids and spherocylinders, but aspect ratios in both studies were well below that of our rods. The corrections concern the prefactor ξ but leave the rest of the predictions unchanged. Figure 3a shows that Eq. (1) fits the experimental data of the mean total energy qualitatively very well after some initial period of ≈ 1.25 s. Kanzaki et al. [33] predicted an exponent −5/3 of the decay of translational DoFs for viscoelastic (ε(v)) anisotropic grains in 2D from numerical simulations. In clear contrast, our experiments are in excellent agreement with an exponent −2 for the long-term decay. The initial discrepancy is easily understood: Immediately after excitation, the system is not spatially homogeneous, for example, particles are 'hotter' near the exciting plates. The individual DoF are at very different granular temperatures. Therefore the Haff fit of the data after Fig. 3a) overestimates the initial E(t). After about 1.25 s, the system is in the homogeneous cooling regime. This is confirmed by the analysis of the mutual dependence of τ H andv ( Fig. 3b: The experimental data were fitted with Eq. (1) starting at different initial times t i > 0 during cooling, and the inverse of the fit parameter τ H (t i ) was related to the momentary mean absolute velocitiesv(t i ). The linear fit confirms Eq. (3) and yields the fit value ξ = 18.9 m −1 λ/(1 − ε 2 ). s (dashed line in A further test of the model is the experimental determination of the cumulated collision number per particle Figure 4). The dashed line is a fit through the data for t > 1.25 s. Again, the initial values deviate from that fit, partially because the homogeneous cooling is not yet reached, and partially because the particles are very fast so that not all collisions may have been detected unambiguously in the videos (both reflected in the offset). Together with the fit in Fig. 3, this yields a value λ = 0.31/18.9 m ≈ 1.64 cm, in excellent agreement with our geometrical estimate. In our experiment, the following complications affect the interpretation of the factor ξ: In Haff's model, the energy loss per collision is distributed among three DoF. In our system, translational energy is partially converted in rotational energy and vice versa [34,36], and the loss of rotational energy per collision is much more difficult to estimate than for frictionless spheres. In addition, the kinetic energy share of rotations about the long rod axis is known only to the order of magnitude. For an estimation of τ H with Eq. (3), ε andv can in principle be determined experimentally, but the factor ξ is unknown. Numerical simulations [35,36] may give some hints, but it is not clear whether these results can be extrapolated to our particles with aspect ratios above 7, and whether the results of Ref. [35] can be applied to rodlike particles at all. An additional problem is that we find a preferential alignment of the rods in flight direction (see below), so that the collision statistics is not the same as for completely random rod orientations. The velocity distributions of the components v x , v y , v z are non-Gaussian. Within our statistical accuracy, their shape remains nearly unchanged during the homogeneous cooling process. The kurtosis has no trend in our experimental data during homogeneous cooling, it fluctuates between 3 and 4, with an average of 3.5. The measured ratio of v 2 and v 2 was between 1.2 and 1.32, slightly increasing with time. This is in clear contrast to Gaussian velocity distributions reported in simulations [35,36]. e. Alignment: One particular feature of anisotropic grains is their tendency to align in shear flow [47], or in active matter [48]. Correlations between velocity and orientation were reported in simulations of hard needles [49]. We find similar correlations in the distribution of relative rod orientations p(θ), where θ = 0 • resembles spear-like orientation. Immediately after excitation, the alignment angles are distributed almost isotropically (p(θ) ∝ sin θ), this can be attributed to the random excitation by the vibrating container walls. During cooling, rods are more often in spear-like orientation, the probability of angles θ < 45 • increases, while it decreases for θ > 45 • (Fig. 5). This can be understood intuitively: rod-rod collisions are more probable when the rod axis is perpendicular to the flight path. The consequence is a slightly larger λ than es-timated for random rod orientations. The effect is small, though, as illustrated in the bottom images of Fig. 5. Note that the highest possible alignment of rotating rods would be an equal distribution p(θ) = 2/π (black line in the image), when all rotations occur about short axes perpendicular to the flight path (like spokes of a wheel). III. CONCLUSIONS AND SUMMARY The experimental study of homogeneous free cooling of a 3D granular gas in micro-gravity demonstrates that Haff's scaling law of the energy loss with time is surprisingly robust, even though several central assumptions are not fulfilled, e.g. friction and shape-anisometry of the grains, and non-Gaussian velocity distributions even in the homogeneous cooling state. Energies become nearly equally distributed among the DoF in the homogeneously cooling state, with a slight excess (≈ 10..20 %) in the translational DoF. Even the purely friction-coupled rotations around the long axis are excited, albeit with one order of magnitude lower mean energy. A gradual alignment of rods in flight direction is also documented, explainable by the lower collision probability in this flight orientation. The detailed mechanisms underlying the collective dynamics of this system e.g. the exact role of particle shape, contact parameters, confinement and spatial inhomogeneities, are still to be explored. The present data may serve as benchmarks for computational studies. In perspective, only the combined efforts of theoretical work allowing for conditions realizable in experiments, and quantitative experiments with materials of different shapes, friction and elastic properties, may bridge the gap from individual grain collision to ensemble dynamics of granular gases in an extendable and realistic manner. ACKNOWLEDGMENTS The authors acknowledge funding by the German Aerospace Center DLR, projects 50WM1241 and 50WM1344, and by the German Science Foundation (DFG) grant STA 425/34-1. We cordially thank the ZARM staff in Bremen for their excellent support in the drop tower experiments. IV. METHODS The setup consisting of a box with two laterally movable side walls, illumination and two cameras recording perpendicular perspectives (similar to Ref. [24]), see Fig. 1 is integrated into a ZARM catapult capsule. When the capsule is released from the catapult, the rods lift off from the base plate. We prepare an initial excited state by injecting energy through vibrations of the lateral walls in x-direction (3 mm amplitude, 30 Hz) for 1.5 . . . 2 s. The velocity distributions in this initial state are spatially inhomogeneous (see, e. g. [41,50]), while the spatial distribution of rod positions and orientations remain almost homogeneous throughout the complete experiment, see Fig. 1c and supplementary movie. After the excitation is stopped, 3D trajectories of the colored rods in the ensembles are tracked in the two perspective views. Black rods provide the thermal background, it is technically too demanding to distinguish and track more than 50...60 rods in the ensemble. Because particles are sometimes obscured by others, individual trajectories cannot always be followed continuously, see Fig. 1. Partially necessary manual tracking is the most time-consuming task of the data analysis. We mark all collisions in the trajectories and reorientations manually and fit segments of the translational motion between collisions linearly to reduce sampling noise. Angular velocities are computed from the vector product of the rod orientations in consecutive time steps, and their absolute values are averaged in the intervals between collisions for the same reason. Only rotations about the short rod axes are analyzed systematically. The statistical analysis combines data from 15 independent experimental runs, providing ≈ 130000 data points (≈ 200 data points for each 10 ms time step). The coefficients of restitution of our particles are not known. A certain estimate of the normal coefficient of restitution was obtained from collisions with a container wall [24], the total loss of kinetic energy per collision yields ε ≈ 0.54. This is not necessarily a good estimate for particle-particle collisions, which are very difficult to evaluate.	2018-06-22T14:36:50.833Z	2018-01-01T00:00:00.000	{ "year": 2018, "sha1": "b248b6a040062361f5a730885bab3bb0bf8e8c5e", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1706.07472", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "2136772c5faf46527ce951d0cfebbaa2e6c938e6", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
16233416	pes2o/s2orc	v3-fos-license	Regional assessment of articular cartilage gene expression and small proteoglycan metabolism in an animal model of osteoarthritis Osteoarthritis (OA), the commonest form of arthritis and a major cause of morbidity, is characterized by progressive degeneration of the articular cartilage. Along with increased production and activation of degradative enzymes, altered synthesis of cartilage matrix molecules and growth factors by resident chondrocytes is believed to play a central role in this pathological process. We used an ovine meniscectomy model of OA to evaluate changes in chondrocyte expression of types I, II and III collagen; aggrecan; the small leucine-rich proteoglycans (SLRPs) biglycan, decorin, lumican and fibromodulin; transforming growth factor-β; and connective tissue growth factor. Changes were evaluated separately in the medial and lateral tibial plateaux, and were confirmed for selected molecules using immunohistochemistry and Western blotting. Significant changes in mRNA levels were confined to the lateral compartment, where active cartilage degeneration was observed. In this region there was significant upregulation in expession of types I, II and III collagen, aggrecan, biglycan and lumican, concomitant with downregulation of decorin and connective tissue growth factor. The increases in type I and III collagen mRNA were accompanied by increased immunostaining for these proteins in cartilage. The upregulated lumican expression in degenerative cartilage was associated with increased lumican core protein deficient in keratan sulphate side-chains. Furthermore, there was evidence of significant fragmentation of SLRPs in both normal and arthritic tissue, with specific catabolites of biglycan and fibromodulin identified only in the cartilage from meniscectomized joints. This study highlights the focal nature of the degenerative changes that occur in OA cartilage and suggests that altered synthesis and proteolysis of SLRPs may play an important role in cartilage destruction in arthritis. R853 The ECM of cartilage also contains the small leucine-rich proteoglycans (SLRPs) biglycan, decorin, fibromodulin and lumican, which have diverse functions as modulators of tissue organization, cellular proliferation, adhesion and responses to growth factors and cytokines [2,3]. The SLRPs all bind to fibrillar type I and/or II collagens [4][5][6] and, in the case of decorin, to fibromodulin and lumican; these interactions modulate the rate and ultimate diameter of collagen fibrils formed in vitro [7][8][9]. Decorin, biglycan and fibromodulin can also form complexes with transforming growth factor (TGF)-β and modulate the action of this growth factor [10,11]. The physical presence of the SLRPs, in addition to the minor type IX and XI collagens, on the surface of type II collagen fibrils has been proposed to restrict sterically the access of collagenases to sites of cleavage on the collagen fibrils [12]. Complexes of matrilin-1 and decorin or biglycan have also been reported to connect type VI collagen to aggrecan and type II collagen, further stabilizing the cartilage ECM [13]. It is evident that there is a complex interplay between the collagenous and proteoglycan components of the cartilage ECM that produces a biocomposite material with unique mechanical properties. Disruption of the normal balance of ECM components through altered synthesis or degradation will have important ramifications for the loadbearing capacity of cartilage. Chondrocytes, the highly differentiated cells of cartilage, are responsible for maintaining a homeostatic balance between production and degradation of cartilage ECM [14,15]. The metabolic status of the chondrocyte is central to our understanding of the initiation and progression of osteoarthrits (OA) [16]. An initial anabolic response of chondrocytes in OA includes an upregulation of mRNA levels for the major structural components type II collagen and aggrecan, with an associated elevation in synthesis [17,18]. Degradation of the ECM is also elevated in these early stages in OA. Eventually, the biosynthetic machinery of the chondrocyte is unable to keep up with the anabolic demands and a net depletion of ECM occurs during the later stages of OA. Loss of key functional components combined with a disrupted architecture result in compromised tissue function, cell death and, eventually, cartilage loss down to subchondral bone. Changes in SLRP metabolism in human OA are relatively poorly characterized, with both increased synthesis and degradation of individual molecules reported in arthritic human cartilage [19,20]. Their function within the collagen network means that changes in their tissue content may significantly alter the biomechanical integrity of cartilage. However, because SLRPs are also regulators of growth factor activity, changes in their synthesis and degradation may have significant effects on chondrocyte metabolism. It is unclear whether the changes in SLRP metabolism are restricted to the cartilage undergoing OA degeneration or are more generalized within arthritic joints. An understanding of the changes that occur with the onset and progression of cartilage degeneration in OA may provide important insights into potential regulatory steps in this process. Animal models of OA have permitted longitudinal evaluation of spatial and temporal changes in joint tissues that occur during the development of joint disease. Total or partial removal of knee joint meniscus in humans commonly results in degeneration of articular cartilage, leading to osteoarthritic changes [21]. In sheep, lateral meniscectomy has been shown to reliably reproduce biochemical, biomechanical and histopathological alterations typical of OA [22,23]. In the present study we used this established model of OA to study the changes in expression of key structural molecules (aggrecan and type II collagen), the collagen-associated SLRPs (biglycan, decorin, lumican and fibromodulin), TGF-β 1 and its associated downstream signaling molecule connective tissue growth factor (CTGF), and markers of altered chondrocyte phenotypetypes I and III collagen. The expression levels were compared with protein levels in cartilage extracts or by immunohistochemistry in tissues with various histopathological grades of OA in the medial and lateral joint compartments. Animal model Twelve 7-year-old female pure-bred Merino sheep were used in the present study. Six of the sheep underwent open lateral meniscectomy of both stifle joints, as previously described [24], whereas the remaining six served as nonoperated controls. Following recovery from surgery, the animals were maintained in an open paddock for 6 months before sacrifice. The protocol used for the present study was approved by the animal ethics committee of Murdoch University, Western Australia (AEC 832R/00). Tissue preparation Full depth articular cartilage from the medial tibial plateau (MTP) and lateral tibial plateau (LTP) was sampled from either the right or left stifle (knee) joint, randomly selected. Care was taken not to sample tissue from the joint margins or osteophytes. Tissue samples were snap frozen in liquid nitrogen before storage at -80°C until they were required. The tibial plateaux from the contralateral joints were isolated by a horizontal cut through the tibia below the epiphyseal growth plate using a band saw. Full thickness coronal osteochondral slabs (5 mm) were subsequently prepared through the mid weightbearing region of the tibial plateau. Histology The coronal tibial osteochondral slices were fixed in 10% (vol/ vol) neutral buffered formalin for 48 hours then decalcified in 10% formic acid (vol/vol)/5% formalin (vol/vol) for 5 days. The specimens were then dehydrated in graded alcohols and double-embedded in celloidin-paraffin blocks. Tissue sections (4 µm) were cut using a rotary microtome and attached to microscope slides. They were then deparaffinized in xylene and washed in graded alcohols to 70% (vol/vol) ethanol and then stained for 10 min with 0.04% (weight/vol) toluidine blue in 0.1 mol/l sodium acetate buffer (pH 4.0) to visualize the tissue proteoglycans. This was followed by 2 min counter-staining in an aqueous 0.1% (weight/vol) Food Drug and Cosmetic Green Nos. 3 stain. The slides were subsequently evaluated by bright field microscopy using a Leica MPS-60 (Leica Microsystems, Gladesville, New South Wales, Australia) photomicroscope system by two independent observers using a modified Mankin scoring scheme, previously developed in our laboratory for this ovine model [22]. In each compartment the worst score evident across the width of the tibial plateau was used to calculate the mean score for MTP and LTP of control and meniscectomized joints (n = 6 for each group). Immunohistochemistry Immunostaining was performed using monoclonal antibodies against type I collagen (ICN Biomedicals, Aurora, USA; code no. 63170; clone no. I-8H5) and type II collagen (ICN Biomedicals, North Ryde, New South Wales, Australia; code no. 63171; clone no. II-4CII), and a polyclonal antibody against type III collagen (Cedarlane, Hornby, Ontario, Canada; code no. CL50321AP). Endogenous peroxidase activity was initially blocked by incubating the tissue sections in 3% (vol/vol) H 2 O 2 for 5 min and the sections were rinsed in TBS-Tween. For type I and II collagen localizations, the sections were predigested with proteinase K (Dako, Glostrup, Denmark; code no. S3020) for 6 min at room temperature, followed by bovine testicular hyaluronidase (Sigma, St Louis, MO, USA; code no. H-3506) 1000 U/ml for 1 hour at 37°C in phosphate buffer (pH 5.0). The type III collagen localizations were predigested with hyaluronidase alone. The sections were then incubated in 10% (vol/vol) swine serum for 10 min at room temperature to block any nonspecific binding. Incubations with the primary antibodies were performed overnight at 4°C with type I (5 µg/ml), type II (10 µg/ml) and type III (1:500 dilution) collagens. Detection of primary antibody was undertaken using a 20 min incubation with a cocktail of biotinylated anti-rabbit and anti-mouse immunoglobulin secondary antibodies (Dako; code no. K1015), followed by a 20 min incubation with streptavidin-conjugated horseradish peroxidase (Dako; code no. K0690). Staining was undertaken using NovaRED substrate (Vector, Burlingame, CA, USA; code no. SK-4800) for 15 min, which gives a red-brown end product. Sections were counter-stained in Mayer's haematoxylin for 1 min, washed in H 2 O, dehydrated in ethanol, cleared in xylene and mounted. Negative control sections were prepared using irrelevant isotype matched primary antibodies (Dako; code no. X931 or X0936) in place of authentic primary antibody. RNA extraction Approximately 100 mg of frozen cartilage samples was fragmented in a Mikro-Dismembrator (Braun Biotech International, Melsungen, Germany), 1 ml of TRIzol Reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) was added, and the mixture was allowed to defrost to room temperature. Total RNA was isolated using the RNeasy Mini Kit from Qiagen (Valencia, CA, USA). Chloroform (300 µl) was subsequently added to the samples and the tubes vortexed vigorously before centrifugation to pellet the tissue residue. The clear supernatant solution (aqueous phase) was recovered and mixed by inversion with an equal volume of 70% ethanol, and then loaded onto spin columns. Following several washing steps and an on-column DNase digestion (Qiagen, Hilden, Germany), RNA was eluted from the column with 32 µl of RNAse free distilled H 2 O. Total RNA was quantified using a flourimeter (Perkin Elmer, Beaconsfield, UK) using SYBR ® Green II colour reagent (Cambrex Bio Science, Rockland, ME, USA), and each sample was assessed for purity to confirm the absence of detectable DNA. Semiquantitative RT-PCR RT reactions were undertaken with 1 µg total RNA using the Omniscript RT kit from Qiagen (Germany). Using specific primer sets (Sigma Genosys, Castle Hill, New South Wales, Australia; Table 1), aliquots of cDNA were amplified by PCR, with initial denaturation at 94°C for 5 min, followed by cycles of 30 s of denaturation at 94°C, 30 s annealing at variable primer specific temperatures (Table 1), 30 s for extension at 72°C, and a further 7 min extension at 72°C on completion of the cycles. Reactions generated single PCR products that were identified by sequencing (SUPAMAC, Sydney, Australia) and specificity confirmed by BLAST searches. Cycle optimization was performed for each primer set before PCR, and for all reported experiments amplification levels were compared in the linear range of the PCR reaction. All samples underwent RT and cDNA amplification at the same time to avoid potential variations in experimental conditions. The amplified products were electrophoresed on 2% (weight/ vol) agarose gels, stained with ethidium bromide, imaged using a Fujifilm FLA-3000 fluorescent image analyzer and integrated densities calculated using One-Dscan, 1-D gel analysis software (Scanalytics, Fairfax, VA, USA). Sample loadings were normalized to the housekeeping gene GAPDH (glyceraldehyde-3-phosphate dehydrogenase) to permit semiquantitative comparisons in mRNA levels, as previously described [25,26]. Cartilage extraction, SDS-PAGE and Western blotting of the small leucine-rich proteoglycans Pooled cartilage samples from all meniscectomized and nonoperated control LTPs were finely diced and extracted with 10 volumes of 4 mol/l GuCl and 50 mmol/l Tris HCl (pH 7.2) in the presence of proteinase inhibitors at 4°C with end over end stirring for 48 hours before dialysis of the extract against ultrapure water, as described previously [27]. Insufficient cartilage was available from MTPs for extraction and Western blot analyses. Dialysed extracts corresponding to equal dry weights of tissue were predigested with either chondroitinase ABC (Seikagaku) 0.1 U/ml alone or in combination with keratanase II (Seikagaku) 0.01 U/ml and endo-β-galactosidase (Seikagaku, Tokyo, Japan) 0.01 U/ml in 0.1 mol/l Tris/0.1 mol/ l sodium acetate (pH 7.0) overnight at 37°C before electrophoresis. Electrophoresis was conducted under reducing conditions in 10% NuPAGE Bis-Tris resolving gels (Invitrogen), using MOPS SDS running buffer at 125 V constant voltage for 1 hour. The gels were then electroblotted to nitrocellulose membranes in NuPAGE transfer buffer with 20% (vol/vol) methanol at 200 mA for 2 hours and blocked overnight in 5% (weight/vol) BSA in 50 mmol/l Tris-HCl (pH 7.2) and 0.15 mol/l NaCl 0.02% (weight/vol) NaN 3 (TBSazide). The blots were probed overnight with affinity purified polyclonal antibodies directed against the carboxyl-terminus of decorin, biglycan, fibromodulin and lumican (0.3-1 µg/ml) [12] followed by washing in TBS-azide and detection using alkaline phosphatase conjugated anti-rabbit secondary antibodies and the nitro blue tetrazolium/4-bromo-1-chloro-indolyl phosphate substrate system (BioRad, Hercules, CA, USA). A sample of human OA cartilage harvested from the tibial plateau at the time of joint replacement surgery also underwent identical processing as a positive control. Statistical analysis All RT-PCR data were normalized to the housekeeping gene GAPDH (glyceraldehyde-3-phosphate dehydrogenase) to facilitate equal loading of gels for quantitative comparisons of amplified PCR products. Comparison of parametric data from the nonoperated and meniscectomized sample groups were undertaken using the unpaired Student's t-test with Benjamini-Hochberg correction [28] for multiple comparisons. Comparisons of nonparametric data from the modified Mankin histological scoring of the stained tissue sections were assessed using the Mann-Whitney U-test. Histology Lateral meniscectomy resulted in macroscopic joint changes characteristic of the early and middle phases of OA with cartilage fibrillation and erosion, in addition to formation of marginal osteophytes, particularly in the lateral compartment ( Fig. 1f; arrowheads). The histopathological lesions varied between animals, between medial and lateral joint compartments, and across the width of the tibial plateaux. A significant loss of proteoglycan was evident in the superficial cartilage of both the LTP (Fig. 1d, e) and, to a lesser extent, the MTP of the meniscectomized joints (Fig. 1b) compared with nonoperated controls (Fig. 1a, c). Chondrocyte cloning was also a prominent feature in the LTP specimens after meniscectomy ( Fig. 1d; asterisk), which is in keeping with the validity of this model's representation of human OA. The most severe lesions were Shown are the details of the primers used for RT-PCR, including annealing temperatures, size of the amplified products, forward (F) and reverse (R) sequences, and primer source. CTCG, connective tissue growth factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; TGF, transforming growth factor. confined to the weight-bearing region of the LTP, with significant proteoglycan loss and surface fibrillation (Fig. 1e, f). Histological grading of the meniscectomized and nonoperated control cartilage specimens confirmed and quantitated the histological observations. In control sheep the modified Mankin score (mean ± standard deviation) was significantly higher in the MTP specimens than in the LTP ones (9.3 ± 1.9 versus 3.1 ± 1.1; P < 0.01). Following meniscectomy there was a slight although not statistically significant change in the modified Mankin score for the MTP specimens (10.7 ± 3.3). The same could not be said of the LTP specimens, in which meniscectomy resulted in a significant increase from 3.1 ± 1.1 to 23.3 ± 1.8 (P < 0.01). Immunolocalization of types I, II and III collagens An increase in type I collagen matrix immunostaining was evident following meniscectomy in the most superficial cartilage of the LTP specimens (Fig. 2d) and, to a lesser extent, in the MTP specimens (Fig. 2b), corresponding to areas of degenerative change. In nonoperated control sections (Fig. 2a, c), type I collagen was restricted to the uppermost surface lamina, as reported previously [29]. Type III collagen, which is typically seen pericellularly in normal cartilage [30], also exhibited increased matrix staining after meniscectomy (Fig. 2j, l) compared with nonoperated control (Fig. 2i, k). Type II collagen was immunolocalized in the matrix throughout the depth of the cartilage in both MTP and LTP, and there was a generalized decrease in staining following meniscectomy (Fig. 2e-h). As expected [31,32], types I and III collagens were also prominently immunolocalized in the marginal osteophytic fibrocartilaginous regions in the meniscectomized joints (data not shown). RT-PCR It was not possible to undertake all procedures with some of the cartilage samples that did not yield at least 1 µg total RNA. This resulted in four samples being excluded, all from MTP cartilage (one from the meniscectomy group and three from the nonoperated control group). Statistical comparisons of mRNA levels following meniscectomy as a percentage of control values were undertaken separately for LTP and MTP cartilages and are presented graphically in Fig. 3. Following lateral meniscectomy, mRNA levels in LTP cartilage were found to be upregulated for the following molecules: aggrecan (1.5 fold; P < 0.01), type I collagen (11.7-fold; P < 0.01), type II collagen (3.9-fold; P < 0.01), type III collagen (2.3-fold; P < 0.05), biglycan (1.8-fold; P < 0.01) and lumican (14.6-fold; P < 0.01). In the same region there were downregulations of decorin (1.6-fold; P < 0.01) and CTGF (2.1-fold; P < 0.05), and unchanged expression of fibromodulin and TGF-β. In the MTP cartilage samples, none of the changes in mRNA levels following meniscectomy relative to nonoperated controls were statistically significant. Western blotting of the small leucine-rich proteoglycans Western blot analysis of extracts of an equivalent dry weight of pooled LTP cartilage from control and meniscectomized joints and OA human cartilage are shown in Fig. 4. There was little difference in total staining intensity between nonoperated and meniscectomized cartilage for the 45 kDa intact core protein of decorin that was also evident in human OA cartilage. Additional fragmented forms of decorin core protein (32 and 20 kDa) were evident in the cartilage extracts from both the control and meniscectomy specimens, whereas a 40 kDa fragment was identified only in meniscectomized cartilage extracts ( Fig. 4; asterisk). Blotting for biglycan identified intact core protein (43 kDa) and a number of fragments (39,32,28 and 26 kDa) in all of the specimens. There was an increase in staining intensity for all biglycan core protein species in menis- cectomized cartilage. The predominant fibromodulin core protein species identified in all specimens was about 55 kDa in size, with a slight increase in staining following meniscectomy. This 55 kDa fibromodulin band is consistent with full-length core protein [12]. A 28 kDa fibromodulin fragment was detected only in the extract from meniscectomized joints ( Fig. 4; asterisk). Lumican electrophoresed as two predominant species, a 60-64 kDa band with similar staining intensities evident in control and meniscectomy extracts. A smaller, approximately 50 kDa band, which was the predominant species in the human OA sample, exhibited greater staining intensity after meniscectomy compared with cartilage from nonoperated joints. Removal of KS side-chains with keratanase II/endo-β-galactosidase treatment resulted in all of the lumican migrating at 50 kDa, suggesting that the 60-64 kDa band represented KS substituted lumican. Discussion Our laboratory previously reported biochemical, biomechanical and histological changes that occur in the articular cartilage in the ovine lateral meniscectomy model of OA [22,23,33]. The present study extends these earlier investigations by examining the expression of a number of important extracellular matrix components at the mRNA level. One of the difficulties we encountered was relatively low average RNA yields (0.85-9.13 µg per 100 mg), which resulted in exclusion of some MTP samples. Studies utilizing other animal models of OA have reported RNA yields from 2.5 to 21 µg/100 mg of normal cartilage [34,35], but the animals used in those studies (rabbit and canine) were of a much younger age than ours. Studies using aged human cartilage report much lower average yields, ranging from 0.669 to 0.839 µg/100 mg of OA and 'normal' cartilage [36]. We attributed the low RNA yield in our study to our use of an aged population of sheep, although other factors such as species differences, RNA degradation and technical factors cannot be excluded. Although we were able to analyze medial and lateral tibial cartilage separately, the low yields of RNA from the older sheep precluded further topographical separation. Future studies using younger animals may permit analysis of affected and unaffected cartilage within one joint area. Although morphological and histological changes in cartilage were most notable in the lateral compartment, changes in the medial femoro-tibial joint were nevertheless still evident but of a markedly lesser magnitude, as previously reported [23,37]. In the present study the MTP cartilage in control joints had significantly worse histopathological scores than did LTP from the same joints, which is consistent with age-related change in the more heavily loaded compartment of these old animals. The histopathology scores did not increase significantly in the medial compartment following meniscectomy, and this was consistent with the lack of change in mRNA levels. Our inability to detect differences in mRNA expression in the medial compartment might have resulted from the small number of samples evaluated. However, the standard deviation of the MTP samples was similar to that of the LTP, suggesting that the lower number of MTPs studied did not contribute to the lack of statistical significance. Changes in mRNA levels for a number of molecules were significant in the lateral compartment following meniscectomy. Although our findings are limited to a single time point following induction of OA, restriction of significant alterations in gene expression to the LTP indicates that the changes observed were likely associated with active degradation of cartilage primarily due to altered biomechanical forces rather than humoral factors. In the present study the changes observed in the expression of aggrecan and type II collagen probably reflect an anabolic response by the chondrocytes to the altered mechanical stresses imposed by this surgical procedure, as well as early OA degeneration. The increase in expression is consistent with an attempted 'repair' response in early OA, as described in other animal models [34,35,38]. Levels of mRNA for a particular molecule may not reflect protein synthesis or its accumulation in tissue, with post-transcriptional regulation and post-translational processing playing significant roles. Indeed, we previously demonstrated increased degradation of newly synthesized aggrecan in cartilage after lateral meniscectomy in sheep [24]. Furthermore, the changes in mRNA levels observed in the present study were representative of the entire MTPs or LTPs and therefore probably included cartilage from areas with different stages of OA. In addition to the increase in mRNA for the major cartilage matrix components aggrecan and type II collagen, significant increases in expression and protein levels of types I and III collagen were observed following meniscectomy. Type III collagen is present pericellularly in small amounts in normal articular cartilage [16,30], and type I collagen is is evident in the most superficial layer [29]. Contrary to early reports [39], evidence now suggests that both types I and III collagens are significantly increased in OA cartilage, both at the expression and protein levels [40,41]. It has been suggested that a major phenotypic shift occurs in OA toward a de-differentiated chondrocyte [40]. Interestingly, in the present study we observed increased amounts of types I and III collagens by immunohistology in both compartments following meniscectomy, despite increased mRNA levels only being evident in the lateral compartment. A probable explanation was that the increased types I and III collagens observed with immunohistochemistry represented the cumulative changes throughout the course of the disease process while expression levels reflected chondrocyte metabolism at a specific point in time (i.e. 6 months following meniscectomy). Changes in collagen subtypes in pathological cartilage may not only influence the biomechanical integrity of the tissue but may sequester and modulate the actions of cytokines, with types I and III collagen shown to bind oncostatin M specifically [42]. Selective modulation of SLRP mRNA levels in OA cartilage was observed in the present study, with increased biglycan and lumican, decreased decorin, and little or no change in fibromodulin. Additionally, we have shown for the first time that these changes in SLRP expression are confined to the cartilage in the compartment undergoing active OA degeneration. The differential regulation contrasts with the reported increase in expression of all four SLRPs in late-stage human OA in one study [19], but it is consistent with another study [43] that reported no change in decorin but increased biglycan message in late stage OA. In the canine anterior cruciate ligament transection model, increased cartilage mRNA for biglycan, decorin and fibromodulin have been described [38,44]. The reported differences in mRNA expression may relate to variable stages of disease, methods of quantitation and species evaluated. The SLRPs have been shown to influence cartilage metabolism indirectly via actions on growth factors such as TGF-β, which they inactivate through sequestration and thereby potentially mitigate its effects in OA [11,45]. Although we found no change in the expression of TGF-β following meniscectomy, there was a significant decrease in mRNA levels of CTGF. We speculate that sequestration of TGF-β by the SLRPs may have accounted for the decrease in CTGF expression. Our results contrast with human cartilage, in which an increase in CTGF in OA was recently reported [46], and this could be associated with species differences or the stage of disease. CTGF, a secretory protein involved in fibrotic response mechanisms in tissues, is an important downstream effector of TGF-β [47] and is thought to be involved in promoting the proliferation and/or differentiation of chondrocytes [48][49][50][51]. Further investigation of the specific relationships between growth factors, collagens and the SLRPs in normal and diseased cartilage is warranted. Significant proteolysis of the SLRPs was evident in the present study. SLRP degradation was previously reported in both human OA [20] and spontaneous canine OA [52], but not in a canine cruciate ligament transection model of OA [52]. The catabolites that were identified in meniscectomized cartilage in the present study were also generally evident in normal cartilage, indicating similar proteolytic processes in health and disease. However, in the case of decorin and fibromodulin, fragments unique to the meniscectomized cartilage were identified, suggesting the presence of disease-specific proteolytic processes. In this regard, a specific proteolytic fragment of fibromodulin was recently identified from interleukin-1 stimulated but not normal cartilage [53]. The cleavage site(s) and proteinase(s) responsible for extracellular SLRP breakdown in arthritic cartilage have yet to be identified and are the subject of further investigation. A particularly novel finding in the present study was the increased lumican core protein present in degenerative cartilage following meniscectomy, which is consistent with the significant increase observed in mRNA levels. Furthermore, the increased lumican observed by Western blotting was present in a non-KS substituted form. Limited studies [19,54] have suggested that lumican primarily exists lacking KS in adult cartilage, but cultured chondrocytes have been observed to produce a KS-substituted form that appeared to be the default synthesis preference [55]. The catabolic cytokine interleukin-1β, which may be present in OA joints, stimulates secretion of lumican deficient in KS [55]. It has been shown that OA chondrocytes synthesize SLRPs that are differently glycosylated, and that nonglycosylated biglycan and decorin are more abundant in OA cartilage [20]. Changes in glycosylation of the SLRPs, whether by altered synthesis or subsequent degradation, are likely to influence the functional properties of these molecules in cartilage. Conclusion We showed that degradation of cartilage in OA is associated with significant focal changes in expression and content of matrix proteins. Accelerated proteolysis of aggrecan and type II collagen overwhelms the increase in expression of these major structural proteins. Furthermore, there is a shift in chondrocyte phenotype, with increased synthesis of collagens types I and III and a change in the relative levels of the fibrilassociated SLRPs. In particular there is decrease in synthesis of decorin and an increase in biglycan and lumican, with the latter lacking KS substitution. It seems likely that the altered pattern of SLRP synthesis, which is localized to the diseased joint compartment, along with an increase in SLRP proteolysis, modifies the biomechanical properties of the matrix and contributes to cartilage breakdown. Changes in SLRP levels could also significantly modulate the action of potential anabolic factors such as TGF-β and its downstream effector CTGF, possibly adding to disease development. An understanding of the relationship between SLRP metabolism and progressive cartilage breakdown in OA may provide both novel diagnostic markers of disease and therapeutic targets for the treatment of this disorder.	2017-06-25T11:12:50.478Z	2005-05-12T00:00:00.000	{ "year": 2005, "sha1": "949174373b102318be99bf3cb171d12577661048", "oa_license": "CCBY", "oa_url": "https://arthritis-research.biomedcentral.com/track/pdf/10.1186/ar1756", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "949174373b102318be99bf3cb171d12577661048", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Chemistry", "Medicine" ] }
235828080	pes2o/s2orc	v3-fos-license	Common Fixed Point Results on Generalized Weak Compatible Mapping in Quasi-Partial b-Metric Space In the early years of 20 century, the French mathematician Fréchet [1] commenced the concept of metric space, and due to its consequences and practicable implementations, the idea has been enlarged, upgraded, and generalized in different directions. In 1922, Banach [2] introduced the very important Banach contraction principle which holds a remarkable position in the field on nonlinear analysis. One such generalization was established by Künzi et al. [3] known as quasi-partial metric space by Karapinar et al. [4, 5]. In 1993, Czerwik [6] introduced the concept of b-metric space. Later, Gupta and Gautam [7, 8] generalized quasi-partial metric space to quasi-partial b-metric space and proved some fixed point results for such spaces. Several authors [9–18] have already proved the fixed point theorem in metric space, partial metric space [19], quasi-partial metric space, quasi-partial b-metric space [7], and many different spaces. After these classical results, some researchers [20–25] introduced the distinctive concepts and used fixed point theorems to demonstrate the uniqueness of a solution of the equations in different metric spaces such as multivalued contractive type mappings, Reich–Rus–Cirić and Hardy–Rogers contraction mappings, and Chatterjea and cyclic Chatterjea contraction. In this paper, we have introduced the generalized condition (B) in quasi-partial b-metric space to obtain coincidence and common fixed points. Moreover, some examples are given to exemplify the concept followed up with pictographic grid. Introduction In the early years of 20 th century, the French mathematician Fréchet [1] commenced the concept of metric space, and due to its consequences and practicable implementations, the idea has been enlarged, upgraded, and generalized in different directions. In 1922, Banach [2] introduced the very important Banach contraction principle which holds a remarkable position in the field on nonlinear analysis. One such generalization was established by Künzi et al. [3] known as quasi-partial metric space by Karapinar et al. [4,5]. In 1993, Czerwik [6] introduced the concept of b-metric space. Later, Gupta and Gautam [7,8] generalized quasi-partial metric space to quasi-partial b-metric space and proved some fixed point results for such spaces. Several authors [9][10][11][12][13][14][15][16][17][18] have already proved the fixed point theorem in metric space, partial metric space [19], quasi-partial metric space, quasi-partial b-metric space [7], and many different spaces. After these classical results, some researchers [20][21][22][23][24][25] introduced the distinctive concepts and used fixed point theorems to demonstrate the uniqueness of a solution of the equations in different metric spaces such as multivalued contractive type mappings, Reich-Rus-Cirić and Hardy-Rogers contraction mappings, and Chatterjea and cyclic Chatterjea contraction. In this paper, we have introduced the generalized condition (B) in quasi-partial b-metric space to obtain coincidence and common fixed points. Moreover, some examples are given to exemplify the concept followed up with pictographic grid. Preliminaries Let us recall some definition. Definition 1 (see [19]). A partial metric space on a nonempty set X is a function M: X × X ⟶ R + satisfying Definition 2 (see [4]). A quasi-partial metric on a nonempty set X is a function q: X × X ⟶R + satisfying Definition 3 (see [20]). A quasi-partial b-metric on a nonempty set X is a function qp b : X × X ⟶ R + such that for some real number ρ ≥ 1 for all τ, υ, Υ ∈ X. e infimum over all reals ρ ≥ 1 satisfying condition (30) is called the coefficient of (X, qp b ) and represented by R(X, qp b ). Lemma 1 (see [6]). Let (X, qp b ) be a quasi-partial b-metric space. en the following hold: Definition 4 (see [6]). Let (X, qp b ) be a quasi-partial b-metric. en (1) A sequence τ n ⊂ X converges to τ ∈ X if and only if (1) (3) e quasi-partial b-metric space (X, qp b ) is said to be complete if every Cauchy sequence τ n ⊂ X converges with respect to x qp b to a point τ ∈ X such that (4) A mapping f: X ⟶ X is said to be continuous at Lemma 2 (see [6]). Let (X, qp b ) be a quasi-partial b-metric space and (X, Definition 5 (see [26] Following Babu et al. [26], Abbas et al. [27] and Abbas and Illic [28] extended the concept of condition (B) to a pair of mappings. Abbas et al. [27] called it generalized condition (B), and Abbas and Illic [28] called it generalized almost A-contraction. Definition 6 (see [27]). Let P and Q be two self-mappings on a metric space (X, d). e mapping Q satisfies generalized condition (B) associated with P if there exist δ ∈ (0, 1) and Clearly condition (B) implies generalized condition (B). e mapping R satisfies generalized condition (B) associated with P (R is a generalized almost P contraction) if there exist δ ∈ (0, 1), ρ ≥ 1, and M ≥ 0 such that for all τ, υ ∈ X, we have Definition 10. Let P, Q, R, S be four self-mappings on a quasi-partial b-metric space (X, qp b ). e pair of mapping (P, R) satisfies generalized condition (B) associated with (Q, S) ((P, R) is generalized almost Theorem 1. Let P, Q, R, S be four self-mappings on quasipartial b-metric space (X, qp b ) and if we take the mappings in pair as (P, R) associated with (Q, S) for all τ, υ ∈ X, δ ∈ (0, 1), and M ≥ 0, ρ ≥ 1 and then the pairs (P, R) and (Q, S) have a coincidence point. Also P, Q, R, S have a unique common fixed point, providing that pairs (P, R) and (Q, S) are weakly compatible. Proof. Let τ * ∈ X. Since RX ⊂ QX there exists τ 0 ∈ X such that υ 0 � Qτ 0 � Rτ * . Suppose there exists a point υ 1 ∈ Sτ 0 corresponding to the point υ 0 . Also since SX ⊂ PX there exist τ 1 ∈ X such that υ 1 � Pτ 1 � Sτ 0 . Going this way we get a sequence υ n ∈ X as is condition gives 4 cases. Also, which implies Let Also, which implies, Let μ 2 � ((δ + M)/ρ), ((δ + 2M)/ρ) < 1 then μ 2 < 1. Also, which implies Let Also, which implies Let Using mathematical induction, which tends to 0 as m tends to ∞ So, υ m and its subsequence is convergent Let PX be closed. erefore, τ ∈ PX, that is, there exists Υ ∈ X such that τ � PΥ, and we need to show τ � RΥ By definition, which is a contradiction. Hence, 4 Journal of Mathematics So, PΥ � RΥ, that is, P and R have a coincidence point. Similarly, Q and S have a coincidence point. If we also assume QX is closed, then (P, R) and (Q, S) have a coincidence point. Since (P, R) and (Q, S) are weakly compatible, we can prove there exists a common fixed point for P, Q, R, S by contradiction. Let P, Q, R, S be selfmappings on quasi-partial b-metric defined by Here, e point 0 is a coincidence point of these mapping. Furthermore, PR0 � RP0 � 0 and SQ0 � QS0 � 0, that is, the two pairs (P, R) and (Q, S) are weakly compatible. Dominance of right-hand side of equation (27) is easily visually checked in Figure 1. us the inequality required in Definition 10 holds for τ, υ ∈ [0, 2]. Dominance of right-hand side of equation (30) is easily visually checked in Figure 4. us the inequality required in Definition 10 holds for τ, υ ∈ [2,4]. If P � Q and R � S, we get a corollary. Corollary 1. Let P and S be self-mappings on quasi-partial b-metric space (X, qp b ). If for all τ, υ ∈ X, P satisfies the following conditions: then P and S have a coincidence point. Also P and S have a common fixed point if (P, S) are weakly compatible. Proof. Taking P � Q and R � S in eorem 1, the above result can be obtained. Proof. is can be done following the same steps as the proof of eorem 1. Dominance of the right-hand side of equation (34) is easily visually checked in Figure 5. us the inequality required in theorem holds for τ, υ ∈ [0, 2]. Dominance of the right-hand side of equation (35) is easily visually checked in Figure 6. us the inequality required in theorem holds for τ ∈ [0, 2], υ > 2. Journal of Mathematics Dominance of the right-hand side of equation (36) is easily visually checked in Figure 7. us the inequality required in theorem holds for τ > 2, υ ∈ [0, 2]. Dominance of the right-hand side of equation (37) is easily visually checked in Figure 8. us the inequality required in theorem holds for τ, υ > 2. If P � Q and R � S, we get a corollary. Proof. Taking P � Q and R � S in eorem 2, the above result can be obtained. Conclusion is paper expounds a new notion in quasi-partial b-metric space which is generalized condition (B) that helped to demonstrate coincidence and common fixed point for two weakly compatible pairs of self-mappings. e incentive behind using quasi-partial b-metric space is the fact that the distance from point x to point y may be different to that from y to x, and the self-distance of a point need not always be zero; also the distance between two points x and z is not equal to the sum of the two distances having a point y in between x and z. Furthermore, the results acquired are validated by explanatory examples. Data Availability No data were used to support this study.	2021-07-15T09:15:12.069Z	2021-06-11T00:00:00.000	{ "year": 2021, "sha1": "b0061d43707e1ace68290b677763e790a6196466", "oa_license": "CCBY", "oa_url": "https://downloads.hindawi.com/journals/jmath/2021/5526801.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "b0061d43707e1ace68290b677763e790a6196466", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [ "Mathematics" ] }
118688776	pes2o/s2orc	v3-fos-license	Superradiant scattering of dispersive fields Motivated by analogue models of classical and quantum field theory in curved spacetimes and their recent experimental realizations, we consider wave scattering processes of dispersive fields exhibiting two extra degrees of freedom. In particular, we investigate how standard superradiant scattering processes are affected by subluminal or superluminal modifications of the dispersion relation. We analyze simple 1-dimensional toy-models based on fourth-order corrections to the standard second order wave equation and show that low-frequency waves impinging on generic scattering potentials can be amplified during the process. In specific cases, by assuming a simple step potential, we determine quantitatively the deviations in the amplification spectrum that arise due to dispersion, and demonstrate that the amplification can be further enhanced due to the presence of extra degrees of freedom. We also consider dispersive scattering processes in which the medium where the scattering takes place is moving with respect to the observer and show that superradiance can also be manifest in such situations. Introduction and motivation Experimental realizations of analogue black holes [1,2] and their associated effects have drawn a lot of attention in the past few years [3,4,5,6,7,8,9]. Probably the most discussed results have been the first observation of the classical analogue of Hawking radiation in an open channel flow [9] and the still controversial observation of radiation in ultrashort laser pulse filaments [8,10,11,12,13]. Even though it remains an open question whether or not real black holes emit Hawking radiation, calculations involving analogue black holes suggest that the emission process is most probably unaffected by transplanckian effects that could, in principle, alter or even exclude the radiation process [14,15,16]. Superradiance [17,18,19] is another typical phenomenon of black hole physics [20,21] which is also manifest in analogue models of gravity [22]. In standard scattering processes, the ratio between the reflected and the incident particle number currents [23] (i.e. the reflection coefficient) is smaller than one. This is directly encoded by the fact that the amplitude of the reflected wave is usually smaller than the amplitude of the incident one in non-dispersive normalized scattering processes. However, in some special situations (e.g. wave scattering in a Kerr black hole spacetime), low-frequency incident waves can be amplified in the scattering process. This amplification effect, known as superradiance, was first discovered by Zel'dovich in the context of electromagnetic waves [24], and later shown to be a more general phenomenon in physics [18,19], in which classical as well as quantum field excitations can be amplified. The main goal of our work is to analyze the robustness of the amplification process for dispersive fields. The dispersive fields considered in this paper exhibit two extra scattering channels, such that one can study the overall robustness of the amplification process for multiple superradiant scattering. Additionally, in some simplified cases we also investigate the specific deviations in the amplification spectrum due to dispersion. The examples discussed in this paper are motivated by analogue models of gravity, where fields exhibiting sub or superluminal dispersion relations arise naturally. Superradiant systems Before starting our analysis of superradiance for dispersive fields, it is important to review the basic ingredients that make the phenomenon possible in some simple systems. For example, in a Kerr black hole, incident scalar field modes of frequency ω and azimuthal number m = 0 are known to be superradiant for sufficiently low frequencies. Although the Klein-Gordon equation is very complicated when expressed in standard Boyer-Lindquist coordinates, the radial part of the equation of motion for these modes can be written, after a change of variables, very simply as d 2 u dr * 2 + V ω,m (r * )u = 0, where u is related to the radial part of the separated scalar field and r * is a tortoise-like coordinate (which goes to −∞ at the event horizon and reduces to the standard radial coordinate r at spatial infinity). V ω,m (r * ) is the effective potential and possesses the following asymptotics, where Ω h is the angular velocity of the black hole at the event horizon. It can be shown that, for 0 < ω < mΩ h , incident modes are superradiantly scattered by the black hole [25,20]. The corresponding reflection coefficient is given by where ω−mΩ h ω \|T \| 2 is the transmission coefficient. Two factors are responsible for the occurrence of the effect [19]: first, the event horizon behaves as a one-way membrane that allows only ingoing transmitted waves (defined by the group velocity) near the black hole; second, because of the ergoregion of the rotating black hole, low-frequency ingoing waves are associated with an outgoing particle number current at the horizon (the current being proportional to ω − mΩ h ) [23]. This outgoing flux at the horizon is compensated by the superradiantly reflected modes in such a way that the total particle number current is conserved during the process. Another simple system exhibiting superradiance is a massless scalar field φ with electric charge e and minimally coupled to an electromagnetic potential A µ = (V (x), 0) in 1+1-dimensions [17]. The evolution of this system is determined by the Klein-Gordon equation, which, after separation of the temporal dependence (i.e. φ = f (x) exp(−iωt)), reduces to d 2 f dx 2 + (ω − eV (x)) 2 f = 0, (5) where ω is the frequency of the mode in question. If the electromagnetic potential V (x) has the following asymptotic behaviour, one can calculate the relation between the reflection and transmission coefficients for incident waves [17]: Therefore, low frequency (0 < ω < eΦ 0 ) ‡ right-moving modes originating from −∞ interact with the scattering potential, resulting in transmitted right-moving modes ‡ It is important to remark that the electric potential is defined only up to an arbitrary constant C. Therefore, if we redefine V (x) → V (x)+C, the superradiant condition will be given by 0 < ω−C < eΦ 0 . Throughout this paper, we fix the potential by requiring that V (−∞) = C = 0. Consequently, the condition for superradiance, which is based on the smallness of ω − C, is reduced to a condition on the smallness of ω. and superradiantly reflected (\|R\| 2 > 1) left-moving modes. This is basically the Klein paradox [26] -see [17] for a detailed explanation of the relationship between superradiance, the Klein paradox and pair creation. Note that the important ingredients for superradiance are essentially the same as in the rotating black hole case: firstly, there is no left-moving mode at +∞ (this is similar to the boundary condition at the event horizon of a black hole, where no outgoing solutions are allowed); secondly, the form of the electromagnetic potential allows for low-frequency right-moving modes to be associated with left-moving particle number currents at +∞. Modified dispersion relations In order to study how a modified dispersion relation affects superradiance, we shall first generalize the simple, non-dispersive, 1+1-dimensional model above [17] by including fourth-order terms in (4). Inspired by the quartic dispersion relation Ω 2 = k 2 ± k 4 /Λ 2 , where Λ is a dispersive momentum scale and the ± notates super and subluminal dispersion respectively, we propose the following generalization of (4), which can be obtained from the following action for the complex scalar field φ, After separating the temporal dependence in (8), instead of (5), one obtains where f represents the field mode with frequency ω > 0 and prime denotes derivative with respect to x. We choose the effective frequency Ω(x) = ω − eV (x) to satisfy asymptotic relations similar to those in (2) and (6), where eΦ 0 is a positive constant. In the asymptotic regions, the solutions of (10) are simple exponentials, exp (ikx), whose wavenumbers k satisfy the dispersion relations below, In the subluminal case (lower sign), real solutions (ω, k) to the dispersion relation correspond to the intersections of a lemniscate (figure-eight) and a straight line, see In order to relate the asymptotic solutions at +∞ and −∞ without having to solve the differential equation for all values of x, one needs a conserved quantity analogous to the Wronskian for second order wave equations. Since our model is non-dissipative, we expect such a quantity to exist [27]. In fact, by calculating the x-component of the Noether current associated with the symmetry φ → e iα φ, it is possible to show that the expression where generalizes the notion of the Wronskian § to our dispersive model, i.e. dZ/dx = 0 for any solution f of (10). Furthermore, it is convenient to work with the scaled functional X whose action on a function f is defined by In particular, the action of X on a linear combination of 'on shell' plane waves (wavenumbers satisfying the dispersion relation) is simply We would like to emphasize the simplicity of the algebraic expression above, which only depends on the amplitudes, effective frequencies and group velocities of the various scattering channels participating in the scattering process. Throughout the paper, we shall refer to (17) as the particle number current since it generalizes the usual notion of particle number current associated with a complex Klein-Gordon field. Subluminal scattering Based on the general notion of superradiance described in the introduction, we will study the scattering process of incident waves originating from x → −∞ in the presence of a subluminal dispersion. In realistic scenarios, we do not expect the dispersion relation Ω 2 = k 2 −k 4 /Λ 2 to have a maximum/minimum value above/below which only imaginary solutions of the dispersion relation are possible. Therefore, in order to guarantee that at § For the standard 2nd order wave equation, W 1 = W 2 = 0 and the functional W 3 is the conserved quantity commonly referred to as Wronskian. For Klein-Gordon fields, the Wronskian can be interpreted as the particle number current. least one real mode is present in the dispersion relation, we assume that the dispersion parameter Λ is large compared to the electromagnetic interaction, i.e. we assume that eΦ 0 < Λ/2. In such situations, the behaviour of the solutions to (10) in the asymptotic limits is captured in FIG. 1, where the green dash-dotted (Ω = ω) and the blue dashed (Ω = ω − eΦ 0 ) lines represent the effective frequency when x → −∞ and x → +∞, respectively. Note that there exist four roots, corresponding to four propagating modes as one would expect from a fourth order differential equation. We also note that, when ω < eΦ 0 , the blue dashed line is located below the Ω = 0 axis and the roots in the region x → +∞ inherit a relative sign between their group and phase velocities with respect to the roots at x → +∞ (compare the intersections of the green dash-dotted line (x → −∞) and the blue dashed line (x → +∞) with the red solid curve). We label the four roots associated with a frequency 0 < ω < eΦ 0 as k a < k b < k c < k d when x → −∞ and as k A < k B < k C < k D when x → +∞ Since, in this setup, there is only one source of waves, located at x → −∞, and no incoming signal from x → +∞, we impose the boundary condition that the modes A and C are unpopulated in the scattering process. Furthermore, we choose the incoming mode at x → −∞ to be entirely composed of low-momentum c modes with no highmomentum a mode component. The corresponding solution of (10) in the asymptotic limits is where the wavenumber of the incident mode is given by k in = k c , the reflected modes are given by k r 1 = k b , k r 2 = k d , and the transmitted modes are k t 1 = k B , k t 2 = k D . In addition, the coefficients R 1 , R 2 , T 1 and T 2 can be related to reflection and transmission coefficients (see (20) below). In the asymptotic regions, it is possible to solve exactly the dispersion relation and find the following explicit expressions for the roots, Note that here, and everywhere else in this paper, we consider only stimulated scattering processes in which the ingoing high momentum channels are suppressed. This is not applicable to spontaneous scattering processes, where the quantum vacuum naturally supplies all ingoing, low and high, momentum modes. In order to compare the particle number current of the reflected waves with the incident and transmitted currents, we substitute (18) into the expression for the conserved generalized Wronskian (16) and find, after straightforward algebraic manipulation, the following relation between the coefficients R 1 , R 2 , T 1 and T 2 , This relation should be compared to the standard result for non-dispersive 1D scattering, \|R\| 2 = 1 − \|T \| 2 , and its generalization in the presence of an external potential, \|R\| 2 = 1 − (ω − eΦ 0 )\|T \| 2 /ω, see (7). As discussed above, in the non-dispersive case it is possible to achieve \|R\| > 1 for sufficiently low frequency scattering with 0 < ω < eΦ 0 . For fields with subluminal dispersion relations, the conclusion is similar. From expression (20) above, which is valid only when 0 < ω < eΦ 0 < Λ/2, we conclude that the total reflection coefficient (i.e. the ratio between the total reflected particle number current and the incident current) is given by the LHS of (20) and is always greater than one, characterizing a generalization of the usual superradiant scattering which involves extra scattering channels. Note that, in the general case, without an exact solution we have no information about how this total reflection is distributed between the low and high wavenumber channels represented, respectively, by \|R 1 \| 2 and \|k r 2 /k in \|\|R 2 \| 2 . However, by looking at the Λ series expansion of the reflected and transmitted wavenumbers, we can draw some interesting conclusions about the regime Λ ≫ 1. Using (19), we obtain the following expansions for the relevant wavenumbers, As mentioned before, all possible modes of the system are characterized by the same conserved frequency ω. The momentum of each mode, on the other hand, is determined by the wavenumber k and, therefore, is not the same for every mode. From the expansions above, we note that k t 2 , k r 2 ∼ O(Λ) while k in , k r 1 , k t 1 ∼ O(Λ 0 ). In particular, the difference between the high momentum modes k r 2 and k t 2 is k r If the potential is sufficiently smooth, this suggests that the creation of a pair of these modes (k r 2 , k t 2 ) should be favoured since it requires a negligible momentum change in the system if Λ ≫ 1. It is also interesting to analyse the case of a non-smooth potential, by solving the idealized problem of a step function, i.e. V (x) = eΦ 0 Θ(x) (see the appendix for a discussion concerning the appropriate boundary conditions used at the discontinuity The reflection coefficient \|R 1 \| 2 for the subluminal case as a function of ω/eΦ 0 calculated for the step potential V (x) = eΦ 0 Θ(x) with eΦ 0 = 1. Note the singular behaviour at ω = eΦ 0 /2, which is present even in the non-dispersive case. This divergence is directly related to the discontinuity in V (x) and would be cured by smoothing the step potential at x = 0. point x = 0). In such a case, one can show that the reflection and transmission coefficients are given by, where the wavenumbers k are explicitly given in (19). Note that the expressions above are valid for any value of Λ, not only in the regime Λ ≫ 1. In order to observe the effects of the dispersive parameter, we plot the reflection coefficient then the total reflection coefficient is always superradiant, as proven in (20). Note that the presence of extra scattering channels even enhances the amplification process in comparison with the non-dispersive case. This can be understood in the sense that the amplifier is more effective if the number of 'accessible' channels increases. Note also that the total reflection coefficient in the dispersive case is not continuously connected with the non-dispersive regime. More precisely, as Λ is increased in FIG. 3, the total reflection coefficient becomes larger and larger and moves away from the non-dispersive coefficient (Λ = ∞). A possible explanation for this behaviour resides in the fact that the subluminal dispersion relation itself is not continuously connected to the linear dispersion relation in the limit Λ → ∞. In other words, the modes k r 2 and k t 2 which are always absent in the non-dispersive case, will be present in the dispersive regime no matter how large Λ is. Superluminal scattering Let us now turn our attention to the case of a superluminal dispersion. Once again, we can understand much of the scattering process from the dispersion relation, see FIG. 4. The main difference with the subluminal case is that now there are only two real roots in either asymptotic regions, corresponding to one left-moving and one right-moving mode. The other two solutions to the fourth order equation are imaginary roots, corresponding to exponentially decaying and exponentially growing modes. Similarly to the subluminal case, we are interested in the scattering of an incident low-frequency (0 < ω < eΦ 0 ) wave originating from x → −∞, which is converted into a reflected left-moving mode, a transmitted right-moving mode and exponentially decaying modes (as a boundary condition, we impose the fact that there can be no exponentially growing modes in the asymptotic regions). Therefore, the solution of (10) corresponding to this scattering process is given by where the coefficients R and T are related, respectively, to reflection and transmission coefficients (see (25) below), and the coefficients E r and E t are the coefficients of the exponentially decaying modes. The wavenumbers k in (23) are obtained directly from the dispersion relations (12) and (13), The conservation of particle number current can be obtained by inserting (23) into (17) and equating the generalized Wronskian at ±∞. Compared to the nondispersive result, one might expect extra terms related to the exponentially decaying modes. However, these extra contributions are also exponentially decaying and, therefore, their particle number currents are negligible at ±∞. As a consequence, one obtains the following reflection coefficient (valid for 0 < ω < eΦ 0 ), which, similarly to the subluminal case, is always greater than one. Note also that the expression above reduces to the usual non-dispersive reflection coefficient (7) in the limit Λ → ∞. Repeating the analysis used in the subluminal case, one can solve exactly the problem for a step potential given by V (x) = eΦ 0 Θ(x). Using appropriate boundary conditions at x = 0 (see the appendix), one can relate the coefficients in (23) to the wavenumbers given in (24), Once again, it is useful to plot the reflection coefficient as a function of ω for different values of the dispersive parameter Λ, see FIG. 5. The results obtained for the scattering of sub and superluminal dispersive fields in our 1+1-dimensional toy-model demonstrate that superradiance is possible and that the amplification is enhanced due to the extra scattering channels in comparison with non-dispersive superradiance. These results are interesting for experimental attempts of detecting superradiance, since they indicate that dispersion may increase the amplification rates and, consequently, make the effect easier to observe. Generalization to non-zero flows Having in mind moving media in analogue models of gravity, we extend the ideas of the previous section by including a position-dependent flow velocity W (x) in our model. In other words, the medium where the scattering process takes place (fluid frame) is in relative motion with respect to the observer (lab frame). If we require that the dispersion relation be unaltered in the comoving frame of the fluid, the action (9) generalizes to Consequently, the modified Klein-Gordon equation (8) generalizes to where the derivative operator ∂ x is understood to act on everything to its right. It is important to remark that the reference velocity c = 1 in such analogue systems corresponds to the velocity of the scalar perturbations (e.g. the sound speed in hydrodynamical systems). Usually, such velocities (and also W (x)) are much smaller than the speed of light and, therefore, the system is nonrelativistic. For relativistic analogue models of gravity, we refer the reader to [28,29,30]. After separation of variables, φ(t, x) = e −iωt f (x), the wave equation (28) becomes, (29) which can be written as where Observe that these coefficients are not independent but satisfy the following relations, We choose the functions W (x) and V (x) to be asymptotically constant, so that, at ±∞, any solution to (29) can be decomposed into plane waves satisfying the dispersion relation below, where the ± stands for super and subluminal respectively. As previously, we work under the assumption that ω, eΦ 0 ≪ Λ. Similarly to the zero-flow case, we can obtain a conserved quantity by calculating the x-component of the Noether current associated with the symmetry φ → e iα φ. After separating the temporal dependence in φ, one can show that a modification of (14), and the corresponding scaled quantity X = Z/(2iΛ 2 ) are independent of x. The action of the functional X on a linear combination of 'on-shell' plane waves takes precisely the same form as in the zero-flow case, see (17) (the only difference is that the effective frequency is now given by Ω = ω − eV − kW instead of Ω = ω − eV ). This result highlights the generality of the algebraic structure of the particle number currents given by (17). Additionally, the existence of superradiance in the zero-flow case was related to the condition that ω − eΦ 0 < 0. We can anticipate that the occurrence of superradiance in non-zero flows will be favoured by modes for which ω − eΦ 0 − kW 0 < 0. Subluminal dispersion Let us consider first the case of a subluminal dispersion relation. As shown in FIG. 6 (red solid curve), for fixed eΦ 0 /Λ < W 0 < 1, there are two distinct intervals of frequencies separated by a critical frequency ω crit < eΦ 0 in which we expect superradiance: for 0 < ω < ω crit (region I) two propagating modes are admitted, both right-moving in the lab frame; for ω crit < ω < eΦ 0 (region II) there are four real roots of the dispersion relation corresponding to four propagating modes, three right-moving and one leftmoving (with respect to the lab-frame). Note that the requirement that W 0 is not too small, specifically W 0 > eΦ 0 /Λ, is necessary in order to guarantee that the left-most root in region I (i.e. the circular dot labeled k t 2 in region I of FIG. 6) has positive group velocity and hence defines a true transmitted mode. If 0 < W 0 < eΦ 0 /Λ, the situation is basically the same as the one discussed in section 2.1. On the other hand, if W 0 > 1 (see the black dashed curve in FIG. 6), there is only one possible regime: for all frequencies 0 < ω < eΦ 0 , two right-moving modes are admitted. An interesting fact is that, due to the absence of left-moving modes at +∞ when ω < ω crit , this system is a model for the event horizon of an analogue black hole with modified dispersion relations. However, the precise location of the horizon, besides being ω-dependent, is also rather ill-defined, relying on a global solution to the equation of motion in the vicinity of a classical turning point. This region can be studied by WKB methods and Hamilton-Jacobi theory [31], but it is not of specific interest to us here. According to the analysis above, the scattering of an incoming wave from −∞ will result in transmission through two or three channels, depending on whether ω < ω crit or not. An exact solution to the scattering problem can be decomposed as when 0 < ω < ω crit (if W 0 < 1) or 0 < ω < eΦ 0 (if W 0 > 1) and as when ω crit < ω < eΦ 0 (only possible if W 0 < 1). The wavenumbers k in , k r 1 and k r 2 are given by (19) and are labeled in FIG. 1, while the transmitted wavenumbers are not, in general, expressible as simple functions of the parameters. Note that we do not explicitly keep track of possible exponential decaying solutions in (38) (corresponding to complex solutions to the dispersion relation), since they do not contribute directly to (40). We insert the solutions (38) and (39) into the functional X of (17) and find, after algebraic manipulations, the following relationship between the transmission and reflection coefficients, where the sum is over all (2 or 3, depending on ω and W 0 ) transmission channels. Here, v gn = v g (k tn ) are the group velocities of the transmitted modes at +∞. The LHS of (40) can be interpreted as the total reflection coefficient associated with the incident modes. Since the group velocities v gn of the transmitted modes are, by definition, always positive, the sign of the contribution from each channel k tn to the RHS of (40) is determined by the factor Ω(k) = ω − eΦ 0 − kW 0 evaluated at k tn , as we anticipated previously. When only two transmission channels are admitted, one of the two factors Ω(k) is strictly negative while in the case of three transmission channels, two of the three factors Ω(k) are strictly negative. Therefore, in both situations there is one root which contributes an overall negative amount to the RHS of (40) and thus reduces the magnitude of the total reflection. Because of these troublesome modes, the RHS is not strictly greater than 1 and we cannot straightforwardly conclude superradiance. In general, in order to fully answer the question of superradiance, one would need to specify W (x) and V (x) in all space and solve for the coefficients R n and T n . Large Λ approximation: To better understand the relation between (40) and superradiance, we now focus on small deviations (Λ ≫ 1) from the non-dispersive case. For fixed W 0 < 1, the size of region I in FIG. 6 becomes zero when Λ is greater then eΦ 0 [2(1 − W 0 )/3] −2/3 . We therefore assume that the frequency ω lies in region II, where three transmission and two reflection channels are available, see (39). As explained above, in the most general case one would need to solve the equation of motion for all x in order to conclude superradiance or not from (40). Similarly to the zero flow case, by looking at the series expansions of the relevant wavenumbers, we can make useful predictions about the transmission coefficients in this Λ ≫ 1 regime. We start the analysis by solving (36) in the asymptotic region x → +∞ and expressing the obtained transmitted wavenumbers as power series in Λ, The series expansions of the incident wavenumber k in and of the reflected wavenumbers, k r 1 and k r 2 , are given, as previously, by (21). From these Λ expansions, we note that, unless W 0 ∼ 1, we have k t 2,3 , k r 2 ∼ O(Λ) and k in , k r 1 , k t 1 ∼ O(Λ 0 ). However, unlike in the zero-flow case, the difference between the high momentum transmitted modes k t 2,3 and the high momentum reflected modes k r 2 is not negligible for Λ ≫ 1, being O (Λ). Hence, the appearance of such modes requires a large momentum change in our system. Therefore, if the potential is sufficiently smooth, we expect the conversion of incident modes k in into transmitted modes k t 2 and k t 3 and into reflected modes k r 2 to be disfavoured in comparison with the low momentum transmission/reflection channel involving k t 1 and k r 1 . In other words, we expect that the Wronskian condition (40) will include only the low momentum channel, i.e. From this relation for the reflection coefficient, we would conclude that superradiance occurs for all frequencies in region II (ω crit < ω < eΦ 0 ) when Λ ≫ 1. Note that this conclusion certainly does not hold in the case of a general dispersive parameter. If the condition Λ ≫ 1 is not satisfied, there can be a mixture between the high and low momentum channels. Consequently, as discussed before, the total reflection coefficient given by (40) is not necessarily larger than one. In such a case, a definite answer about superradiance can only be obtained by solving the differential equations at every spatial point x. Having analyzed the case of W 0 < 1, let us now fix W 0 > 1 and assume 0 < ω < eΦ 0 . As discussed previously, there are two transmission and two reflection channels available when W 0 > 1 and the scattering problem is now described by (38). Since we are interested in the regime of large Λ, we calculate the wavenumber of the transmitted modes up to next-to-leading order terms, By direct substitution of these expressions into (40), one can straightforwadly determine the reflection coefficient in powers of Λ, plus terms of O (Λ −2 ). Note that this second channel T 2 is present even in the absence of dispersion, being an upstream mode which is swept downstream by a superluminal flow. From the equation above, it also becomes evident that the relation between the norms \|T 1 \| and \|T 2 \| of the two transmission channels determines the occurrence or not of superradiance. Critical case: An interesting situation to be analyzed is the critical case ω = ω crit , which corresponds to the boundary between regions I and II in FIG. 6. In this scenario, the background flow W 0 , when expanded in powers of Λ, relates to the critical frequency according to the following expression, Note that the previous analysis leading to (43) relied on series expansions (see (42)) which are not valid when Therefore, in order to analyze the possibility of superradiance in the critical case, we cannot use (43); instead, we have to start from the original Wronskian relation (40). In the critical regime, the dispersion relation (see FIG. 7) has three distinct solutions. Two of these solutions, denoted by k t 1 and k t 2 , have positive group velocities in the lab frame and, therefore, are identified as transmitted modes. The other solution, denoted by k 0 , is a degenerate double root and, consequently, has a vanishing group velocity in the lab frame. In order to obtain the reflection coefficient for the scattering problem, we first expand k t 1 and k t 2 as power series in Λ, and then substitute the obtained expressions into (40). The final result is given by plus terms of order O Λ − 2 3 . Observe again the importance of the relative sign of the norm in the two transmission channels. Note that the scattering process converts incident modes with wavenumber , which can only be balanced by the high momentum reflected modes k r 2 ≈ O (Λ). Another possibility is the conversion of the incident modes k in ≈ ω + O(Λ −1 ) into transmitted modes of wavenumber k t 1 ≈ (ω − eΦ 0 )/2, which is comparable to the low momentum k r 1 channel. Because of the high momentum change required by the second channel (involving k r 2 and k t 2 ), we expect the first channel (involving k r 1 and k t 1 ) to be favoured in our scattering experiment. Since ω crit < eΦ 0 , we therefore deduce that the RHS of (48) is always greater than one in the limit Λ ≫ 1. In other words, low-frequency waves in the critical regime are superradiantly scattered in our toy-model if small subluminal corrections are added to the dispersion relation. Superluminal dispersion We now turn to superluminal scattering processes, which can be quite different compared to subluminal ones since there does not exist any notion of a horizon or modeindependent blocking region for high momentum incident modes (the group velocity dω/dk is unbounded as a function of k and only the low frequency modes which possess quasilinear dispersion experience a blocking region in such flows). We will follow the standard treatment [31] of analogue black holes with superluminal dispersion and analyze the transmission of an incoming wave from −∞ through to +∞. The relevant dispersion relation in the superluminal case is depicted in FIG. 8. Given a superluminal flow W 0 > 1 (solid red curve in FIG. 8), there exists an interval of frequencies 0 < ω < ω crit (region I in FIG. 8) for which only two propagating modes are admitted, one right-moving and one left-moving in the lab frame. For ω crit < ω < eΦ 0 (region II in FIG. 8), however, there are four propagating modes, two transmitted rightmovers and two left-movers. The third possibility is a subluminal flow W 0 < 1 with 0 < ω < eΦ 0 , for which there are always two propagating modes (see the black dashed curve in FIG. 8). Let us discuss first the cases in which only two propagating modes are available in the asymptotic limit x → +∞. Hence, the scattering is produced by an incident wave from −∞ whose frequency ω satisfies 0 < ω < ω crit (if W 0 > 1) or 0 < ω < eΦ 0 (if W 0 < 1). Note that exactly one of these two propagating modes is a left-moving mode. Imposing the boundary condition that no incoming mode is allowed at +∞, we obtain the solution of (29) corresponding to the scattering problem, plus exponentially decaying channels which do not contribute directly to the generalized Wronskian current calculated at x → ±∞. Note that the wavenumbers k in and k r are the same ones that appear in (24) for the superluminal W = 0 case and the wavenumber k t represents the only available transmission channel. On the other hand, if W 0 > 1 and the frequency ω of the incident wave satisfies ω crit < ω < eΦ 0 (region II in FIG. 8), then there are, in principle, two extra propagating channels available (four in total, as discussed above). However, because of the boundary condition imposed at x → +∞, only one extra transmission channel has to be considered (the other extra channel is always left-moving at x → +∞). The scattering solution is then given by where k in and k r are again given by (24) and k t 1 and k t 2 are the wavenumbers of the transmitted modes. Note that we have once again omitted the exponential decaying mode at x → −∞ since it does not affect directly the generalized Wronskian. Using (17) to evaluate the functional X in both asymptotic regions, we obtain, similarly to the subluminal case, the following relation between the reflection and transmission coefficients, where the sum is over one or two transmission channels, depending on ω and whether W 0 > 1 or W 0 < 1. Here, v g 1 and v g 2 are the group velocities of the transmitted modes k t 1 and k t 2 , which are always positive by definition. Furthermore, it is possible to show, for frequencies 0 < ω < eΦ 0 , that the effective frequency Ω = ω − eΦ 0 − k tn W 0 is always negative for k t 1 modes and always positive for k t 2 modes. Therefore, since only the n = 1 transmission channel is available for frequencies lying in region I of FIG. 8, we conclude that the RHS of (51) is greater than 1 and, therefore, the scattering is always superradiant. The situation for W 0 < 1 and 0 < ω < eΦ 0 is similar: only the first transmission channel is available and superradiance always occurs. However, for frequencies located in region II, we cannot so easily conclude superradiance since the extra transmission channel k t 2 contributes an overall negative factor in (51). To obtain a conclusive answer, one would need to know the detailed structure of W (x) and V (x) in the intermediate regime and solve the equations not only in the asymptotic regions but at every point x. Large Λ approximation: In order to better understand the scattering of an incident wave whose frequency is located in region II of FIG. 8, we shall consider small deviations from the non-dispersive limit, i.e. Λ ≫ 1. In such a case, we can expand the two transmission channels, k t 1 and k t 2 , in powers of Λ, and substitute the obtained wavenumbers into (51) in order to determine the reflection coefficient for the scattering, Since we assume ω < eΦ 0 in region II, this reflection coefficient is larger than 1 whenever \|T 2 \| < \|T 1 \|. As explained above, whether this condition is satisfied or not in a general model would depend on the detailed structure of W (x) and V (x) in the intermediate regime [31]. From a practical point of view, in order to maximize the potential for superradiance in an experiment with a superluminally dispersive medium, one should choose the asymptotic flow W 0 as small as possible while still being superluminal (W 0 > 1) as this would minimize the size of region II and the extra positive-effectivefrequency transmission channels therein. Choosing the flow as such to maximize the T 1 channel is also consistent with our intuition that scattering favors the channel which most closely matches the momentum of the reflected mode; in this case the wavenumber k t 1 is closer to k r than k t 2 is. This prediction is confirmed in the step function model for which V (x) = eΦ 0 Θ(x) and W (x) = W 0 Θ(x). In such a case, one can impose the appropriate boundary conditions at x = 0 discussed in the appendix to obtain the following reflection and transmission coefficients, We can also show that the coefficient correponding to the omitted exponential decaying mode in (50) is of order O (Λ −2 ). Comparing T 1 and T 2 above and using the fact that W 0 > 1, one can see that \|T 1 \| > \|T 2 \| at zeroth order in Λ, which implies superradiance and confirms our expectations. Alternatively, one can verify the occurence of superradiance by directly analyzing the reflection coefficient R above. It is straightforward to see that \|R 2 \| > 1 at lowest order in Λ. Critical case: Another interesting possibility that we now consider in detail is the critical regime ω = ω crit . This situation corresponds to the boundary between regions I and II in FIG. 8 and is depicted, in the fluid frame, in FIG. 9. The relation between the background flow W 0 and the critical frequency ω crit is given by the following expression, Note that, since W 0 −1 ∼ O Λ − 2 3 , the Λ expansion (54) of the generalized Wronskian obtained previously is not valid in the present case (check the denominators in (53)). Consequently, we shall need different Λ expansions in order to obtain an appropriate expression for the reflection coefficient. Like in the critical subluminal case, the dispersion relation has three distinct roots: the double root k 0 (with vanishing group velocity in the lab frame), a right-moving mode (with negative group velocity in the lab frame) and a transmitted mode (with positive group velocity in the lab frame) whose wavenumber k t is given by Applying as a boundary condition the fact that only right moving modes are allowed at +∞, we obtain, after substituting the relevant quantities into (51), the relation between the reflection and transmission coefficients for the scattering process, Since ω crit < eΦ 0 , we conclude that the RHS of the equation above is always greater than one when Λ ≫ 1. In summary, superradiance is expected to occur in the superluminal critical case for small deviations from the non-dispersive regime. Inertial motion superradiance Throughout this paper, inspired by the usual condition for rotational superradiance in the black hole case, i.e. ω −mΩ h < 0, we analyzed only scattering problems in which the frequency of the incident mode satisfies ω − eΦ 0 < 0. However, since it is the effective frequency Ω that appears in (17), we conclude that the amplification of an incident mode can also occur for ω − eΦ 0 > 0 given that ω − eΦ 0 − kW 0 < 0. In particular, even when Φ 0 = 0 superradiant scattering will be possible. However, being due exclusively to inertial motion in the system, this kind of superradiance is outside the scope of our work. In fact, inertial motion superradiance has long been known in the literature as the anomalous Doppler effect and the condition for negative effective-frequency modes is referred to as the Ginzburg-Frank condition [32]. Several phenomena in physics, like the Vavilov-Cherenkov effect and the Mach cones (which appear in supersonic airplanes) can be understood in terms of inertial motion superradiance [18]. For a detailed analysis of inertial motion superradiance, we refer the reader to [18]. Applications: axisymmetric systems Having analyzed superradiance in simple 1+1-dimensional toy models with modified dispersion relations, we will now discuss how the ideas presented in this paper can be generalized to more realistic situations based on analogue models of gravity. Our starting point is a general 2+1-dimensional, axisymmetric and irrotational fluid flow with background velocity v given by where A and B are constants and (r, φ) are the usual polar coordinates. Velocity perturbations δv of the background flow can be conveniently described by a scalar field ψ, which relates to δv through equation δv = ∇ψ. We denote the propagation speed of these perturbations by c. The idea of analogue gravity is derived from the observation that the differential equation satisfied by the perturbations ψ, i.e. can be cast into a Klein-Gordon equation in an effectively curved spacetime geometry. This connection between hydrodynamics and gravity is responsible for many important results, see [2] for a detailed review. One of the successes of Unruh's [1] original idea of using sound waves to study gravitational phenomena is that it can be extended to many other physical systems, like gravity waves in open channel flows [33] and density perturbations in Bose-Einstein condensates [34]. An important feature of such systems is that (62) is only accurate in certain regimes; at sufficiently small distance scales (e.g. wavelengths comparable to the fluid depth in open channels), the dispersion relation is not linear anymore and (62) has to be replaced by ¶ where Λ is a dispersive parameter and the upper (lower) sign corresponds to subluminal (superluminal) dispersion. Even though superradiance has been subjected to extensive studies in the linear regime of analogue models, it has never been analysed before in the context of modified dispersion relations, as opposed to Hawking radiation (see e.g. [2] and references therein). The remarkable fact about the toy models introduced in this paper is that they can be used to analyze superradiance in realistic dispersive analogue models of gravity satisfying (63). Firstly, the electromagnetic interaction term eΦ 0 appearing in our toy models is analogous to the rotational term mΩ in axisymmetric analogue models, where m is the azimuthal number and Ω is the angular velocity. It is interesting to note that this duality is manifest in real black holes: both electromagnetic [35] and rotational [25] superradiance are possible. Another essential ingredient for the occurrence of superradiance in analogue models of gravity is the presence of an event ¶ Note that this equation includes only fourth order corrections. A full description of the system will possibly include also higher order terms. horizon, which allows no mode to escape from inside the analogue black hole. In our toy model, such behaviour is mimicked by an appropriate boundary condition imposed in the asymptotic limit x → ∞. It is also important to address the usefulness of the generalized Wronskian (37) in the context of axisymmetric systems. More precisely, we are going to show that, if all derivatives with respect to x are replaced by derivatives with respect to r, then the Wronskian (37), when applied to solutions of (63), is independent of r. Indeed, by applying the ansatz ψ = (H(r)/ √ r) e imφ e −iωt , we are able to separate (63) and are left with a radial equation for H, where the coefficients α(r), β(r) and γ(r) are given by The functions P (r) and Q(r) appearing in the coefficients above are, up to a factor (c 2 − v 2 r ), the same functions P and Q defined in Refs. [36,37]. They can be expressed as where the upper (lower) sign corresponds to subluminal (superluminal) dispersion. Finally, we note that (64) is exactly the same as (30) and, more remarkably, that the coefficients α, β, γ above satisfy conditions (32) and (33). Consequently, the generalized Wronskian defined in (37) can also be used in the context of axisymmetric analogue models of gravity, thus completing the connection between our toy models and realistic physical systems. Summary and final remarks We have proposed idealized systems to investigate multi superradiant scattering processes that are applicable to sub and superluminal dispersive fields. Perhaps the most important theoretical result obtained is related to the simplicity of the analytic expression for the particle number currents J n of the scattering channels n (see (17)), Notice that the expression above depends only on the amplitude, group velocity and effective frequency of the particular scattering channel. Moreover, this result is universal to all scattering processes discussed in this paper. Note also that, in principle, we could have normalized the modes so that the group velocity in (67) is absorbed into the coefficients A n . Doing that would make the conservation equations (see e.g. (20)) look much simpler. However, their dependence on the dispersive parameter Λ would then also be hidden in these new coefficients. Our findings link to standard scattering processes, allowing a deeper insight into superradiance. Let us consider scattering of up to four incident modes by a general scattering potential. There are up to four channels to the left {a, b, c, d} and four to the right {A, B, C, D} of the scattering potential. As a lesson from the analysis carried out before, one needs to be careful when assigning the propagation direction of each mode since its group velocity is dependent on the particular scattering potential and type of dispersion, see FIG. 10. Furthermore, as long as the scattering potential is real, the total current to the left equals the total current to the right of the potential, The scattering potential is amplifying classical and quantum field excitations if, at Figure 10. The scattering processes for sub and superluminal dispersive fields, with two low (red solid) and two high (blue dashed) momentum degrees on each side of the potential, are represented by a general scattering process whose particle number currents are given by (67). the left side of the potential, the total outflux (i.e. the reflected current) is larger then the total in-flux, J ref total > J in total . There are several scattering coefficients that can be considered: the reflection coefficient in each individual scattering channel, Figure 11. Schematic representation of some of the scattering processes investigated in the paper. The arrows indicate the group velocity propagation. The particle number current direction is related to the relative sign between this group velocity and the effective frequency Ω. This information, together with (68), can help determine if superradiance occurs or not. In addition, the present work can be related to recent experimental realizations of analogue black holes, some of which have even studied the Hawking emission process. In particular, [9] exhibits the first detection of the classical analogue of Hawking radiation using dispersive gravity waves in an open channel flow. One of the most important lessons to be learned from that work is that even though vorticity and viscosity effects cannot be completely removed from the experimental setup, they can be made extremely small. In fact, they can be reduced to the point that the results predicted by the irrotational and inviscid theory match the results obtained experimentally with considerable accuracy. Based on this fact, together with our present results, one might ask whether superradiance occurs (and can be observed) in the laboratory using dispersive gravity waves. The connection between such system and our analysis can be seen directly from the full dispersion relation for gravity waves in a fluid of constant depth h [38], where g is the gravitational acceleration and σ and ρ correspond, respectively, to the surface tension and the density of the fluid. If first-order deviations from the shallow water limit (kh ≪ 1) are considered, we recover the quartic dispersion relation analyzed in this paper. In other words, non-shallow gravity waves impinging on a rotating analogue black hole (e.g. a draining 'bathtub' vortex) satisfy (63) with c 2 = gh and a dispersive parameter Λ −2 = \|gh 3 /3 − σh/ρ\|. Based on the existent analysis of superradiance of linear fields in open channel flows [33,37] together with our discussion in Section 4, we expect superradiance to also be manifest for non-shallow gravity waves. This relation between dispersion and superradiance in open channels is currently being further investigated by the authors and will be the subject of a future work. Another class of analogue black holes, which was recently set up in the laboratory [6] and which might be used in the future to produce superradiant scattering processes, consists of Bose-Einstein condensates [34,2]. The analogy with gravity arises when one considers the Gross-Pitaevskii equation and uses the Madelung representation of the condensate wave function. If the eikonal approximation is used and axisymmetry is assumed, it is possible to show that perturbations around a background condensate obey a superluminal dispersion relation given by ω 2 = 4π 2 n 0 a m 2 k 2 + 2m where m is the mass of a single boson, a is the scattering length and n 0 (r) is the background density. Such perturbations are described by (63) with c 2 = 4π 2 n 0 a/m 2 and Λ = 2m/ . Based on our work, the most obvious conclusion we can draw is to expect superradiance to be manifest also in low-frequency BEC scattering experiments. However, since the dispersion relation is superluminal, there is no notion of a mode independent blocking region (see section 3.2) and, consequently, it is not clear if the appropriate boundary conditions will be sufficient to guarantee superradiance. Additionally, quantized vortices may be present in such systems, restricting the angular momentum to integer multiples of . The physics of these quantized vortices with respect to superradiance and instabilities is also unclear at this point and more investigation is needed to understand their role in possible BEC scattering processes.	2013-03-25T23:36:56.000Z	2012-08-17T00:00:00.000	{ "year": 2012, "sha1": "f2d98772fd919e4f68ff065fa5a7a997316d2a77", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1208.3601", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "f2d98772fd919e4f68ff065fa5a7a997316d2a77", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
9721382	pes2o/s2orc	v3-fos-license	Overlaps Between Autism and Language Impairment: Phenomimicry or Shared Etiology? Traditionally, autistic spectrum disorder (ASD) and specific language impairment (SLI) are regarded as distinct conditions with separate etiologies. Yet these disorders co-occur at above chance levels, suggesting shared etiology. Simulations, however, show that additive pleiotropic genes cannot account for observed rates of language impairment in relatives, which are higher for probands with SLI than for those with ASD + language impairment. An alternative account is in terms of ‘phenomimicry’, i.e., language impairment in comorbid cases may be a consequence of ASD risk factors, and different from that seen in SLI. However, this cannot explain why molecular genetic studies have found a common risk genotype for ASD and SLI. This paper explores whether nonadditive genetic influences could account for both family and molecular findings. A modified simulation involving G × G interactions obtained levels of comorbidity and rates of impairment in relatives more consistent with observed values. The simulations further suggest that the shape of distributions of phenotypic trait scores for different genotypes may provide evidence of whether a gene is involved in epistasis. Introduction Specific language impairment (SLI) refers to a condition where a child fails to develop spoken language on the normal schedule, for no obvious reason (Bishop and Norbury 2008). Potential causes such as hearing loss, low general ability, or physical impairment of articulators are excluded. Development in areas such as skills of daily living and nonverbal ability is age-appropriate. Autism is also excluded, and the textbook picture of SLI is of a child with normal social interaction and nonverbal communication, but with specific difficulties in mastering structural aspects of language, especially syntax (grammatical devices such as word order and inflectional endings) and phonological skills (identification and production of speech sounds). Autistic disorder also involves impairments of communication, but these are much broader, affecting pragmatics, i.e., the appropriate use of language in context, as well as nonverbal communication. In addition, there are impairments in social interaction and understanding, and the repertoire of behaviour and interests is unusual and often restricted (Dover and Le Couteur 2007). The diagnostic manuals ICD-10 and DSM-IV (American Psychiatric Association 1994; World Health Organization 1993) use different terminology, but both make a clear diagnostic distinction between a specific developmental disorder affecting language and the more pervasive difficulties seen in autism. The past few decades have seen two major changes in our conceptualisation of the etiology of both autism and SLI. The first breakthrough was the abandonment of purely environmental explanations for these conditions as it became clear from twin studies that they were both highly heritable. Several studies showed that identical, monozygotic (MZ) twins were significantly more concordant than fraternal, dizygotic (DZ) twins for autism (Bailey et al. 1995;Folstein and Rutter 1977;Hallmayer et al. 2002;Steffenburg et al. 1989) and SLI (Bishop et al. 1995;Bishop and Hayiou-Thomas 2008;Lewis and Thompson 1992;Tomblin and Buckwalter 1998), despite growing up together and sharing many environmental influences. The next shift in understanding was from a focus on single genes of large effect to an etiological model of these conditions as complex and multifactorial, resulting from the influence of many genes of small effect, combined with environmental influences (see review by Bishop 2009). There were several reasons for this development. First, the high heritability estimates from twin studies did not translate into discoveries of common single gene mutations of large effect, as might have been expected. In the field of SLI, discovery of a mutation of the FOXP2 gene in one multigenerational family led to a fascinating series of studies exploring the gene's evolution and mode of action (see Fisher, 2007, for review andKonopka et al. 2009 for more recent work), but it has become clear that it is a rare cause of speech and language impairments (Newbury et al. 2002). Second, pedigree studies showed that in both autism and SLI, rates of impairment in first degree relatives are higher than in the general population, but it is unusual to observe a classic Mendelian pattern of inheritance (Lewis et al. 1993;Rutter 2005a). In short, these conditions aggregate but do not segregate (Sing and Reilly 1993). Third, family studies indicated that first degree relatives of affected individuals often manifest subthreshold symptoms, such as subtle phonological difficulties in relatives of children with SLI ), or mild social and communicative difficulties in relatives of those with autism (see review by Bailey et al. 1998). This suggested that these conditions correspond to points on a continuum of impairment, rather than all-or-none diseases. Finally, studies in clinical medicine led to growing awareness that complex multifactorial etiology is the rule rather than the exception for disorders that are common in the general population, consistent with arguments by evolutionary scientists that genetic variants that had large effects on reproductive fitness in ancestral humans would be unlikely to persist in modern humans (Keller and Miller 2006). A model of the etiology of SLI and autism is shown in Fig. 1. This is based on the original approach to complex multifactorial disorders proposed by Falconer (1965), and subsequently developed to contrast different models of comorbidity by Neale and Kendler (1995). The model assumes that underlying a categorical diagnosis is a normally distributed liability continuum. Figure 1 differs from the traditional models by having disorder result when the value on the continuum falls below a threshold, whereas Falconer treated liability as a positive risk variable with disorder resulting from a high value. The reason for this change is to make the model more compatible with conceptualisations of developmental disorders, where disorder is typically recognised when a score on a normally distributed ability is unusually low. In effect, we work with negative liability, but this affects only the graphical depiction of the model, and has no computational consequences. In the classic Falconer model, liability is a theoretical construct, not directly observed, but inferred from the frequencies of disorders in probands and their relatives. Nevertheless, it should be possible to identify measures that are correlated with the causal trait, which will be termed 'liability markers' for disorder. These could be indices of neurological or cognitive function. The construct of liability marker overlaps with the notion of endophenotype (Gottesman and Gould 2003), but is more general because it is not restricted to genetically determined traits. An additional point to note is that although the model specifies a single causal trait, risk factors are not unitary: on the contrary, they can be partitioned into genetic and environmental risks, with the latter subdivided into those shared between relatives and those specific to the individual. For each type of risk, there are many influences, each of small impact, whose combined effect gives rise to the continuous distribution on the relevant causal trait. The cutoff on the causal trait will determine the frequency of disorder. The prevalence of SLI depends on the operational criteria that are used; an epidemiological study in the US estimated prevalence as 7% in kindergarten (Tomblin et al. 1997). This prevalence corresponds to a z-score threshold of around -1.5. Estimates of the prevalence of autism have mounted steadily (Rutter 2005b), from 4 per 10 000 in the 1960s up to 38.9 per 10 000 in a recent epidemiological survey in the UK (Baird et al. 2006). This is usually thought to reflect changing diagnostic criteria (King and Bearman 2009), although a genuine increase cannot be ruled out. In addition, autism is now regarded as a spectrum disorder rather than an all-or-none disease, with milder and partial forms of disorder being labelled as cases of 'autism spectrum disorder' (ASD) or pervasive developmental disorder not otherwise specified. When these cases are included, the prevalence rises to 116.1 per 10 000 (Baird et al. 2006), with corresponding z-score = -2.3. Nevertheless, even this estimate is lower than the prevalence of SLI and the cutoffs are shown as different in Fig. 1 to reflect this. In simulations discussed below we use the latter prevalence rate, and hence refer to ASD rather than autism. The neat divide between independent disorders, shown in Fig. 1, has been questioned in recent years. It has been argued that the conditions of ASD and SLI may be less distinct than the textbooks imply. There are three lines of evidence that any etiological model has to account for: (a) apparently above chance levels of comorbidity between SLI and ASD; (b) rates of language impairment in relatives of probands with SLI and ASD; (c) molecular genetic findings of shared genetic risk factors for ASD and SLI. Comorbidity between SLI and ASD According to conventional diagnostic frameworks, SLI and ASD are mutually exclusive diagnoses-ASD is explicitly excluded when making a diagnosis of SLI, which is, by definition, a 'specific' developmental disorder. From this perspective it does not make sense to talk of comorbidity. Nevertheless, diagnostic frameworks do not necessarily reflect clinical reality, and there has been interest over many years in the idea that there might be overlapping language deficits in the two conditions. In discussing overlaps between SLI and ASD, it is important to distinguish between different aspects of communication. On the one hand, children need to master the structural aspects of their language-phonology and syntax. These are the domains that are most often noted to be impaired in SLI. On the other hand, children need to use those skills to communicate with others-pragmatics. Although there are exceptions, most formal language tests focus either on vocabulary or structural aspects of language, but do not assess how effectively language is used to communicate in everyday situations. The conventional view of SLI maintains that pragmatic skill is intact and the child may communicate reasonably despite having limited structural language skills (Bishop 2000). As demonstrated in a landmark study by Bartak et al. (1975), many children with ASD are poor at both structural and functional aspects of communication. These authors compared children with severe receptive SLI (termed 'developmental dysphasia'), and children with autism. They documented similarities for the two groups on language milestones and measures of language structure, but striking differences in the functional use of language (Cantwell et al. 1978). Children with autism had much broader communicative difficulties than those with SLI, extending to encompass nonverbal as well as verbal communication. Nevertheless, while poor functional communication is a hallmark of ASD, not all cases are impaired on formal language tests. Kjelgaard and Tager-Flusberg (2001) noted substantial variation in language abilities in a group of 89 children with autism. On a wide-ranging battery of language tests commonly used to diagnose SLI, 76% of them performed in the impaired range. The remaining 24% had no evidence of structural language deficits. In an epidemiological sample, Loucas et al. (2008) found that 41 of 72 (57%) children with autism and normal nonverbal IQ had impaired performance on a language battery. Subsequently, Tager-Flusberg and Joseph (2003) noted that many children with ASD were particularly poor at repeating nonsense words, a measure that has been proposed as a marker of heritable SLI (Bishop et al. 1996). Furthermore, these children tended to make similar morphosyntactic errors to those seen in SLI, i.e., omission of verb inflectional endings. These studies indicate that structural and pragmatic language deficits are logically separable, but often co-occur. Tager-Flusberg and Joseph (2003) argued that the existence of cases of ASD whose language features resembled those of SLI suggested overlaps between these disorders at a deeper level. In this paper, a distinction is drawn between pure ASD, SLI and the apparently 'comorbid' cases, referred to here as ASD?LI, who have classic autism together with language impairments of the kind seen in SLI. Rates of language impairment in relatives of probands with SLI or ASD There have been several studies of parents and siblings of people with autism on language measures, with rather mixed results (see Bailey et al. 1998). In general, relatives of people with autism are more likely than control relatives to report a personal history of language or literacy problems, but these have proved harder to demonstrate on formal testing, especially when the measures come from instruments that are sensitive to SLI (Bishop et al. 2004;Whitehouse et al. 2007). A study by Lindgren et al. (2009), which is noteworthy for its methodological rigour and relatively large sample size, explicitly compared parents and siblings from three groups of probands: pure SLI (N = 36), pure ASD (N =20), and comorbid ASD?LI (N = 32). Particular care was taken to exclude from the SLI group any individuals with autistic features. The probands with SLI and comorbid ASD?LI were similar in their language profiles, but their siblings and parents differed. The relatives of those with SLI had language deficits, in line with previous studies of SLI relatives ), but relatives of those with ASD?LI had language scores in the normal range. Relatives of the pure ASD group tended to obtain higher scores than the ASD?LI parents on language measures, but both the language scores and nonverbal IQ of the pure ASD group were above average. A supplementary analysis in which relatives were categorised according to whether they met criteria for language impairment showed the following rates of language impairment in relatives of those with pure ASD, ASD?LI and SLI respectively: siblings, 11%, 16% and 42%; fathers, 21%, 35% and 54%; mothers, 5%, 29% and 60%. The rates of LI in relatives were significantly lower in ASD?LI than in SLI probands for all relatives except fathers. Molecular genetic risks for SLI and ASD Vernes et al. (2008) studied a sample of individuals with SLI and demonstrated association between nonword repetition skills (a marker of SLI) and polymorphisms of CNTNAP2, a gene on chromosome 7q35 that is a downstream target of FOXP2. CNTNAP2 encodes a neurexin and is expressed in the developing human brain. These authors noted that association with ASD had been demonstrated for the same locus (Arking et al. 2008), with the strength of association greatest when cases were restricted to probands with severely delayed language milestones . Taken together, the evidence from comorbidity, impairments in relatives and molecular genetic risks presents a puzzling picture. The comorbidity and molecular genetic findings would appear to point to overlapping etiology, yet the data from relatives are inconsistent with that picture. In this paper, I shall first present a formal simulation of overlapping etiology through additive pleiotropic effects to demonstrate the problems this model has in accounting for observed data, before going onto consider two radically different accounts of the etiological relationship between the two disorders. A 'correlated additive risks' model The 'correlated additive risks' (CAR) model shown in Fig. 2 corresponds to Neale and Kendler's (1995) 'correlated liability model' of comorbidity. It depicts the same causal routes as Fig. 1 except for the two-headed pathway between risk factors. According to the model, each condition has its own separate risk factors and causal traits, but the risk factors for the two conditions co-occur at above chance levels. Such a model predicts that relatives of individuals affected with one disorder will be at increased risk for the other; the simulation allows us to quantify this prediction depending on the extent of correlation between risk factors, and the relative frequency of the two disorders. Before describing the simulation, it is worth noting the different routes by which risk factors may be correlated: In the simulation presented here, genetic correlation is induced by including pleiotropic genes that lead to increased risk for both disorders. However, there are other possibilities: risk genes for the two disorders may be transmitted together because they are close together on a chromosome (linkage): in general, if linkage is tight, then predictions are similar to those from a pleiotropic model. Furthermore, nonrandom (assortative) mating could lead to different risk genes being contributed by each parent. In that case, predictions about affectedness in relatives may differ for parents and siblings. Environmental risks could also be correlated, but are not discussed here because evidence from twin studies suggests that environmental factors play a relatively minor role in the etiology of both SLI and ASD (Bishop 2006;Newschaffer et al. 2002). Methods A Matlab program that simulates the CAR model is available from http://psyweb.psy.ox.ac.uk/oscci/Miscellaneous. htm. The simulation starts by assigning a set of probands values for a set of genotypes, aa, aA and AA, whose frequency is determined by the user, with default minor allele frequency of .5. The 'a' allele is designated the risk allele, and has an additive impact on the causal trait for one or both disorders, with genotype aA having an effect intermediate between aa (low) and AA (high). The user specifies the number of probands, the number of genes affecting a causal trait, and the proportion of genes that affect both traits (pleiotropy). If we have 10 genes, when pleiotropy is set to zero, there are five genes affecting each trait, and none affecting both, and r g is zero (corresponding to Fig. 1). If the proportion of pleiotropic genes is set to .2, then genes 1, 2, 3 and 4 affect SLI only, genes 7, 8, 9 and 10 affect ASD only, and genes 5 and 6 affect both traits, giving a computed value for r g of .33. The model was designed to simulate twin data, and can be set to give estimated trait scores for MZ as well as DZ twins. The focus of interest here, however, is in similarities between probands and their first degree relatives, and so only the predictions for DZ twins are considered. In addition, environmental effects that are shared between relatives are modelled by a random normal variable that is identical for first degree relatives (and for MZ twins), and nonshared environmental effects (including measurement error) are modelled with a random normal variable that has no correlation between relatives. The genetic effects, shared and nonshared environmental effects are then combined in a weighted sum where the weights reflect the values of h 2 , c 2 and e 2 input by the user. The diagnostic categories of probands and their relatives are then assigned depending on whether or not the standardized score on the liability distributions for ASD and SLI fall below threshold. Results Figure 3 shows the predicted rates of pure and comorbid disorders in probands and their relatives, at different levels of pleiotropy, when settings for h 2 , c 2 and e 2 are .7, .1 and .2, respectively for both traits, with 100,000 simulated probands, cutoffs of -1.5 z for SLI and -2.3 z for ASD, and 10 genes in total. With pleiotropy set to zero, five genes affect the SLI liability, and five affect ASD liability. When pleiotropy is greater than zero, a subset of genes affects both traits, with r g increasing with number of pleiotropic genes. As shown in Fig. 3, the prevalence of comorbid cases increases as the genetic correlation increases. Regardless of the size of genetic correlation, pure disorders tend to 'breed true', but the less extreme threshold for SLI means that probands with ASD?LI will have more relatives with SLI than with pure or comorbid ASD. The proportion of relatives with language impairment is similar for the SLI and comorbid proband groups, at around 22-23%. As well as looking at categorical outcomes, we can inspect mean scores on liability markers for language impairment, which may be seen as proxies for language test scores or a quantitative measure of autistic features. Figure 4 shows how the profiles seen in probands (unbroken lines) and their first degree relatives (dashed lines) change, as degree of genetic correlation increases. Note that relatives who themselves meet criteria for ASD are excluded, as is usually done in family studies (e.g., Lindgren et al. 2009). Relatives are not, however, usually screened for SLI, and so those who meet criteria for this condition (i.e., have language liability scores below -1.5) are retained in the sample. (As is evident from Fig. 3, the number of relatives with ASD was very small, and including them makes no visible difference to the plots). The plots of mean liability markers show that (nonautistic) relatives have a profile resembling that of probands, but less extreme. As the genetic correlation increases, probands with 'pure' disorders and their first degree relatives show a tendency to be impaired on the liability marker for the other condition, i.e., pure ASD probands and their relatives have depressed scores on the language liability marker, and pure SLI probands and their relatives have depressed scores on the ASD liability marker. The relatives of comorbid cases score as poorly as relatives of pure cases on the relevant liability marker (i.e., mean scores are comparable to pure SLI relatives on the language trait, and comparable to pure ASD relatives on the ASD trait). Comparison with observed data As we have seen, the CAR model can explain why rates of comorbidity occur in excess of the chance rate that would be expected if the two disorders were independent. It also is compatible with molecular genetic evidence of common risk variants for ASD and SLI. However, it further predicts that relatives of those with comorbid ASD?LI should resemble relatives of those with pure ASD on ASD trait markers, and resemble relatives of those with pure SLI on language measures. As noted above, this is inconsistent with observed data from studies by Bishop et al. (2004), Lindgren et al. (2009) and Whitehouse et al. (2007). The CAR model therefore fails to provide a plausible account of etiological overlaps. Fig. 3 Simulated data from CAR model showing proportions of relatives of probands with SLI, ASD and ASD?LI (labelled above histograms), with diagnosis in relatives indicated by shading, for four levels of r g . Note that the scale for the relatives of SLI probands is broken so that the full range can be shown for all groups. SLI specific language impairment, ASD autism spectrum disorder, ASD?LI 'comorbid' cases There are other models of overlapping risk factors that have been proposed in the literature, but in all cases they predict that relatives of those with comorbid ASD?LI should have an increased rate of language impairments. Thus, Tager-Flusberg and Joseph (2003) suggested that pure ASD and ASD?LI might be distinct subtypes, with a common neurocognitive phenotype for ASD?LI and SLI: however the subtype hypothesis predicts that ASD?LI will 'breed true', so relatives of affected individuals would show language difficulties. Another model of overlapping risk factors is the dimensional account of ASD by Ronald et al. (2006). Rather than assuming two liability distributions, one for autism and one for SLI, they suggest that the different components of autism (social interaction, communication and behavioural repertoire) are independently heritable, and only when all dimensions were impaired would they qualify as cases of autism. A problem for this view is that the communication dimension is not well-specified. As noted above, while an autism diagnosis requires that the child have difficulties with communication, this does not necessarily mean problems with structural aspects of language of the kind seen in SLI. Thus, to represent the full range of observed phenotypes, we need at least four dimensions, corresponding to language structure, pragmatic aspects of communication, social interaction and behavioural repertoire. As noted by Bishop (2003), such a model has the advantage of being able to capture a wide range of clinical conditions, including classic SLI (only language structure impaired), pragmatic language impairment (communication impaired, with or without poor language structure), Asperger syndrome (impairment in all domains except language structure) and autism (impairment in pragmatics, social interaction and behavioural repertoire, with or without poor language structure). The model runs into difficulties, however, in accounting for patterns of trait markers in relatives, because, like the CAR model, it predicts that the different dimensions should 'breed true', so relatives of those with comorbid autism?LI should resemble relatives of those with SLI on language trait markers, with both being impaired. Furthermore, it is hard to specify thresholds on the four causal traits that can generate plausible prevalence rates for the different types of disorder. If the traits are semi-independent, then the more traits that are impaired, the rarer the disorder will be. Autism without structural language impairment should therefore be far more common than autism with language impairment, which is not what is found (Loucas et al. 2008). Phenomimicry A very different way of accounting for similar language problems in ASD and SLI is to argue that phenotypic continuities between conditions are more apparent than real, resulting from 'phenomimicry'. I use this term to refer to the situation when the causal route for one disorder can lead to an outcome resembling the other disorder. Unlike 'phenocopy', which refers to an environmentally caused disorder that resembles a genetically-determined disorder, the term 'phenomimicry' makes no assumptions about whether etiology is genetic or environmental, or, as in this case, complex and multifactorial. It corresponds to what Neale and Kendler (1995) termed 'multiformity'. In the case of ASD and SLI, this would mean that having a causal trait for one disorder could lead to a clinical picture resembling the other disorder. Figure 5 shows one version of phenomimicry where a child with ASD develops language deficits similar to those in SLI. The converse is also possible: a child with SLI might develop a clinical picture resembling ASD, perhaps because social interaction is difficult and stressful. Note that a phenomimicry account essentially is a variant of the traditional model (Fig. 1) and treats any phenotypic overlap between disorders as superficial. An account similar to that shown in Fig. 5 was proposed by Williams et al. (2008), who argued that apparently similar language deficits in SLI and ASD had different underlying causes. They argued that, although some children with ASD made errors in using inflectional endings on verbs, the types of errors differed from those seen in SLI. In a similar vein, Whitehouse et al. (2007) noted that the pattern of errors on a nonword repetition test was different in the two disorders, and argued that deficient phonological memory, indexed by disproportionate difficulty with long nonwords, may be implicated only in pure SLI. Problems for a phenomimicry account Although data on relatives appear compatible with a phenomimicry account, there are two problems for this explanation of comorbidity. First, it cannot readily explain the finding that CNTNAP2 is implicated as a risk factor for both SLI and ASD. Furthermore, we also have to explain why only a subset of individuals with ASD have SLI-like language problems, if such problems are a consequence of having ASD. An obvious possibility is that the likelihood of language problems increases with severity of ASD, as suggested by Whitehouse et al. (2007). However, this was not found in the larger study by Lindgren et al. (2009), nor in the epidemiologically-based sample studied by Loucas et al. (2008). The evidence for etiological overlap between ASD and SLI is therefore somewhat inconsistent. On the one hand, phenotypic similarities between language deficits in the two disorders, and findings that CNTNAP2 variants confer risk for both ASD and SLI, suggest a common causal pathway. On the other hand, qualitative differences in language phenotypes, coupled with the relatively spared language abilities in relatives of those with ASD points to distinct etiologies. A modified model: correlated risks with epistasis (CRE) A possible way of resolving the inconsistencies is to incorporate nonadditive interactions between genes in an etiological model. There are several lines of evidence that suggest that a genetic model of autism needs to include interactions between genes, rather than just additive effects. The first is an analysis by Pickles et al. (1995), who considered frequency of autism in relatives (twins and other family members) for probands with autism. If the etiology involves many genetic variants with additive effects, then the prediction would be that the rate of autism in DZ twins or sibs of those with autism should be around 50% of the rate seen in MZ twins. In fact, the rate in these first degree relatives is considerably less than that (with around 10% of siblings and DZ affected with ASD, compared to around 80% in MZ twins; Pickles et al. 1995). The authors concluded that nonadditive genetic influences must be implicated in the etiology. A similar conclusion was reached by Risch et al. (1999), who compared the proportion of alleles with a common identity by descent (IBD) in affected sib-pairs vs. discordant sib pairs. They found that there was a small increase in IBD-sharing for affected sib pairs across all 360 markers that they considered, rather than an effect confined to a few loci. They concluded that the etiology of autism involved a large number of loci, perhaps more than 15, and probably involves interactions between genes (i.e., epistasis) as well as additive effects. A third line of evidence comes from consideration of the functional networks in which genes are involved; Bill and Geschwind (2009) noted that many autism susceptibility genes are involved in the same pathways, suggesting possible interactions between proteins in signalling pathways. The models in Figs. 1 and 2 are derived from the classic ACE model, which assumes only additive influences. The question then arises as to whether modifying the model to allow for nonadditive effects in the form of gene-gene (G 9 G) interaction might accommodate findings such as those of Lindgren et al. (2009), given that one signature of epistasis is reduced similarity between probands and their first degree relatives. The simulation was therefore modified to incorporate interaction between risk genes for SLI and ASD. Figure 6 shows one specific version of the model in which presence of autism risk factors enhances the impact of a risk factor for LI, but it would be logically possible to have the opposite situation (impact of autistic risk factor increased if LI risk factors present), or epistatic interaction affecting both language and ASD traits. All Phenomimicry model: the phenotype of autism spectrum disorder (ASD) can lead to language impairment (LI), but the resemblance with specific language impairment (SLI) is superficial and those with ASD?LI do not have risk factors for SLI these options can be explored using the simulation; the one shown in Fig. 6 is selected here because it gives a reasonable account of observed data. Methods The Matlab script in the Appendix includes the option of specifying G 9 G interactions. Each gene is identified by number, and the user specifies a list of genes involved in each interaction. The first gene in the list has the effect of its risk genotype increased by a specified amount if and only if all the subsequent genes in the list have the risk genotype. For instance, if a G 9 G term is specified as [1 7 8] this means that an individual with a homozygous risk genotype (aa) for gene 1 will have the effect of that genotype amplified if a homozygous risk genotype is also present for genes 7 and 8. This means that even if two relatives have the risk genotype for language impairment, they may differ in terms of the effect of that genotype, because the added effect depends on the presence of a constellation of genotypes on other (ASD) genes. The probability of relatives sharing such a constellation decreases with the number of genes involved in epistasis. Table 1 shows the categorical results for diagnoses in relatives in relation to proband diagnosis for a model where there are 10 genes in total, one of which is pleiotropic, and has its impact doubled when its homozygous risk genotype (aa) occurs in combination with risk genotypes of any one of two pairs of ASD genes. It can be seen that relatives of comorbid ASD?LI probands now have a reduced rate of language impairment compared with the CAR model, and mean scores of relatives on the language liability measure are also less impaired. The pattern of results is thus more compatible with empirical data than was seen with the CAR model. The relatives of comorbid cases still show more language impairment than relatives of unaffected individuals; however, the relatively small increase in rates of LI over population prevalence, and more modest deficit on the language trait would only be detectable with a large sample. Although the overall pattern of affectedness in relatives is similar to that of Lindgren et al. (2009), the rates of language impairment in relatives are lower in the CRE model. Thus, the percentages of relatives with pure LI are 21%, 6% and 14% for the SLI, ASD and ASD?LI probands in the CRE model, compared with 52%, 12% and 27% for the sample of Lindgren et al. To some extent the discrepancy may reflect the fact that the criterion of LI used by Lindgren et al. was a language score of 1 SD below average on one of two measures, whereas the model cutoff was at 1.5 SD. Re-running the CRE model with cutoff for LI of -1 SD improved the fit, giving rates of affected relatives in SLI, ASD and ASD?LI of 34%, 10% and 25%, but the observed rate in SLI relatives was still higher than predicted by the model. Note, however, that the Lindgren et al. sample included several relatives from each family, and so rates could be inflated if there were clustering of affectedness within families, either because of assortative mating, or shared genes. In addition, affectedness in siblings was determined on the basis of a low score on either one of two tests, whereas in parents, a single test (nonword repetition) was used. This complicates comparisons of absolute frequencies with simulated data. Another inconsistency with data from Lindgren et al. (2009) in ASD?LI cases than SLI for some receptive language measures (Loucas et al. 2008;Rapin and Dunn 2003). The simulation was re-run with different values specified for G 9 G interactions, number of genes, and the size of epistatic effect. Examples of outputs are shown in Supplementary Material. Including more than two ASD genes in interaction with a LI gene makes outcomes of relatives less similar to the proband, but also makes it less likely that conditions for epistasis will be met, and so has little impact unless the size of epistatic effect is large. Altering the number of genes did not, in general, affect outcomes, unless the proportion of genes involved in epistasis was too low to exert much influence on overall trait scores. Increasing the size of epistatic effect decreased similarity between relatives, but large values induced strong skew, and sometimes bimodality, in the data. Results The skew in liability trait markers induced by epistasis is of potential interest for those who aim to detect epistasis using association analysis. Figure 7 shows boxplots of phenotypic scores on SLI and ASD liability markers for aa, aA and AA genotypes for each gene in the simulation described above, based on 5000 probands (to give a sample size that is in the range of realism). A plot shown in red denotes that the mean on the trait for aa genotype is significantly lower than for other genotypes on t-test, and magenta indicates that the mean for the aA genotype is lower than for AA. This kind of comparison relates to the statistical testing for association for a quantitative trait that is performed in conventional molecular genetic analyses. As expected, the left-hand set of genes (1-5) show differences between genotypes for the language trait, and the right-hand set (6-10) as well as the pleiotropic gene (1) show differences between genotypes for the ASD trait. Gene 4 gives a false positive result on the ASD trait, which disappears if a larger sample is taken. Overall, the t-test comparisons show that if we simply compare mean values of language trait scores for different genotypes for genes 6-10, which interact with gene 1, we are unlikely to detect an effect of an ASD gene on a language phenotype, with gene 8 being the only one to show a significant (though very small) effect. This confirms the low power of means comparisons for detecting epistasis. It is striking, however, that the distributions of phenotype scores are more skewed for risk genotypes that enter into G 9 G interactions. Bold lines denote cases where the variance on a trait for the aa genotype is significantly greater than for other genotypes on F-test. It is evident from inspection that this is a feature that characterises just those genes that are implicated in epistasis. This is because the genotype usually has no effect on language ability, but occasionally exerts a large effect, when occurring in the context of a set of other risk genotypes. This skew is distinctive compared with that seen for the other genotypes of this gene, and for distributions of genotypes of other genes. If this kind of G 9 G interaction is in play, then significant differences in variances of phenotypic trait scores between genotypes could provide evidence that a gene may be implicated in epistasis. General discussion Although ASD and SLI have traditionally been regarded as distinct disorders, they often involve similar language deficits, raising the question of whether this is merely a superficial resemblance, or indicative of a deeper similarity, with overlap in etiology. Three models of etiology were considered. The first, of CAR, was simulated to test its predictions. The simulation confirmed that a model of correlated additive genetic risks can explain the relatively high rate of comorbid ASD?LI cases, but it does not predict observed data showing that relatives of people with ASD?LI tend to do better than relatives of those with pure SLI on language measures. The second model, in terms of 'phenomimicry', could account both for comorbidity and the patterns of deficit in relatives, but is unable to explain why the CNTNAP2 gene has been found, in independent samples, to be associated with both ASD and SLI. It also leaves unexplained why only a subset of those with ASD have language difficulties resembling SLI. The final model was a modified version of the CAR model that incorporated G 9 G interaction. With one pleiotropic gene, whose impact was enhanced when a risk genotype occurred in the context of ASD risk genotypes, the model gave a pattern of results more in line with observed findings. In particular, the model could account for (a) comorbidity of ASD?LI at above chance levels, (b) similar or more severe levels of language impairment in ASD?LI as in SLI probands, while at the same time predicting (c) higher rates of language impairment in relatives of SLI cases than in relatives of ASD?LI cases. Of course, the fact that a simulation can fit a pattern seen in observed data does not mean that the model is correct. Phenomimicry could also be implicated: we need more studies of qualitative aspects of language phenotypes in ASD and SLI to test this hypothesis convincingly. Other mechanisms, such as gene-environment interaction or assortative mating, could also be involved. Nevertheless, the simulation program used here showed that incorporating epistasis allows us to retain a model that postulates overlapping genetic etiology for ASD and SLI, in line with the molecular genetic findings on CNTNAP2. The G 9 G interaction reduces the correlation between probands and first degree relatives, and so can accommodate the result observed by Lindgren et al. (2009) whereby relatives of those with comorbid ASD?LI were less impaired on the language trait than relatives of pure LI cases, even though the comorbid probands themselves were at least as impaired as SLI cases on language measures. Although a model with G 9 G is less parsimonious than the basic CAR model, it is compatible with evidence from other sources that point to epistasis being implicated in the etiology of complex disorders in general (Carlborg and Haley 2004) and in ASD in particular (Bill and Geschwind 2009;Pickles et al. 1995;Risch et al. 1999). The model not only provides a better fit to the observed data on relatives; it also suggests a way to identify genes that are involved in epistatic interactions. The CRE model included a single gene whose effect was magnified when its risk genotype co-occurred with risk genotypes on other genes. When this condition was met, this gene had a substantial impact on the pattern of observed data, introducing a skew in the tail of the distribution of liability markers. Nevertheless, when effect size is measured simply by comparing overall liability marker scores for those with the risk and nonrisk versions of the gene, the effect sizes of the interacting genes were small. Importantly, as shown in Fig. 7, the genes involved in epistatic interaction could, however, be distinguished from other genes in terms of the shape of the distributions of liability marker scores for risk and nonrisk genotypes. It follows that if such mechanisms do operate, then one could identify genes likely to be involved in epistasis by considering the distribution of liability marker scores associated with different allelic variants. In practice, non-normal distributions are often regarded as a nuisance to be corrected by removal of outliers, transformation or use of nonparametric tests (see also Bishop 2005). The sheer scale of information now available about the human genome means that one needs a strategy for prioritising which genes to analyse in order for association studies to be both statistically and methodologically tractable (Tabor et al. 2002). These simulations suggest that increased skew in phenotypic distributions for one genotype vs. others may be an indicator that a gene is likely to be involved in epistatic interactions. Acknowledgments The author is supported by a Principal Research Fellowship from the Wellcome Trust (ref 082498/Z/07/Z). The ideas in this paper were stimulated by discussions at a meeting organised by Gina Conti-Ramsden and funded by the Economic and Social Research Council, 'Language and Social Understanding in Developmental Fig. 7 Simulated data from correlated risks with epistasis model showing distributions of scores for 5000 probands on language trait (three leftmost plots) and ASD trait (three rightmost plots) for aa, aA and AA genotypes of 10 genes. For genes 1-5, the a allele confers risk for language impairment, and for genes 1 and 6-10 the a allele confers risk for ASD. G 9 G interaction is specified so that when the aa genotype is present for gene 1 in the context of either aa genotypes for genes 7 and 8, or for genes 9 and 10, its effect is doubled. Red denotes that the mean on a trait for aa genotype is significantly lower than for other genotypes on t-test, and magenta that the mean for the aA genotype is lower than for AA. Bold lines denote cases where the variance on a trait for the aa genotype is significantly greater than for other genotypes on F-test Disorders'. The author likes to thank Andrew Whitehouse and Courtenay Frazier Norbury for their insightful comments on an earlier draft of this paper. Open Access This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.	2014-10-01T00:00:00.000Z	2010-07-18T00:00:00.000	{ "year": 2010, "sha1": "d88a0e47e3f231da6c423486abf43c84910e24e7", "oa_license": "CCBYNC", "oa_url": "https://link.springer.com/content/pdf/10.1007/s10519-010-9381-x.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "d88a0e47e3f231da6c423486abf43c84910e24e7", "s2fieldsofstudy": [ "Linguistics", "Psychology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
233520864	pes2o/s2orc	v3-fos-license	Extended Reality in Educating the Next generation of health professionals COVID-19 has severely impacted health professions education and assessment has mostly shifted online. Major challenges remain especially about teaching-learning and assessment of clinical skills. Interacting with an individual online does not provide the range of information provided by an in-person meeting. There have been dramatic advances in computers and the internet in the last 60 years. Extended reality (XR) is all real-and virtual-combined environments and human-computer interactions generated by computers and wearable devices. The future physical world is likely to face multiple challenges. It is likely that human interactions and learning will increasingly occur in XR spaces. Three-dimensional holographic avatars and interacting and manipulating objects in XR spaces will become easier. Most medical education can occur in these spaces. Universities will invest substantially in these spaces. The safety of the physical world, the cost and usability will determine the extent of use of these learning spaces. INTRODUCTION COVID-19 continues to ravage the world. On 8th November 2020, there were 48.5 million confirmed cases and 1.2 million deaths attributed to the pandemic (1). The pandemic has severely impacted education. The deleterious effects on different aspects of health professions education range from suspension of classes, practical, small group sessions and stoppage of clinical postings (2,3). Education and assessment have mostly shifted online, and this has been extensively described in the literature. Major challenges remain especially about teaching-learning and assessment of clinical skills. Online learning has played a vital role in sustaining student learning during the pandemic preventing a total collapse of education. LEARNING ONLINE However, concerns remain that online learning and interactions may not capture ABsTRACT COVID-19 has severely impacted health professions education and assessment has mostly shifted online. Major challenges remain especially about teaching-learning and assessment of clinical skills. Interacting with an individual online does not provide the range of information provided by an inperson meeting. There have been dramatic advances in computers and the internet in the last 60 years. Extended reality (XR) is all real-and virtual-combined environments and human-computer interactions generated by computers and wearable devices. The future physical world is likely to face multiple challenges. It is likely that human interactions and learning will increasingly occur in XR spaces. Three-dimensional holographic avatars and interacting and manipulating objects in XR spaces will become easier. Most medical education can occur in these spaces. Universities will invest substantially in these spaces. The safety of the physical world, the cost and usability will determine the extent of use of these learning spaces. the nuances of non-verbal communication and in-person interactions (4). One of the major challenges is that the computer screen provides only an image of the person's face and upper body. We are relying exclusively on voice, facial and upper body gestures in understanding the other person. Regarding the student-teacher interaction, this raises several problems. As teachers, we may have felt weird speaking to a microphone in front of a webcam either during a live teaching session or while recording a session for later student viewing. We are used to an audience and during an online session, we cannot always see the audience. Keywords The principle of focused interaction concentrates on each person noticing how s/he is being experienced by other people. We use what we see to make decisions on how to react and about the future course of action when we are meeting in person. Another challenge with online interactions is peripheral participation which plays an important role in educating health professionals. Newcomers often start at the periphery and slowly develop "who knows what" knowledge directories and move slowly towards the centre of the profession. In-person informal interactions play an important role in developing the knowledge directory and in increasing participation. DRAMATIC ADVANCES IN COMPUTERS AND THE INTERNET Computers have been in use since the 1960s and Moore's law which states that the processing power of computers will double every two years has mostly held true (5). There has been a trillion-fold increase in the processing power of computers from 1956 to 2015 (6). The Apollo Guidance Computer that got man to the moon had less processing power than today's smartphones. So, with the steady increase in processing power computer-created worlds will become richer and more detailed. The physical world can be created down to the minute detail in these virtual worlds and the real world can even be modified. Accessing these worlds, manipulating and "living" in them will require increasing internet speeds and bandwidth. Internet speed has steadily increased from about 50 kilobytes per second in 1980 to around 500 megabytes per second or 1 gigabyte per second today (7). Network 2030 envisages a massive increase in our capabilities with the internet of things, holographic avatars, and a massive increase in speed and bandwidth (8). EXTENDED REALITY Extended reality (XR) refers to all realand virtual-combined environments and human-computer interactions generated by computers and wearables (9). XR is an umbrella term for all immersive technologies including augmented reality (AR), virtual reality (VR), mixed reality (MR) and others that have yet to be developed. Virtual information and objects are overlaid on the real world in AR. In VR, users are fully immersed in a simulated virtual environment, while in MR real-world and virtual objects can co-exist and interact with each other. We will use the term XR space to refer to a computer-generated space combining real-world and virtual objects, keeping in consideration that in the next two decades newer methods of creating these spaces may be developed. THE FUTURE PHYSICAL WORLD The future world is likely to have several challenges. The COVID-19 pandemic continues and there are predictions of more frequent pandemics with rising global temperatures. Increasing population density and destruction of animal habitats make the transmission of diseases between animals and humans and between humans easier. Climate change might make a large part of the world difficult to live in. Storms, wildfires, droughts and floods will become more frequent (10). In the future with increasing internet bandwidth and the decreasing cost of computing power, many human interactions will likely take place in computer-generated spaces. The current methods of producing XR are cumbersome though displays are getting lighter. In the future, creating an XR space may be easier and we can more easily flip between the real and the XR space. We could create them on a smaller scale to save space and cost. AVATARS Currently, we depend on "avatars" to represent us in virtual worlds, and though steadily improving, these are still clunky and difficult to manipulate. Holographic avatars, a three-dimensional representation of one's body placed in virtual space will offer a much better representation. Holographic telepresence projects full-motion, realistic, three-dimensional images in real-time (11). People can interact, learn and work in the virtual world just like they do in the physical one. XR SPACES How common XR becomes and the extent of their use in work, education and recreation may depend on the state of the physical world and on improvements in the technology to create and interact with XR. If the current COVID-19 pandemic becomes endemic and other frequent pandemics happen, then there will be a big push towards XR spaces. I believe a proportion of our teaching and learning will happen in XR spaces and we will continue to interact in the real world when required. We face challenges in manipulating objects in the virtual world and obtaining the sensation of touch which plays an important role in constructing reality. Researchers are working on several ways to deliver the sensation of touch. Electronic skin with artificial sensors can enable us to manipulate real and even virtual objects (12). An article (12) mentions the creation of highly compliant magnetosensitive skins that were used by wearers to manipulate virtual knobs and virtual dialling pads. Over millions of years, we have interacted with people face-to-face in the real world and not as flat two-dimensional images on computer screens and other devices. The cost will be an important factor in determining how many spaces we can create and support. With steady advances in computer technology, we believe soon it may become cheaper in many situations to create an XR space rather than a detailed real one. XR IN MEDICAL EDUCATION In medical education online classes, small group sessions, virtual labs and problembased learning can take place online in XR spaces. It is highly likely that a new set of interactive learning tasks will be created. Examples may range from constructing an artificial joint to creating a mitochondrion. As technology advances, these tasks can become richer and more complex. With more ability to receive tactile and visuospatial sensations, and manipulate objects in virtual spaces simulation can shift online and students can learn to start an intravenous line and insert an airway in an XR space. A simulated patient can be examined and feedback will be provided. XR spaces can overcome many of the challenges in present-day online interactions. Medical schools and other institutions may have to decide on how much of their learning and interactions should happen in the real world and how much in XR ones. Most institutions will continue to have physical spaces where faculty, students and other staff can meet and interact face-to-face. The nature of these XR spaces and how they will be regulated are important issues to work out. We are likely to interact with computers using voice and touch rather than through a keyboard as we do now. Three-dimensional multi-user virtual worlds can connect multidisciplinary learning communities in which users (students and instructors) can learn together more easily (13). We always believed our physical environments will become better, richer and safer with the passage of time and the development of technology and resources. The COVID-19 pandemic may lead us to question this belief and make resources available towards learning and working in XR spaces. Economics, physical safety and the advance in technology will have an important role in determining the extent of XR spaces use. Careful thought and consideration should be given to this future development in education, living and working. CONCLUSION XR learning spaces will play an increasingly important role in health professions education. Rapid increases in computing power, internet bandwidth and speed, holographic telepresence, the ability to deliver the sensation of touch and to manipulate objects in XR spaces will make the experience more realistic and common. Regulation of these spaces will be an important issue. Economics, physical safety, and the advance in technology will have an important role in determining the extent of their use.	2021-05-04T22:06:29.989Z	2021-03-31T00:00:00.000	{ "year": 2021, "sha1": "482b4d11b6510cdd46f9a4af6fa0eb3145780771", "oa_license": null, "oa_url": "https://doi.org/10.21315/eimj2021.13.1.8", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "b0ee5082e03fafbe075f616a382d56f6c2ec5e63", "s2fieldsofstudy": [ "Education" ], "extfieldsofstudy": [ "Sociology" ] }
269049518	pes2o/s2orc	v3-fos-license	Molecular maps of synovial cells in inflammatory arthritis using an optimized synovial tissue dissociation protocol Summary In this study, we optimized the dissociation of synovial tissue biopsies for single-cell omics studies and created a single-cell atlas of human synovium in inflammatory arthritis. The optimized protocol allowed consistent isolation of highly viable cells from tiny fresh synovial biopsies, minimizing the synovial biopsy drop-out rate. The synovium scRNA-seq atlas contained over 100,000 unsorted synovial cells from 25 synovial tissues affected by inflammatory arthritis, including 16 structural, 11 lymphoid, and 15 myeloid cell clusters. This synovial cell map expanded the diversity of synovial cell types/states, detected synovial neutrophils, and broadened synovial endothelial cell classification. We revealed tissue-resident macrophage subsets with proposed matrix-sensing (FOLR2+COLEC12high) and iron-recycling (LYVE1+SLC40A1+) activities and identified fibroblast subsets with proposed functions in cartilage breakdown (SOD2highSAA1+SAA2+SDC4+) and extracellular matrix remodeling (SERPINE1+COL5A3+LOXL2+). Our study offers an efficient synovium dissociation method and a reference scRNA-seq resource, that advances the current understanding of synovial cell heterogeneity in inflammatory arthritis. . Representative synovial biopsy images from a patient with psoriatic arthritis.The upper (lower) whisker extends from the hinge to the largest (smallest) value no further than 1.5 * interquartile range from the hinge.Individual dots represent data from different samples.Our data and data from Stephenson W et al.S2 is derived from unsorted synovial cells, while Wei K et al.S3 dataset includes sorted CD45 neg CD235 neg synovial fibroblasts, pericytes/mural cells and synovial endothelial cells. SUPPLEMENTAL TABLES AND LEGENDS Figure S18 Figure S2 . Figure S2.Quality control of integrative protocol scRNA-seq analysis of 18 human synovial biopsies of patients with inflammatory arthritis.Related to Fig. 2. a) The number of genes vs. total counts per sample coloured by filter (left) and percentage mitochondrial counts (right).b) The number of cells per sample (left) and per protocol (right), coloured by the number of genes (top) and by the number of genes in at least 1% of cells (bottom).c) Distribution of counts per protocol (left) and per sample (right).d) Distribution of the number of genes per protocol (left) and per sample (right). Figure S3 . Figure S3.UMAPs of integrated scRNA-seq data from integrative protocol analysis showing main synovial cell types.Related to Fig. 3. Data derive from 18 synovial tissue biopsies of patients with inflammatory arthritis and are coloured by top marker genes of major synovial cell populations including a) T cells, B cells and plasmablasts; b) macrophages and plasmacytoid dendritic cells; c) myeloid dendritic cells, mast cells and neutrophils; d) endothelial cells, pericytes/mural cells and synovial fibroblasts. Figure S4 . Figure S4.UMAPs of scRNA-seq data from integrative protocol analysis showing synovial cell subsets.Related to Fig. 3. Data from 18 synovial tissue biopsies of patients with inflammatory arthritis.a) T cells/NK cells, color by the expression of key marker genes CD3, CD4, CD8, NKG7 or by the geometric Figure S5 : Figure S5: The heatmap of major neutrophil genes, detected in the integrative protocol scRNA-seq analysis.Related to Fig. 3. Data derive from 18 synovial tissue biopsies of patients with inflammatory arthritis.Expressions are aggregated by sample and cell type. Figure Figure S6 Multicolor flow cytometry analysis of freshly isolated synovial tissue cells using protocol 2. Related to Fig. 3. Synovial cells were dissociated from a knee synovial biopsy of a patient with septic oligoarthritis.The cells were labeled with Zombie Green™ Fixable Viability Dye, APC anti-human CD45, BV421™ anti-human CD11b, PE anti-human CD64, BV785™ anti-human CD19, and Alexa Fluor® 700 antihuman CD3.a) Total synovial cells, excluding cell debris.b) Gating single cells and doublet exclusion.c) Gating live cells and excluding dead cells, with a high Zombie Green™ Fixable Viability Dye signal.d) Live synovial cells gated for CD45+ leukocytes (violet) and CD45-structural cells (green).e) Gating of CD45+ leukocytes into CD11b+ CD64+ macrophages.f) CD11b -CD64-cells were gated further into CD3+ CD19-T cells and CD3-CD19+ B cells.g) A total of 10121 events were analyzed, among which 74.2% were gated as synovial cells, consisting of 93.2% of single cells with 80.2% viability. Figure Figure S7 Spectral flow cytometry analysis of freshly isolated synovial tissue cells using protocol 2. Related to Fig. 3. Synovial cells were dissociated from a biopsied wrist synovial tissue of a patient with early RA.Fixed unsorted synovial cells were labeled with a cocktail of antibodies targeting surface leukocyte (CD45), myeloid (CD14), lymphocyte (CD3, CD4, CD8, CD19) and structural cell (CD31, PDPN) markers followed by analysis on spectral analyser Sony ID7000.a) Division of synovial cells based on the expression of CD45.CD45+ leukocytes included CD14+ macrophages, CD19+ B cells and CD3+ T lymphocytes, dividing further into CD4+ and CD8+ T cells.CD45neg synovial cells contained two structural cell clusters, including PECAM1+ endothelial cells and PDPN+ synovial fibroblasts.There was a considerable amount of cell debris, which might have been attributed to the sample shipment from Portugal to Germany.b) Unstained synovial cells were used for setting the flow cytometry gates. Figure S8 . Figure S8.Quality control of scRNA-seq analysis of the single cell reference map of fresh human synovium from 25 synovial biopsies of patients with inflammatory arthritis.Related to Figs. 4-8.a) A cell-level summary of the total number of counts, the number of detected genes and the percentage of mitochondrial counts.The number of genes vs. total counts per sample coloured by filter (left) and percentage mitochondrial counts (right).b) Sample summary statistics with the number of cells and number of detected genes after filtering of low-quality cells.The number of cells per sample is coloured by the number of genes (top) and by the number of genes in at least 1% of cells (bottom).c) Distribution of counts per sample.d) Distribution of the number of genes per synovial tissue sample. Figure S9 . Figure S9.Integrated scRNA-seq dataset from 25 synovial tissue samples from patients with inflammatory arthritis.Related to Figs. 4-8.a) UMAP of annotated main synovial cell populations coloured by main cell type.b) Bar plots of relative abundances of main cell types per sample coloured by main cell type.c) The variability of the proportion of cell types across patient synovial tissues.The box plot visualises 5 summary statistics: the median; two hinges, corresponding to the first and the third quartiles; two whiskers.The upper (lower) whisker extends from the hinge to the largest (smallest) value no further than 1.5 * interquartile range from the hinge.Individual dots represent data from different samples.See STAR Methods for details. Figure S10 . Figure S10.A heatmap of top marker genes of the main synovial cell types identified in the integrated scRNA-seq dataset.Related to Figs. 4-8.Data are derived from 25 synovial tissue samples from patients with inflammatory arthritis (see STAR Methods for details).Expressions are aggregated by sample and cell type. Figure S11 . Figure S11.T cell, natural killer (NK) cell and innate lymphoid cell sub clustering.Related to Fig. 4. The integrated scRNA-seq dataset consists of scRNA-seq profiles from 25 synovial tissue samples from patients with different types of inflammatory arthritis (see STAR Methods for details).a) UMAPs showing the small population of proliferating TOP2A+ CENPF+ T cells (cluster 2).b) A heatmap of the cluster-enriched genes and marker genes in synovial T cell, NK cell and innate lymphoid clusters in patients with inflammatory arthritides.Expressions are aggregated by sample and cell type.c) UMAPs demonstrating the expression of FOXP3, PCDC1 and CXCL13 genes in the TIGIT+ CTLA4+ T cells (cluster Figure S12 . Figure S12.Synovial fibroblast sub clustering.Related to Fig. 5.The integrated dataset from 25 synovial biopsies of patients with inflammatory arthritis (see STAR Methods for details) with a) violin plots showing the expression (log counts) of the lining marker PRG4 and the sublining marker THY1 across the seven synovial fibroblast clusters, and b) UMAPs showing the small population of proliferating TOP2A+CENPF+ synovial fibroblasts, co-clustering with the cluster four SERPINE1+ COL5A3+ synovial fibroblasts (see also Fig.5a, d). Figure Figure S13.A heatmap of the top cluster genes and known markers of synovial fibroblast clusters detected in the synovium of patients with inflammatory arthritides.Related to Fig. 5.The integrated scRNA-seq dataset consists of 25 synovial tissue samples from patients with different types of inflammatory arthritis (see STAR Methods for details).Expressions are aggregated by sample and cell type. Figure S14 .Fig. 5 . Figure S14.The expression of a selected set of genes in synovial fibroblast sub clustering.Related to Fig. 5.The integrated dataset from 25 synovial biopsies of patients with inflammatory arthritis (see STAR Methods for details) with a) UMAPs showing the highest enriched expression of the genes involved in MHCII class mediated antigen presentation in HLA-DRA high synovial fibroblasts (cluster 7), and b) UMAPs showing the enriched expression of IL6 and NOTCH3 genes primarily in sublining GGT5+ synovial fibroblasts (clusters 6). Figure S15 . Figure S15.Synovial macrophage and myeloid dendritic (DC) cell sub clustering.Related to Fig. 6.A heatmap of the top cluster genes and known marker genes of synovial macrophage and DK subclusters, detected in the integrated scRNA-seq dataset from 25 synovial tissue samples of patients with different types of inflammatory arthritis (see STAR Methods for details).Expressions are aggregated by sample and cell type. Figure Figure S16 Trajectory analysis of synovial endothelial cells.Related to Fig. 7. Inferred pseudotime versus gene expression in synovial endothelial cells.Start of the trajectory are GJA4+ CLDN5+ arterial Figure S17 . Figure S17.Endothelial cell sub clustering.Related to Fig. 7. Heatmap of the top cluster genes and known markers of synovial vascular and lymphatic endothelial cell subclusters detected in the integrated scRNA-seq dataset from 25 synovial tissue samples of patients with different types of inflammatory arthritis (see STAR Methods for details).Expressions are aggregated by sample and cell type. Figure S18 . Figure S18.Integration of our synovial scRNA-seq from fresh synovium with publicly available synovial scRNA-seq datasets.Related to Fig. 9.We integrated our scRNA-seq data from 25 synovial tissue samples of patients with inflammatory arthritis with published data from Stephenson W and colleagues S2 and Wei K et colleagues S3 (see STAR Methods for details).a) UMAP of annotated main synovial cell populations coloured by main cell type across the three studies.b) Bar plots of relative abundances of main cell types per sample coloured by main cell types across the 3 studies.c) The variability of the proportion of cell types across patient synovia in the 3 studies.The box plot visualises 5 summary statistics: the median; two hinges, corresponding to the first and the third quartiles; two whiskers. Table S1 Detection of neutrophils in 18 synovial tissue biopsies included in integrative protocol analysis using histology and scRNA-seq analyses. Related to Figure 2. For the detection of neutrophils in formalin-fixed, paraffin-embedded synovial biopsy fragments, tissue sections were labeled with CD15 antibodies (see STAR Methods for details).In scRNA-seq data, the quantity of neutrophils was estimated by calculating the proportion of neutrophils per sample.Data pertains to 18 samples included in integrative protocol analysis.Y: yes, N: no. Table S2 . Demographic, clinical, therapy, histology and cell characteristics for 2 synovial tissue biopsy samples from patients with arthritis, included in the proof of principle flow cytometry analyses. Related to Figures 6 , 7. F: female, M: male, *Automatic cell counting with the Luna-FL Dual Fluorescence Cell Counter.	2024-04-12T15:09:05.420Z	2024-04-01T00:00:00.000	{ "year": 2024, "sha1": "4ef3a74471b1b4ebfaebfb58605b3f761f6fe4d5", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1016/j.isci.2024.109707", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "4620e71b95b164b32441fa5706f42e66cedfb0d8", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [] }
220277213	pes2o/s2orc	v3-fos-license	Primary gas discharge transducer and its application The aim of this study is to develop a high-speed fault transducer and, based on it, to create an automated system for monitoring and repairing faulty insulation of winding wires in electrical equipment. The fault transducer under development is based on a gas discharge process that occurs between a high-voltage electrode and a grounded wire conductor in a damaged insulation segment. Oscillographic measurements show that there is a complex signal that appears on the fault transducer when a damaged insulation segment is passed through it. The basic patterns of electrodeposition of dielectric films onto metals were determined. The conditions of this process were optimized. The features of the above-mentioned signal that can be used to monitor the number of faults and lengths of damaged segments, as well as the features of the deposition of dielectric films onto metals were justified. Based on these features, a faulty wire insulation repair process was developed and can be controlled. Introduction At present, fault monitoring of enamel wire insulation in Russia is regulated by GOST IEC 60851-5-2011. According to this document, the integrity of insulation is expressed in terms of the number of point faults detected by an electrical testing device per wire of a certain length. Point faults for wires with a rated conductor diameter of up to and including 0.050 mm are identified at a low voltage using a transducer consisting of two felt plates immersed in a solution of aqueous sodium sulfate Na 2 SO 4 (30 g/L), between which a controlled wire is passed at a rate of (275 ± 25) mm/s. A test DC voltage (50 ± 3) V is applied between the wire conductor and the solution connected to the electrical circuit. Point faults are recorded by an appropriate relay with a counter. A disadvantage of this monitoring technique is low sensitivity of the transducer and contamination of the insulation with electrolyte during the monitoring. This makes it impossible to test all wires completely, and they are tested selectively on individual segments and from different coils. For wires with a conductor diameter of 0.050 to 0.25 mm, a transducer electrode made in the form of two rollers is used. A controlled wire is passed through the transducer formed by four rollers, two of which are guides and the other two are electrodes. For wires with a diameter of 0.25 to 1.60 mm, this transducer cannot be used. Instead of it, one high-voltage roller electrode of a larger diameter is used. In this monitoring procedure, the wire is repeatedly bent. This leads to high mechanical stresses on the wire insulation on the electrode side, which not only weakens mechanical and electrical strength of the insulation in a controlled wire, but also causes additional faults in the insulation. The fault monitoring techniques described above are not able to determine the length of damaged segments, another indicator of insulation failure which is equally important as the number of faults. Another disadvantage of the monitoring technique described in GOST IEC 60851-5-2011 is that it is applied selectively to a segment of the wire from a coil that has been randomly chosen from a batch and, therefore, leaves the bulk of the wire on the chosen coil untested. The same applies to other coils in the batch that have not been chosen for random inspection, thereby reducing its reliability. Due to the low quality of enamel insulation, electric motors may fail at any time [1][2][3][4][5][6][7], potentially leading to an accident. Rejection of faulty wires and their use in products lead to significant economic losses (since such expensive materials as enamel, wire, etc. are wasted) and unreasonable recycling costs. The studies [8][9][10][11] are dedicated to this problem. The aim of this study is to develop a high-speed fault transducer and, based on it, to create an automated system for monitoring and repairing faulty insulation of winding wires in electrical equipment. Research Materials and Methods The tests were performed in two stages. At the first stage, the gas discharge fault transducer was tested. The design of the transducer is justified in detail in [12,13]. The transducer consists of two cylindrical electrodes that are contiguous to each other along the generatrix and mounted on axes through bearings. The bearing axes are fixed on movable levers (rockers) that allow roller electrodes to make vertical movements in synchrony with wire vibrations. Electrodes are pressed to each other by blade springs. Voltage is applied to electrodes through sliding contacts. For testing purpose, a controlled wire was passed through the above-mentioned electrodes pressed to each other. The signal generated on the transducer when a damaged segment of wire insulation was passed through it, was recorded with one oscilloscope beam from the low-voltage arm of the voltage divider connected to the transducer. The discharge current was measured on the limiting resistance in the discharge circuit of the transducer. In order to repair faulty wire insulation, it was first necessary to develop a technique for applying an insulating film to a damaged segment. This task was solved at the second stage of the study. The second stage involved searching for and optimizing components that, when combined, would give electrophoretic properties to a PE-939 lacquer designed for enameling PETV grade wires with a temperature index of 140  180°С. It was empirically found that, in case of the film-forming lacquer PE-939, electrophoresis occurs when using dioxane as a solvent and 1% NH 4 OH as an electrolyte [14]. Results A typical oscillogram of the signal from the transducer when passing a damaged segment of the wire insulation is shown in Figure 1. The oscillogram (Fig. 1) was taken at a negative potential U = -2 kV on the electrode rollers relative to the grounded conductor at wire speed V = 3.07 m/s. Two times are highlighted on the oscillogram: t p = 4.1  10 -3 s and t с = 3.5  10 -3 s. The pattern of the oscillogram (Fig. 1) can be explained as follows. The ignition of the corona discharge upon approaching a fault in the wire insulation can be explained as follows. When the fault is moved towards the transducer, the distance between them decreases and, due to this, the electric field strength increases. At a certain critical distance between the transducer and the fault and when the transducer voltage is equal to the initial corona onset voltage, a corona discharge is ignited in the damaged area. When the fault approaches the transducer, the electric field in the fault area increases, avalanche processes become more intense, and the corona discharge is converted into spark discharge. The capacitance of the circuit begins to discharge and leads to a voltage drop across this gap. The electric field strength in the damaged area falls to a certain value. The current rise rate in the gap decreases and the discharge goes out. In Figure 1, this process is represented by the first pulse on the current oscillogram. After extinction of the discharge, the capacitance of the discharge circuit begins to be charged from the power source. The voltage across the gap between the fault and the transducer increases. After the transducer voltage is restored to the initial value, spark reignition may not occur immediately. It occurs after a certain time  t, which varies significantly and depends on the wire speed, amplitude, period of transverse wire vibration, and many other factors. (On the oscillogram presented in Figure 1,  t= t pt с ). Therefore, when the fault approaches the transducer and leaves it, the spark rate is unstable, and the value  t varies significantly from fault to fault. When the transducer enters the transducer zone, the discharge stabilizes and, as the distance between the fault and the transducer decreases, subsequent spark ignitions occur at lower voltages than those upon approaching and leaving the fault. On the oscillogram, the residence time of the fault under the roller electrodes is indicated as t с . Estimates resulting from the experiments show that the optimal diameter of the transducer rollers should be in the range of 10 to 16 mm. The second stage of the studies was aimed at optimizing components of the electrophoretic composition. Experiments were carried out using the complete factorial design 2 k . The levels of variable factors are given in Table 1 After statistical processing of the obtained results and regression analysis, the following adequate regression equation was obtained: The planning matrix is given in Table 2. From equation (2), we can find the optimal ratio of the components n 1opt and n 2opt (3) By solving the system (3), we obtain n 1opt = 156.7 ml/L; n 2oпт = 333.3 ml/L. The dioxane concentration n 3opt = 500 ml/L. The gas discharge fault transducer and electrodeposition of dielectric films were used as a basis for the wire insulation monitoring and repair system, which is described in detail in the patent [14]. Conclusion The structure of the signal produced by the gas discharge fault transducer when passing a damaged segment of wire insulation through it was studied and presented in the paper. As shown, this signal has a complex structure. This structure was analyzed. It was shown how to single out a useful zone from this structure to exclude errors when calculating the number of faults and lengths of damaged segments. Concentrations of the components in the electrophoretic composition based on the PE-939 lacquer were justified and optimized. The developed primary fault transducer and the proposed electrodeposition technique for dielectric films can be used not only for designing automated wire repair system, but also for complete monitoring of wire insulations during enameling and production of windings. In addition, the composition and the electrodeposition technique can be used for other purposes, in particular, for wire enameling.	2019-04-16T13:27:26.456Z	2018-01-01T00:00:00.000	{ "year": 2018, "sha1": "4a169975d0d54555d148522ddbf2e7e35efe7086", "oa_license": "CCBY", "oa_url": "https://www.matec-conferences.org/articles/matecconf/pdf/2018/02/matecconf_yssip2017_03002.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "7b125b8800d83c7767ccf8739e58ba6b6f45c015", "s2fieldsofstudy": [ "Engineering", "Physics" ], "extfieldsofstudy": [] }
237937949	pes2o/s2orc	v3-fos-license	Exploitative Competition between Mountain Hare and Moose—Qualitative Effects on Hare Winter Forage? Simple Summary Mountain hares in Scandinavia are classified as Near Threatened in the Norwegian and Swedish Redlists assessing the risk of species extinction. This is due to a possible population decline witnessed during the last decades in Scandinavia. Competition between large herbivores such as moose, red deer, roe deer on one hand and hares on the other, is one of several hypotheses that has been put forward to explain this decline. In a cafeteria trial (providing several types of forage to determine food preference) we investigate the effects of previous moose winter foraging on the food selection (i.e., amount consumed, bites per minute and bitediameter) of downy birch and goat willow by captive hares. We find that hares do not differentiate among levels of previous moose foraging on downy birch but have larger bite diameters of goat willow earlier eaten on by moose, compared to plants not fed on by moose. Thus, effects of moose on hare winter food quality seem to be limited. We highlight the need for studies focusing on (1) effects of previous moose foraging using wild hares in a natural experimental design, and (2) effects of moose foraging on available hare food at a landscape scale during winter. Abstract Mountain hares (Lepus timidus) in Scandinavia are classified as Near Threatened in the Norwegian and Swedish Redlists. This is due to a possible population decline witnessed during the last decades in Scandinavia. Competition between large herbivores and mountain hares is one of several hypotheses that has been put forward to explain this decline. In a cafeteria trial we investigate the effects of previous moose (Alces alces) winter browsing on the food selection (i.e., biomass consumed, bites per minute and bitediameter) of downy birch (Betula pubescens) and goat willow (Salix caprea) by captive mountain hares. We find that mountain hares do not differentiate among previous browsing levels of downy birch, but have larger bite diameters of goat willow earlier browsed by moose, compared to non-browsed plants. Thus, effects of moose on mountain hare winter food quality seem to be limited. We highlight the need for studies focusing on (1) qualitative effects of moose browsing using wild mountain hares in a natural experimental design, and (2) quantitative effects of moose browsing on available mountain hare forage at a landscape scale during winter. Introduction In Scandinavia, mountain hares (Lepus timidus), have shown possible population declines over the past decades [1][2][3] (www.viltdata.se accessed on: 27 August 2021) and were classified as Near Threatened in the Norwegian Redlist in 2015 and again in 2021 (www.artsdatabanken.no accessed on: 27 August 2021), as well as the Swedish Redlist in 2020 (www.artfakta.se accessed on: 27 August 2021). Several hypotheses have been put forward to explain this decline including climate change, land use change, parasites, predation, and competition [3]. Climate change, color mismatch and increased predation is currently a topic of general research interest, and we have earlier shown that abundance of mountain hares is negatively affected by an interaction between reduced snow cover duration and abundance of generalist predators; red fox (Vulpes vulpes) and pine marten (Martes martes) [1]. However, the mountain hare may not only be negatively affected by top-down effects, but also bottom-up through interactions with other species from within the herbivore guild [4]. On a landscape scale, moose (Alces alces) is known to reduce abundance of deciduous trees e.g., [5,6], and mountain hares are negatively associated with moose presence at a habitat patch scale [7]. Thus, in areas of high density, moose may be an important competitor towards mountain hares [8]. Herbivore species are not expected to compete if they either differ in digestive systems [9], or in body size [10]. Hence, one would expect moose and mountain hares to have large ecological niche separation in terms of diet overlap and potential for food competition. However, moose [11] and mountain hare [12] consume several of the same plant species (i.e., birch (Betula spp., salix (Salix spp.) and aspen (Populus tremula), especially during winter. Herbivore species' food preferences vary along a gradient from bulk feeding of low nutritious forage to selective feeding of high-quality forage see [13], and references therein. On one hand, hindgut fermenters (such as the mountain hare) have less efficient digestive system compared to ruminants (such as the moose). Due to their relatively inefficient digestive systems, hind gut fermenters select plant quantity over quality. Small herbivores on the other hand, are in general more selective, targeting higher food quality compared to large-bodied herbivores [13]. Ironically, due to this interaction between body size and digestive systems-a large ruminant and a small hindgut fermenter may compete over food sources. Indeed, previous studies have shown mountain hares to be competitively inferior to larger herbivores such as the roe deer (Capreolus capreolus) [14]. Hulbert and Andersen [14] did not detect any feeding-height separation between the two species, however mountain hares switched to smaller bite diameters when in sympatry with roe deer, likely leading to higher concentrations of harmful plant secondary metabolites. The authors suggest that this switch may lead to reduced survival of mountain hares in the presence of roe deer. Large herbivores such as moose and roe deer may therefore limit mountain hare densities, especially in areas of high cervid density. These negative impacts may either be transferred through reduced forage quantity, i.e., deciduous winter browse [5,6] or forage quality through chemical changes in plants [14,15]. To investigate the latter competitive pathway of reduced food quality we conducted cafeteria trials with captive mountain hares, using downy birch (Betula pubescens) and goat willow (Salix caprea), two species which are consumed by moose and mountain hares in winter. For both moose and mountain hares, downy birch is the staple food, while goat willow is among the highly preferred plant species [11,12,15,16]. We predicted that mountain hares would prefer unbrowsed downy birch due to a lower concentration of plant secondary metabolites [17]. For goat willow we predicted that mountain hares would prefer previously browsed shoots, as the related tea-leaved willow (S. phylicifolia) is known to produce shoots with higher biomass and lower levels of chemical defense as a response to moose browsing [18]. Sample Collection This study was carried out in Østerdalen (61 • N, 11 • E), Southeast Norway along a gradient in moose density and browsing pressure. In January 2008, we collected downy birch and goat willow branches (approximately 30-40 cm long), after inset of winter dormancy but without signs of recent browsing. We here consider branches consisting of several years' growth, while shoots are one year's growth. All branches were collected from trees <2.5 m height. Any collected shoots with signs of current moose browsing were discarded. Shoot coloration and bark structure may be used to separate the most current growing season from previous years growing seasons. Browsing levels were estimated as a percentage of available shoots browsed during the previous browsing season. For downy birch, we collected branches without (0%), intermediate (30% < 60%), and high levels of previous moose browsing (>90%). We also collected goat willow without browsing (0%) and more than 30% previously browsed shoots. Browsing levels were determined based on morphology of the branches. After browsing, lateral shoots tend to form leaving a dry «stump» where the moose bit off the shoot last winter. We specifically targeted the mentioned percentage intervals as to avoid overlapping of categories-thus, this approach should be robust towards any slight over or underestimation of percentage browsing. We originally aimed for three categories of goat willow, like that of downy birch, but high moose densities and limited availability of goat willow prevented us from doing so. For the control (no browsing) we collected branches from two military facilities (Mil I and Mil II) within the Østerdalen valley, which were fenced more than 30 years before the experiment. The other treatments were collected just outside the two military facilities as well as two additional areas (Imsdalen and Kopppangskjølen) known to have high moose presence and browsing pressure in winter [19,20]. All four sites lie within the boreal forest and have similar forest composition dominated by Norway spruce (Picea abies), Scots pine (Pinus sylvestris), interspersed with downy birch, silver birch (B. pendula), Salix. spp., aspen, rowan (Sorbus aucuparia) and grey alder (Alnus incana). Samples were collected at these four sites depending on availability. Thus, for Imsdalen we had the categories "willow browsed", "birch intermediate", and "birch high". For Koppangskjølen we had "birch intermediate" and "birch high", For Mil I we had "birch no browsing", "willow no browsing", "birch intermediate" and "willow browsed". For Mil II we collected "birch no browsing", "willow no browsing", and "willow browsed". Cafeteria Test Fourteen captive mountain hares were housed in cages with ad lib. access to water, rabbit pellets, and branches from various deciduous trees. The mountain hares were part of a breeding facility for restocking natural populations. No permit was required under the Norwegian Food Safety Authority as the experiment was part of the natural feeding of the animals. Mountain hares were kept indoors in an unheated barn. For two days prior to the start of the experiment, mountain hares were fed downy birch and goat willow, in order to habituate them to the material to be used in the experiment. Mountain hares were deprived of food (12 h) over the night before the experiment started. We provided the mountain hares with bundles of branches, weighing approximately 100 g (±0.01 g), for 90 min. All 14 mountain hares were given all five categories/species of branches in randomized order, over a six-day period. Immediately after finishing the trial, we measured bite diameter and shoot base diameters (to the nearest 0.1 mm) of the remaining material using callipers. Bundles and all shoot residues on the cage floor were weighed and we calculated wet biomass consumed. We weighed control bundles of goat willow and downy birch not used in the experiment but treated similarly. Mean weight loss due to evaporation was 0.10 g and 0.2 %; thus, we regarded all weight loss as biomass consumed. Additionally, mountain hares were filmed during the cafeteria trial, and we registered time spent feeding, and number of bites. Some of the recordings failed to document all mountain hare activity during the trial and were thus excluded from further analysis. Statistical Analysis The dataset consisted of branches from four different areas, the means of response variables varied among the different areas (Table 1); however, the 95% CI overlapped for tree species as well as browsing levels. Thus, these differences were not considered for further analysis, but our results should reflect general patterns as the material was collected from several sites. For the two species goat willow and downy birch we ran separate mixed effects ANOVAs with grams consumed, bites per minute and bitediameter as response variables, and moose browsing intensity as explanatory variable. We included mountain hare individual as a random term, to account for individual variation among mountain hares. Analyses were performed in R version 4.0.4 [21]. Results For downy birch we found no effects on amount consumed ( Figure 1A), bites per minute ( Figure 1B) or bite diameter ( Figure 1C); thus, mountain hares did not differentiate among level of previous browsing from moose (Table 2). For goat willow we only found an effect of treatment on bite diameter ( Figure 1C), with mountain hares having larger bite diameter for goat willow previously browsed by moose ( Table 2). For the variables biomass consumed ( Figure 1A) and bites per minute ( Figure 1B) we found no effect of treatment (Table 2). Overall, biomass of goat willow consumed was higher than that of downy birch ( Figure 1A). and goat willow (dark bars) ±95% CI in relation to previous moose browsing. Birch no (0% browsed), Birch int (30% < 60% browsed), and Birch high (>90% browsed). Willow no (0% browsing) and Willow browsed (30% < browsed). Estimates from a mixed effects ANOVA with mountain hare individual as a random term. Note that goat willow and downy birch are run in different models due to differences in browsing categories but plotted together. Discussion For downy birch we found no effect of previous moose browsing on mountain hare preference. In a similar field-based study by Danell and Huss-Danell [17], mountain hares also did not differentiate between moderate and low previous moose browsing on birches (B. pubescens and B. pendula). For goat willow we found an effect on bite diameter only, where mountain hares have larger bite diameters on goat willow previously browsed by moose. Herbivores are known to select for large shoot base diameters [22], and our results and goat willow (dark bars) ±95% CI in relation to previous moose browsing. Birch no (0% browsed), Birch int (30% < 60% browsed), and Birch high (>90% browsed). Willow no (0% browsing) and Willow browsed (30% < browsed). Estimates from a mixed effects ANOVA with mountain hare individual as a random term. Note that goat willow and downy birch are run in different models due to differences in browsing categories but plotted together. Discussion For downy birch we found no effect of previous moose browsing on mountain hare preference. In a similar field-based study by Danell and Huss-Danell [17], mountain hares also did not differentiate between moderate and low previous moose browsing on birches (B. pubescens and B. pendula). For goat willow we found an effect on bite diameter only, where mountain hares have larger bite diameters on goat willow previously browsed by moose. Herbivores are known to select for large shoot base diameters [22], and our results corresponds to findings by Stolter (2008) in the closely related tea-leaved willow that is known to respond to moose winter browsing by producing thick shoots, which again is preferred by moose the following winter. Plants may respond to browsing by tolerance or avoidance. Avoidance of herbivory may either be escaping browsing altogether or physical and chemical defense [23]. Chemical defense may be constitutive (i.e., omnipresent) or induced as a response to damage by e.g., herbivores [23,24]. Regrowth after moose browsing on the tea-leaved willow produced shoots with lower concentration of plant secondary compounds (Stolter 2008). This counter-intuitive response led to a positive feedback loop, where moose rebrowsed tea-leaved-willow that were browsed during the previous browsing season (Stolter 2008). Although we did not investigate chemical composition in the treatments, it is possible that such a positive feedback loop with decreased levels of plant secondary compounds following moose browsing is the case also with goat willow in the current study. The lack of a strong effect of previous moose browsing in the current study could be due to several reasons: (1) The mountain hares used in the experiment are captive; thus, their food selection and preferences may not be reflecting that of wild mountain hares. (2) Many plant secondary compounds act as digestive inhibitors e.g., [25]. Thus, if the mountain hares of our study have unlimited access to high quality rabbit pellets, any response in the plants to moose browsing leading to increased concentration of digestive inhibitors may be swamped by the mountain hares' unlimited access to high quality food. Conclusions In recent years, several studies have suggested competition between lagomorphs and cervids. In Italy, European hare (L. europaeus) and roe deer are known to increase their diet overlap in winter [26]; however, they seem to reduce the degree of competition by having low spatial overlap on the landscape scale [27], similarly to that of moose and mountain hares in Northern Sweden [7]. From our results and those of others [17] we suggest that if there is a competitive relationship between moose and mountain hares, this is not likely caused by qualitative changes in mountain hare winter forage, but rather quantitative changes as a result of moose reducing available mountain hare winter forage in the landscape. Future studies should investigate (1) any qualitative effects (i.e., impacts on plant secondary compounds) of moose browsing using wild mountain hares in a natural experimental design, and (2) the possible alternative competitive route of quantitative effects of moose on available mountain hare forage at a landscape scale during winter. Funding: This study received no external funding. Inland Norway University of Applied Sciences covered publication costs. Institutional Review Board Statement: "The Food Safety Authority (i.e., the Norwegian Animal Research Authority (NARA) does not give formal approval of experiments involving no protocols other than daily feeding activity, where the handling of the animals are within the limits of the Norwegian animal welfare act. The current study by Pedersen and Pedersen («Exploitative competition between mountain hare and moose-qualitative effects on hare winter forage?») is not regarded as an animal experiment, and is thus exempted from approval" (2021/180183). Data Availability Statement: The data presented in this study are available on request from the corresponding author.	2021-09-28T05:24:15.679Z	2021-09-01T00:00:00.000	{ "year": 2021, "sha1": "5a138712e09531765a5c9cad63720af568810d9f", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2076-2615/11/9/2638/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "5a138712e09531765a5c9cad63720af568810d9f", "s2fieldsofstudy": [ "Environmental Science" ], "extfieldsofstudy": [ "Medicine" ] }
214519696	pes2o/s2orc	v3-fos-license	Classification of All-Rounders in the Game of ODI Cricket: Machine Learning Approach Player classification in the game of cricket is very important, as it helps the coach and the captain of the team to identify each player’s role in the team and assign responsibilities accordingly. The objective of this study is to classify all-rounders into one of the four categories in one day international (ODI) Cricket format and to accurately predict new all-rounders’. This study was conducted using a collection of 177 players and ten player-related performance indicators. The prediction was conducted using three machine learning classifiers , namely Naive Bayes (NB), k-nearest neighbours (kNN), and Random Forest (RF). According to the experimental outcomes, RF indicates significantly better prediction accuracy of 99.4%, than its counter parts. Introduction Cricket is considered as a bat and ball team game. The game has basically three formats, namely, the test cricket, one-day-international cricket (ODI), and T20. Test cricket, the longest format is regarded by experts of the game as the ultimate test of playing skills. An ODI cricket game is played for 300 legal deliveries (balls) per side, and the shortest format, T20 is played for 120 legal deliveries (balls) per side. A typical cricket team comprises of 11 players and the team batting first is identified by the outcome of tossing a coin. In the game of cricket, there are three major disciplines: batting, bowling, and the fielding. When selecting 11 players for a team, it is necessary to balancing the team by selecting players to represent each of the above three departments. A player who excels in bowling the cricket ball is considered as a bowler, while a player with higher potential of hitting the cricket ball is considered as a batsman. An all-rounder is a regular performer with bat and the ball. According to Bailey (1989), an all-rounder is a player who is able to grasp a position in his team for either his batting or his bowling ability. Though fielding is an integral part in the game, batting and bowling skills are given higher priorities than fielding. A genuine all-rounder is a special all-rounder who is equally capable of batting and bowling, most importantly this player can bat as a quality batsman and bowl as s quality bowler. Majority of all-rounders in the game of cricket dominate either batting or bowling skills, therefore they are named as batting all-rounders or as bowling all-rounders. Identification of all-rounders is very vital for the success of a team. Classifying an all-rounder as genuine, batting, or a bowling is even beneficial for cricket selection panels, coaches, and players. A review at the literature provides evidences of such studies. Using Indian Premier League (IPL) data, Saikia and Bhattacharjee (2011) classified all-rounders into four groups, namely performer, batting allrounder, bowling all-rounder, and under-performer. According to their results, the Naïve Bayes algorithm has given a classification accuracy of 66.7%. In an attempt to rank all-rounders in test cricket, Tan and Ramachandran (2010) utilized both batting and bowling statistics to devise a mathematical formula. In another study, Stevenson and Brewer (2019) derived a Bayesian parametric model to predict how international cricketers' abilities change between innings in a game. Furthermore, Christie (2012) researched physical requirements of fast bowlers and stated the necessity of physiological demands to evaluate bowlers' performances. Saikia et al. (2016) developed a performance measurement using a combination of batting and bowling statistics to quantify all-rounder's performance. Wickramasinghe (2014) introduced an algorithm to predict batsman's performance using a hierarchical linear model. This multi-level model used player-level and team-level performance indicators to predict the player's performance. Selecting a team against a given opposition team is not an easy task, as various aspects including the strengths and the weaknesses of both teams are required to consider. Bandulasiri et al. (2016) identified a typical ODI game as a mixture of batting, bowling, and decision-making. Presence of a quality allrounder in a team is an asset to a team, as it brings huge flexibility in the composition of the team. A good all-rounder makes the captain's job easy as the player can play a dual role, whenever the captain requires (Van Staden 2008). Though the impact of all-rounders towards the success of a team is enormous, there are no underline criteria to identify them. The existence of prior research work in identifying all-rounders in the game of cricket is handful. According to the knowledge of the author, there is no existing study regarding classification of all-rounders in ODI format. Our objective of this study is to device a method to categorize all-rounders in the ODI format of cricket. We use several machine learning techniques to classify an allrounder as a genuine all-rounder, batting all-rounder, bowling all-rounder, and as an average all-rounder. This study brings novelty for the cricket literature in many ways. According to the author's point of view, this is one of the first studies conducted to classify all-rounders in ODI version of the game using machine learning techniques. Furthermore, the selected player-related performance indicators and the used machine learning techniques are unique for this study. Findings of this study can benefit the entire cricket community and cricket industry as always prediction in sports brings an economical value to the industry (Gakis et al. 2016). The rest of the manuscript is organized as follows. Next section will discuss about the data selection procedure and descriptive statistics about the collected data. In the methodology section, three machine learning techniques are discussed. Then, in the following section findings of this study are illustrated. Finally, the discussion and conclusion section will discuss further about the conducted study and concludes the manuscript. Data Collection and Player -Selection Criteria Data for this study was collected using a publically available website, under the following criteria. Players, who have played more than 50 ODI games with an aggregate score of over 500 runs, were selected. Furthermore, it was essential for each player to have at least a half-century under their name, and collected more than 25 ODI wickets. Under the above criteria, a total of 177 players were selected and ten player related performance indicators (features) were recorded. Table 1 summarises these ten features and their descriptive statistics. Saikia and Bhattacharjee (2011) classified all-rounders based on median value of both batting average and bowling averages. In this collected data, the distributions of both batting and bowling follow Gaussian distributions. Therefore, in this study we use the mean values of both batting and bowling averages to classify all-rounders according to the scheme summarised in Table 2. Figure 1 illustrates the joint distribution of batting and bowling averages, and the four categories of players. Based on the Table 2 and Figure 1, we classify each all-rounder into one of the four categories: genuine all-rounder (G), batting all-rounder (B), bowling allrounder (Bw), and average all-rounder (A). The class variable of the data set is named as Type, which represents each of the four classifications. Methodology In this study, we use three machine learning techniques, NB, kNN, and RF to classify all-rounders into one of the four groups. Regression analysis is one of the alternative conventional statistical procedures for an analysis like this. The number of data appoints used in regression analysis is higher, proportional to the number of involved features (Allision 1999, Bai andPan 2009). Furthermore, some of the machine learning algorithms such as NB is considered as a better performer with smaller datasets (Hand 1992, Kuncheva 2006. Under the previously stated constraints, we opt to use these three machine learning approaches to analyse these data. Naïve Bayes (NB) The NB classifier is considered as one of the simplest and accurate data classifying algorithms. The base of this classifier is the well-known Bayes theorem, used in probability theory. The simplicity, the accuracy, and the robustness of NB have made NB a popular classifying technique with various applications (Arar and Ayan 2017). As the literature indicates, NB is one of the top performing classifiers used in data mining (Wu et al. 2008). Let   12 , ,..., n X x x x  be a n-dimensional random vector (features) from domain According to the Bayes theorem, we have   Then we can write   \| P y X as follows. Therefore, our aim is to find y, that maximize the above expression. In another words, we need to find y, which is k-Nearest Neighbour's Algorithm (kNN) The kNN can be considered as one of the simplest machine learning classifiers, which is based on distance matric (Figure 2). Applications of kNN can be found in text categorization (Elnahrawy 2002), ranking models (Xiubo et al. 2008), and object recognition (Bajramovic et al. 2006). If a novel data point is given, kNN attempts to identify the correct category of the novel point, using the characteristics of the neighbouring data points. The main trait of the data points is going to be the distance from novel data point to each of the other data points. When considering the distance metric, Euclidian is the most commonly used one though other metrics such as Manhattan Distance, Mahalanobis Distance and Chebychev Distances are also used in practice. Table 3 shows some other popular distance matrices used in data classification. represents the M categories (class value). Furthermore, let X be a novel data point. The kNN algorithm can be summarised as follows.  Calculating the distance from this novel point X to all other points in the dataset.  Sort the distances from each point to the novel point and select the k (usually an odd number to prevent tie situations) smallest distances, i.e., nearest k neighbours 12 , ,..., i i ik y y y .  Then for each of the above k nearest neighbours, it records the corresponding class (labels) ; 1, 2,.., j c j M  and calculate the following conditional probability. The class j c that has the highest probability is assigned to the novel data point, as the category of the data point. Random Forest (RF) RF algorithm extends the idea of decision trees by aggregating higher number of decision trees to reduce the variance of the novel decision tree (Couronné 2018). Each tree is built upon a collection of random variables (features) and a collection of such random trees is called a Random Forest. Dues to the higher classification accuracy, RF is considered as one of the most successful classification algorithms in modern-times (Breiman 2001, Biau andScornet 2016) (Figure 3). Furthermore, the performance of this classification algorithm is significant for unbalanced and missing data (Shah et al. 2014), compared to its counterparts. RF has been studied by many researchers both in theoretically and experimentally since its introduction in 2001 (Bernard et al. 2007, Breiman 2001, Geurts 2006, Rodriguez 2006. Further studies have been conducted to improve the classify-cation accuracy of RF by clever selection of the associated parameters of RF (Bernard et al. 2007). A handful of applications of machine learning algorithms in the context of cricket can be seen in the literature. Using kNN and NB classifiers, Kumar and Roy (2018) forecasted final score of an ODI score after the completion of the fifth over of the game. NB and RF were two of the machine learning techniques Passi and Pandey (2018) used in their study to predict the individual player's performance in the game of cricket. Using English T20 county cricket data from 2009 to 2014, Kampakis and Thomas (2015) developed a machine learning model to predict the outcome of the T20 cricket game. Findings All the experimental outcomes were tested under the k-fold cross-validation, which is used to generalize the findings of the study to any given independent sample as discussed in the literature (Burman 1989, Kohavi 1995. We executed all of the three machine learning classifiers with the collected data and according to the experimental outcomes, NB classifier reached a maximum of 60.7% prediction accuracy. Furthermore, the maximum prediction accuracy using Knn was 55.08%. In order to see how the prediction accuracy changes with the selection of distance matric with kNN algorithm, we measured the prediction of accuracies with respect to the various distance matrices. Table 4 summarises the percentage of prediction accuracy for each of the selected distance metric and the value k used in kNN. With RF, an initial accuracy rate of 93.34% was recorded, which is the highest among the three classifiers we used. Further investigation was conducted to optimize the prediction accuracy, by varying the associated parameters of RF. Dataset subset subset subset Tree Tree Tree Important Parameters used in RF Among the various parameters used with RF, the following important parameters were changed to see a better prediction rate. n_estimators: This represents the number of trees in the RF. max_features: This represents the maximum number of features when the RF selects the split point. min_samples_leaf: This represents how many minimum number of data points in the end node. After searching for better parameterization, we investigated the associated errors with the RF Regression model. Both Mean Absolute Error (MAE) and Mean Squared Error (MSE) were recorded for the model with previously identified parameters. Figures 6 and 7 Discussion and Conclusion In this manuscript, we discussed how to categorize all-rounders in the game of ODI cricket. Using a collection of 177 players from all the ODI playing countries, ten player-related predictors, together with three machine learning techniques, we investigated how to categorize all-rounders into one of the four categories. In this study, we utilized three machine learning techniques, namely Random Forest (RF), k-nearest neighbours (kNN), and Naïve Bayes (NB) to predict the appropriate category each of the all-rounder should belong. After initial execution of the above three algorithms, the prediction accuracies for kNN, NB, and RF were 50.08%, 59.00%, and 93.34% respectively. Further improvement of the prediction accuracy was able to achieve with the proper selection of the parameters. By changing the distance metric with kNN and the k value, we were able to improve the prediction accuracy up to 55.08%. Similarly, NB was improved up to 60.7%. According to Figures 4-7, it is clear that RF has improved to the highest prediction accuracy of 99.4%, with the selection of appropriate values for the parameters. This can be reached with two different parameter settings. i.e., when n_estimators is 10, max_features is 0.50, min_ samples_leaf is 2, and n_estimators is 10, max_features is 0.75, min_samples_leaf is 1. In addition to the prediction accuracy, an investigation was conducted to find out the relative errors involved with these processes. According to the findings, these errors became minimum when n_estimators is 10, max_features is 0.75 and min_samples_leaf is 1 respectively and the values were 0.68 and 0.86 respectively. In summary, our experimental results indicated that RF algorithm outperformed both kNN and NB by huge margins. The findings of this study benefit the officials of the game of cricket and the players in many ways. Player selection committees, coaches of teams, and even the players can utilize these outcomes to identify appropriate all-rounders. It would be important to include additional performance indicators, including statistics about the opposition teams that the players play against for future studies.	2020-02-06T09:11:37.668Z	2020-01-31T00:00:00.000	{ "year": 2020, "sha1": "d8545bcbddf183ff410c956a49c0fa707ffb4840", "oa_license": null, "oa_url": "https://doi.org/10.30958/ajspo.7-1-2", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "70cddfb060105e08e7cc793fe65982b671ab2c6e", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Computer Science" ] }
248821368	pes2o/s2orc	v3-fos-license	A Morphological Study of Solvothermally Grown SnO2 Nanostructures for Application in Perovskite Solar Cells Tin(IV) oxide (SnO2) nanostructures, which possess larger surface areas for transporting electron carriers, have been used as an electron transport layer (ETL) in perovskite solar cells (PSCs). However, the reported power conversion efficiencies (PCEs) of this type of PSCs show a large variation. One of the possible reasons for this phenomenon is the low reproducibility of SnO2 nanostructures if they are prepared by different research groups using various growth methods. This work focuses on the morphological study of SnO2 nanostructures grown by a solvothermal method. The growth parameters including growth pressure, substrate orientation, DI water-to-ethanol ratios, types of seed layer, amount of acetic acid, and growth time have been systematically varied. The SnO2 nanomorphology exhibits a different degree of sensitivity and trends towards each growth factor. A surface treatment is also required for solvothermally grown SnO2 nanomaterials for improving photovoltaic performance of PSCs. The obtained results in this work provide the research community with an insight into the general trend of morphological changes in SnO2 nanostructures influenced by different solvothermal growth parameters. This information can guide the researchers to prepare more reproducible solvothermally grown SnO2 nanomaterials for future application in devices. Introduction Today we are facing a high demand for energy sources due to the continuous increase of the human population and ever-advancing technologies. The huge energy consumption cannot rely only on the non-sustainable supplies from conventional energy. The development of renewable energy sources is of significant importance to relieve the energy burden and minimize the release of pollutants during energy generation. Recently, perovskite solar cells (PSCs) have made a breakthrough in the field of photovoltaics (PVs). In the last decade, the power conversion efficiency (PCE) of PSCs has increased from the first reported value of 3.8% to the latest record of 25.7% [1]. Such rapid growth of the PCE of PSCs is due to the impressive intrinsic properties of halide perovskite absorber materials with a high absorption coefficient, high charge carrier mobility, tunable bandgap, long charge carrier diffusion lengths, etc. [2]. Nevertheless, a number of challenges, such as material stability, due to the large degree of variation in nanomorphology of ETLs controlled by multiple material growth parameters, causing the complexity in finding the optimal processing conditions for the fabrication of PSCs. The SnO 2 nanostructures can be synthesized in an autoclave reactor via the solvothermal growth method [64]. This technique has gained great success in preparing nanomaterials due to its advantages, such as low processing temperature and energy consumption, low-cost manufacturing (i.e., equipment, raw materials etc.), and the environmental benignity of the process. In this work, the solvothermal technique for preparing SnO 2 nanorod arrays is systematically investigated. The impact of different growth parameters including the (i) growth pressure, (ii) substrate orientation, (iii) deionized (DI) water-to-ethanol ratios, (iv) types of seed layer, (v) amount of acetic acid, and (vi) growth time on the morphology of SnO 2 nanostructures are studied. The SnO 2 nanorod arrays prepared under different conditions are characterized. The obtained results provide thorough information to the community for synthesizing SnO 2 nanostructures with improved reproducibility for the application in PSCs. Synthesis of SnO 2 Nanorod Arrays Fluorine-doped tin oxide (FTO)-coated glasses were used to grow SnO 2 nanorod arrays. The substrates were cleaned by a thorough cleaning procedure using acetone, isopropanol (IPA), and DI water in an ultrasonic bath. The cleaned substrates were airdried by nitrogen and then put in an ultraviolet ozone cleaning system (UVOCS) for 30 min to eliminate organic contaminants. Unless otherwise specified, the precursor solution of SnO 2 was prepared by dissolving 0.15 mmol sodium bromide (NaBr) in 0.75 mL of DI water. An amount of 0.05 mmol tin(IV) chloride pentahydrate (SnCl 4 · 5H 2 O) was dissolved in 6 mL glacial acetic acid under stirring for 10 min. After that, the two prepared solutions were mixed well, and 0.75 mL of ethanol was added into the mixture. Then, the cleaned substrates were loaded into the Teflon-lined autoclave reactor and heated in a muffle furnace at 200 • C for 12 h. The samples with SnO 2 nanorod arrays were finally cleaned by DI water and ethanol in an ultrasonic bath. Growth Pressure The growth pressure was varied by adding the same amount of SnO 2 precursor solution (7.5 mL) into different volume sizes (25 mL, 50 mL, and 100 mL) of Teflon-lined autoclaves. The filling ratios of the reactors were 30%, 15%, and 7.5%, respectively. The built-up pressure in the reaction system increased with the filling ratios. On the other hand, in order to maintain a constant pressure among different volume sizes of Teflon-lined autoclaves, the same filling ratio of 7.5% was obtained by adjusting the total amount of the precursor solution in each Teflon liner. The compositions of the precursor solution are described in Section 2.1 above. Substrate Orientation Three different geometric alignments of the FTO-coated glass substrates were investigated. The substrates with the side of FTO facing down were placed at an angle of 0 • (horizontally facing down), 45 • , or 90 • (vertically orientated) with respect to the bottom surface of the Teflon liner. The polytetrafluoroethylene (PTFE) sample holders were used for holding FTO-coated substrates in place. The filling ratio of the reactor for each condition was maintained at 7.5% using the 100 mL the Teflon-lined autoclaves. DI Water-to-Ethanol Ratios The DI water-to-ethanol ratio was adjusted to 1:9, 3:7, 1:1, 7:3, and 9:1, where the 1:1 ratio was referred to a mixture of 0.75 mL of DI water and 0.75 mL of ethanol (1.5 mL in total). The other ratios of DI water to ethanol were varied according to obtain a 1.5 mL Nanomaterials 2022, 12, 1686 4 of 17 mixture. The same amount of glacial acetic acid (6 mL), SnCl 4 · 5H 2 O (0.05 mmol), and NaBr (0.15 mmol) were used for all conditions of different DI water-to-ethanol ratios. Seed Layers Three different types of SnO 2 seed layers composed of thin films, nanoparticles, or quantum dots were prepared on FTO-coated glasses for subsequent solvothermal growth of SnO 2 nanorod arrays. The SnO 2 thin films were prepared by the magnetron sputtering. The SnO 2 ceramic disc (99.99% purity, Kurt J. Lesker Company, Pittsburgh, PA, USA) with a diameter of 50.8 mm was used as a target. A working pressure of 5 × 10 −3 Torr in a pure argon atmosphere was maintained during the radio frequency sputtering to achieve a SnO 2 layer in 10 nm. For SnO 2 quantum dots, the precursor solution was prepared by dissolving 3.99 mol SnCl 2 ·2H 2 O in DI water. The solution was stirred for 3-4 h under oxygen flow (1L/min), followed by filtering with a syringe filter with a pore size of 0.45 µm. The prepared solution was spin-coated on top of the FTO-coated glass substrates at 3000 rpm for 30 s. After deposition, the substrates were placed onto a hotplate and annealed at 200 • C for 1 h. For SnO 2 nanoparticles, the solution was prepared using the SnO 2 colloidal dispersion solution (Tin(IV) oxide, 15% in H 2 O colloidal dispersion, Alfa Aesar, Heysham, United Kingdom) diluted with DI water in a 1:5 volume ratio. The dispersion solution was spin-coated on top of the FTO-coated glass substrate at 3000 rpm for 30 s, followed by thermal annealing on a hotplate at 150 • C for 30 min. The SnO 2 precursor solution used for solvothermal growth is described in Section 2.1, except that 6.75 mL instead of 6 mL glacial acetic acid was added into the mixture. The filling ratio of the reactor was 8.3%. Glacial Acetic Acid The precursor solution of SnO 2 was prepared by the procedure as described in Section 2.1 (1:1 DI water-to-ethanol ratio). The glacial acetic acid was used in an amount of 6 mL, 6.5 mL, and 9.75 mL. The filling ratio of the reactor was 7.5%, 8%, and 11.3%, respectively. Growth Time The substrates with a 15 nm SnO 2 thin film as a seed layer were placed in the precursor solution as described in Section 2.1, except that 6.75 mL instead of 6 mL glacial acetic acid was added into the mixture for solvothermal growth of SnO 2 nanorod arrays. The growth temperature was set as 200 • C for a duration of 6, 12, and 24 h. Meanwhile, another growth condition was designed as a duration of 12 h followed by a replacement of the original precursor with the freshly prepared solution for an additional 12 h growth. The Perovskite Solution The perovskite precursor solution was prepared by dissolving 1. Louis, MO, USA) (4:1 by volume ratio). The solution was stirred on a vortex mixer for 5 h and filtered using the syringe filters with a pore size of 0.45 µm. Then, 28 µL of 1.5 M CsI solution in DMSO and 28 µL of 1.5 M RbI solution in a mixture of DMF:DMSO (4:1 volume ratio) was added to 940 µL of the filtered precursor solution. The perovskite precursor solution was spin-coated on the top of the optimized SnO 2 nanorod arrays, which were treated by oxygen plasma for 15 s at a power of 70 W using Plasma Prep III Solid State system from SPI Supplies, using a two-step spinning recipe consisting of 1000 rpm for 10 s and 5000 rpm for 30 s. During the last 10 s of the second spinning step, the antisolvent, chlorobenzene (CB, 99.8%, anhydrous, Sigma Aldrich, St. Louis, MO, USA), was dropped onto the sample. Further, the films were placed onto a hotplate and annealed at 105 • C for 75 min. The hole-transport layer (HTL) was prepared by spin coating (3000 rpm for Results and Discussions It is important to understand the effect of each solvothermal growth parameter on the morphology of SnO 2 nanostructures. It is noteworthy that some growth parameters have a correlation with each other. For example, varying the volume amount of acetic acid leads to a change in the total volume of precursor solution, resulting in a change of the solvothermal growth pressure. Despite keeping the same molar ratio among the chemicals, a change of the total volume of precursor solution in Teflon-lined autoclaves will affect the growth pressure as well as the absolute amount of reactants available for growing SnO 2 nanostructures. A systematic study of different growth conditions should be performed experimentally so as to pinpoint their impact on the morphology of the obtained SnO 2 nanostructures, and thus the reproducibility of desired SnO 2 nanostructures can be ensured for certain applications. The pressure effect was firstly investigated by using different sizes of Teflon-lined autoclave reactors (25 mL, 50 mL, and 100 mL), filling them with an identical volume and composition of precursor solutions for solvothermal growth at 200 • C for 12 h. The built pressure inside the reactor during solvothermal growth is inversely proportional to the volume size of the Teflon liner. By using a pressure gauge connected to the autoclave reactor during the solvothermal growth process, the pressure was stabilized at 130.0 psi, 87.0 psi, and 72.5 psi for a filling ratio of 30%, 15%, and 7.5%, respectively. Figure 1 shows the topview and cross-sectional images obtained from the scanning electron microscope (SEM). It is clearly observed that the change of pressure significantly affects the morphology of SnO 2 . The nanorod arrays, in which some of the nanorods are in a small bundle, the structure can be distinguished from the sample prepared in the largest size of the reactor (100 mL). When the pressure is increased by using the smaller reactors, the size of the nanorod bundle increases substantially. Furthermore, the thickness of the nanostructures grown on the FTOcoated glass increases with the growth pressure. Based on the cross-sectional SEM images (Figure 1d-f), an average of thickness was determined as~425 ± 21 nm,~319 ± 19 nm, and 131 ± 12 nm for the samples prepared in 25 mL, 50 mL, and 100 mL reactors, respectively. The difference in nanostructures among the samples prepared in different pressure is due to the changes in Gibbs free energy [65]. A rise in pressure during solvothermal growth leads to an increase of crystallite sizes [65] and induces the coalescence of nanorods to form a bigger bundle [66]. The statistics of the bundle diameters and the length of the SnO 2 nanorod arrays prepared in different sizes of autoclave reactors are summarized in Table S1 (Supplementary Materials). In order to obtain the desired morphology of SnO 2 nanostructures for device fabrication, the volume of precursor for solvothermal growth should be adjusted accordingly with the size of the autoclave reactors so that an optimized growth pressure can be achieved. However, as mentioned earlier, the precursor solutions in different volumes contain a different total amount of solutes available for reaction, which may probably vary the morphology of the solvothermally grown samples. In order to verify this effect, the volume of the precursor solutions was adjusted accordingly with three different sizes of autoclave reactors to achieve the identical filling ratio (i.e., 7.5%), and hence the same pressure, during solvothermal growth for 12 h. The filling ratio of 7.5% was selected to be investigated further as the SnO 2 nanorod arrays grown under this condition are less compact, which allows for the penetration of the subsequent deposited perovskite material into the gaps among SnO 2 nanorods, and thus this structure can facilitate the carrier transport between the SnO 2 ETL and perovskite absorber. The SEM images of the samples grown in 25 mL, 50 mL, and 100 mL autoclave reactors under the same pressure via adjusting the volume of precursor solution are shown in Figure 2. For a duration of the 12-hour reaction, no significant difference was observed in terms of the surface morphology and lengths of SnO 2 nanorods, indicating that the same SnO 2 nanostructure can be reproduced under the same growth pressure regardless of different volumes of precursor solutions (i.e., different total amount of solutes) in this case. In order to obtain the desired morphology of SnO2 nanostructures for device fabrication, the volume of precursor for solvothermal growth should be adjusted accordingly with the size of the autoclave reactors so that an optimized growth pressure can be achieved. However, as mentioned earlier, the precursor solutions in different volumes contain a different total amount of solutes available for reaction, which may probably vary the morphology of the solvothermally grown samples. In order to verify this effect, the volume of the precursor solutions was adjusted accordingly with three different sizes of autoclave reactors to achieve the identical filling ratio (i.e., 7.5%), and hence the same pressure, during solvothermal growth for 12 hours. The filling ratio of 7.5% was selected to be investigated further as the SnO2 nanorod arrays grown under this condition are less compact, which allows for the penetration of the subsequent deposited perovskite material into the gaps among SnO2 nanorods, and thus this structure can facilitate the carrier transport between the SnO2 ETL and perovskite absorber. The SEM images of the samples grown in 25 mL, 50 mL, and 100 mL autoclave reactors under the same pressure via adjusting the volume of precursor solution are shown in Figure 2. For a duration of the 12hour reaction, no significant difference was observed in terms of the surface morphology and lengths of SnO2 nanorods, indicating that the same SnO2 nanostructure can be reproduced under the same growth pressure regardless of different volumes of precursor solutions (i.e., different total amount of solutes) in this case. In order to obtain the desired morphology of SnO2 nanostructures for device fabrication, the volume of precursor for solvothermal growth should be adjusted accordingly with the size of the autoclave reactors so that an optimized growth pressure can be achieved. However, as mentioned earlier, the precursor solutions in different volumes contain a different total amount of solutes available for reaction, which may probably vary the morphology of the solvothermally grown samples. In order to verify this effect, the volume of the precursor solutions was adjusted accordingly with three different sizes of autoclave reactors to achieve the identical filling ratio (i.e., 7.5%), and hence the same pressure, during solvothermal growth for 12 hours. The filling ratio of 7.5% was selected to be investigated further as the SnO2 nanorod arrays grown under this condition are less compact, which allows for the penetration of the subsequent deposited perovskite material into the gaps among SnO2 nanorods, and thus this structure can facilitate the carrier transport between the SnO2 ETL and perovskite absorber. The SEM images of the samples grown in 25 mL, 50 mL, and 100 mL autoclave reactors under the same pressure via adjusting the volume of precursor solution are shown in Figure 2. For a duration of the 12hour reaction, no significant difference was observed in terms of the surface morphology and lengths of SnO2 nanorods, indicating that the same SnO2 nanostructure can be reproduced under the same growth pressure regardless of different volumes of precursor solutions (i.e., different total amount of solutes) in this case. The impact of substrate orientation in the autoclave reactor during solvothermal growth on the morphology of the SnO 2 nanostructures is another concern. The experiment was designed for three different substrate alignments in 100 mL autoclave reactors. (Figure 3c), relatively more tips of the nanorods point in a perpendicular direction with respect to the surface of the substrate. The density of perpendicularly grown nanorods is reduced, as observed from Figure 3a,b, when the samples were aligned at an angle of 45 • or 90 • . Instead, the nanorods tend to grow in different directions and form relatively larger bundles compared to the case as shown in Figure 3c (mounted at 0 • ). The length of SnO 2 nanorod arrays is shortest for the sample mounted at an angle of 90 • (116 nm ± 21 nm) compared to the samples mounted at an angle of 0 • (132 nm ± 13 nm) or 45 • (135 nm ± 30 nm). Table S2 summarizes the dimensions of SnO 2 nanostructures grown on the substrates mounted at different orientations. Considering for the slight difference in morphology, the atomic force microscopy (AFM) was performed on the samples mounted at 45 • , 90 • , and 0 • in order to further distinguish their surface characteristics. The obtained results are shown in Figure S1 in the Supplementary Materials. The results show that the root mean square (RMS) roughness is 0.09 µm, 0.04 µm, and 0.07 µm for the samples mounted at 45 • , 90 • , and 0 • , respectively. In this work, placing the substrate horizontally facing down is more practical than other orientations (45 • or 90 • ) as the latter orientations have more restrictions in the dimensions of substrates based on the design of the autoclave reactors. The substrates that are larger in size cannot be fully immersed into the precursor solution, resulting in non-uniform growth of the SnO 2 nanostructures. Therefore, for further studies of other growth parameters, the substrates were mounted horizontally (0 • ) for solvothermal growth. was designed for three different substrate alignments in 100 mL autoclave reactors. The substrates with the side of FTO facing down were placed at an angle of 0°, 45°, and 90° with respect to the bottom surface of the Teflon liner. The SEM images of the samples aligned in different orientations for solvothermal growth are shown in Figure 3. A subtle difference in the morphology of SnO2 nanostructures can be observed from the top-view SEM images. For the sample mounted horizontally with the FTO side facing down towards the bottom surface of the Teflon liner (0°), as observed from the top-view SEM image (Figure 3c), relatively more tips of the nanorods point in a perpendicular direction with respect to the surface of the substrate. The density of perpendicularly grown nanorods is reduced, as observed from Figure 3a,b, when the samples were aligned at an angle of 45° or 90°. Instead, the nanorods tend to grow in different directions and form relatively larger bundles compared to the case as shown in Figure 3c (mounted at 0°). The length of SnO2 nanorod arrays is shortest for the sample mounted at an angle of 90° (116 nm ± 21 nm) compared to the samples mounted at an angle of 0° (132 nm ± 13 nm) or 45° (135 nm ± 30 nm). Table S2 summarizes the dimensions of SnO2 nanostructures grown on the substrates mounted at different orientations. Considering for the slight difference in morphology, the atomic force microscopy (AFM) was performed on the samples mounted at 45°, 90°, and 0° in order to further distinguish their surface characteristics. The obtained results are shown in Figure S1 in the Supplementary Materials. The results show that the root mean square (RMS) roughness is 0.09 μm, 0.04 μm, and 0.07 μm for the samples mounted at 45°, 90°, and 0°, respectively. In this work, placing the substrate horizontally facing down is more practical than other orientations (45° or 90°) as the latter orientations have more restrictions in the dimensions of substrates based on the design of the autoclave reactors. The substrates that are larger in size cannot be fully immersed into the precursor solution, resulting in non-uniform growth of the SnO2 nanostructures. Therefore, for further studies of other growth parameters, the substrates were mounted horizontally (0°) for solvothermal growth. The acetic acid, ethanol, and DI water form a ternary solvent system for the solvothermal growth. The work of Chen et al. [64] demonstrated that the appropriate mixing ratios of these three components are necessary to obtain SnO2 nanorod arrays on Ti foil. In this The acetic acid, ethanol, and DI water form a ternary solvent system for the solvothermal growth. The work of Chen et al. [64] demonstrated that the appropriate mixing ratios of these three components are necessary to obtain SnO 2 nanorod arrays on Ti foil. In this work, the recipe of solvothermal growth of SnO 2 nanorod arrays reported by Chen et al. [64] was modified accordingly with the size of autoclave reactors and designed experimental conditions. It is noteworthy that the optimization of a ternary solvent system can be complicated as the observed effect from solvothermally grown samples is a result of the interplay of three solvents. Therefore, in this study, the impacts of DI water, ethanol, and the acetic acid on the morphologies of SnO 2 nanostructures were investigated separately via systematically varying the volume amount of each solvent. The demonstration of the solvent effects on SnO 2 nanostructures was based on FTO-coated glass substrates, which are the common substrates used for PSCs. Figure 4 shows the top-view and cross-sectional images of the samples prepared by using different DI water-to-ethanol ratios for solvothermal growth. A 1.5 mL solvent mixture prepared by mixing the DI water and ethanol in different volume ratios (1:9, 3:7, 1:1, 7:3, or 9:1) was used for solvothermal growth, while other constituents of the precursor solution were kept constant in a 100 mL Teflon liner. The DI water in-volves the hydrolysis process of SnCl 4 · 5H 2 O while ethanol facilitates the formation of a well-defined SnO 2 nanorod array [64]. It is known that increasing water content can accelerate the hydrolysis process of SnCl 4 · 5H 2 O. It is consistent to the observation that, when DI water-to-ethanol ratio is 1:9, the SnO 2 in form of nanorods can be distinguished. Chen et al. [64] reported that low water content causes slow hydrolysis, which leads to formation of small cube-like nanoparticles on the substrate after 24 h solvothermal growth. Compared to our results, it is believed that the amount of the smallest portion of water used in this work is still within the tolerance for growing SnO 2 nanorods. On the other hand, the varying amount of ethanol demonstrates a significant impact on the morphology of SnO 2 nanostructures. For the condition of using a very low portion of ethanol (DI water-to-ethanol ratio: 9:1), a thin layer composed of nanoparticles grown on the FTO can be observed (Figure 4i,j). This is consistent with the results reported by Chen et al. [64], for which a lower ethanol content leads to the growth of SnO 2 nanoparticles instead of nanorods. Furthermore, a trend of transformation from nanoparticles to distinguishable nanorods can be observed from our results when comparing the SEM images of the samples prepared in the solvent with an increasing proportion of ethanol. It is interesting to observe that the nanorods tend to form larger bundles when the samples are prepared at a condition of the highest ethanol proportion (Figure 4a), compared to other samples grown at the conditions with a lower proportion of ethanol ( Figure 4c,e). For the condition of DI water-to-ethanol ratio at 7:3, the nanorod structures start to be barely distinguishable from the top-view SEM image (Figure 4g), while the dense It is interesting to observe that the nanorods tend to form larger bundles when the samples are prepared at a condition of the highest ethanol proportion (Figure 4a), compared to other samples grown at the conditions with a lower proportion of ethanol (Figure 4c,e). For the condition of DI water-to-ethanol ratio at 7:3, the nanorod structures start to be barely distinguishable from the top-view SEM image (Figure 4g), while the dense nanostructures composed of clusters and nanorods can be observed from the corresponding cross-sectional SEM image (Figure 4h), which is the result of the interplay between the relatively high portion of DI water and moderately low portion of ethanol. It is noteworthy that the differences in the molecular weight and boiling point of the DI water and ethanol can also affect the vapor pressure. Therefore, the pressure built in the autoclave reactor for the conditions of using 1:9, 1:1, and 9:1 DI water: ethanol ratio was measured by a pressure gauge. It is found that a similar pressure of 72.5 psi was detected for all testing conditions. This result indicates that the pressure effect associated with variations of the DI water-toethanol ratio on the change of SnO 2 nanostructures, as shown in Figure 4, is negligible under current experimental conditions. The dimensions of SnO 2 nanostructures grown in a ternary solvent system with different DI water-to-ethanol ratios are summarized in Table S3 in Supplementary Materials. The ratio of DI water to ethanol was selected as 1:1 for further study of other solvothermal growth parameters. Usually, a compact ETL is used underneath the metal oxide nanostructures. The usage of the compact ETL can avoid forming the shunt paths due to the direct contact of the perovskite layer with the conductive FTO through the gaps of the nanostructures. In this work, three different types of compact layers, which can also act as a seed layer for subsequent solvothermal growth of SnO 2 nanostructures, were prepared by magnetron sputtering, spin coating of the SnO 2 nanoparticle (10-15 nm) dispersion solution, and deposition of SnO 2 quantum dots (~5 nm) via a sol-gel process [12]. The top-view and the cross-sectional images of SnO 2 nanostructures grown on different types of seed layers are shown in Figure 5. Comparing to the samples of SnO 2 nanostructures grown on FTO without a seed layer, it can be noticed that the nanorods grown on the seed layer tend to orientate in a single direction perpendicular to the substrates. Furthermore, the density of the nanorod arrays is obviously increased when they were grown on the seed layer regardless of their types. Ideally, SnO 2 nanorod arrays used as ETL for perovskite solar cells should contain reasonable free space among the nanostructures to accommodate the perovskite grains. The structure of perovskite grains embedded in SnO 2 nanorod arrays increases the interfacial areas between the two materials so that the photogenerated electrons can be transported efficiently between the perovskite and ETL before carrier recombination. In this work, only SnO 2 nanorod arrays grown on the SnO 2 thin film prepared by the magnetron sputtering exhibit more free spaces, as observed from Figure 5a,b. Besides, compared to the samples of using the other two types of seed layers, the diameters of the nanorods grown on the SnO 2 thin film deposited by the magnetron sputtering seem to be the largest, as more SnO 2 nanorods join to form a larger bundle. Based on the SEM images of Figure 5e,f, SnO 2 quantum dots induce the growth of SnO 2 nanorods in the highest density and finest diameter, resulting in the highest compacity of SnO 2 nanostructures. The SnO 2 nanorods grown on SnO 2 nanoparticles are relatively larger in diameter and form relatively less compact nanostructures (Figure 5c,d) compared to the nanorods grown on SnO 2 quantum dots. It is believed that the size of the SnO 2 particles will affect the density of the nucleation sites, leading to the difference in morphology of SnO 2 nanorod arrays grown subsequently. Nevertheless, varying SnO 2 quantum dots and nanoparticle concentrations in the seed layer should be performed further to investigate their impact on the morphology of subsequently grown nanomaterials. The dimensions of SnO 2 nanostructures grown on different types of seed layers are summarized in Table S4 (Supplementary Materials). Considering the criteria of ETL used in PSCs, it is more suitable to prepare the compact layer by the magnetron sputtering, as the morphology of SnO 2 nanorod arrays grown on the top contains more free spaces for subsequent growth of perovskite grains. thy that the differences in the molecular weight and boiling point of the DI water and ethanol can also affect the vapor pressure. Therefore, the pressure built in the autoclave reactor for the conditions of using 1:9, 1:1, and 9:1 DI water: ethanol ratio was measured by a pressure gauge. It is found that a similar pressure of 72.5 psi was detected for all testing conditions. This result indicates that the pressure effect associated with variations of the DI water-to-ethanol ratio on the change of SnO2 nanostructures, as shown in Figure 4, is negligible under current experimental conditions. The dimensions of SnO2 nanostructures grown in a ternary solvent system with different DI water-to-ethanol ratios are summarized in Table S3 in Supplementary Materials. The ratio of DI water to ethanol was selected as 1:1 for further study of other solvothermal growth parameters. Usually, a compact ETL is used underneath the metal oxide nanostructures. The usage of the compact ETL can avoid forming the shunt paths due to the direct contact of the perovskite layer with the conductive FTO through the gaps of the nanostructures. In this work, three different types of compact layers, which can also act as a seed layer for subsequent solvothermal growth of SnO2 nanostructures, were prepared by magnetron sputtering, spin coating of the SnO2 nanoparticle (10-15 nm) dispersion solution, and deposition of SnO2 quantum dots (~5 nm) via a sol-gel process [12]. The top-view and the crosssectional images of SnO2 nanostructures grown on different types of seed layers are shown in Figure 5. The impact of different amount of acetic acid (6 mL, 6.5 mL, and 9.75 mL) on the morphology of SnO 2 nanostructures was investigated based on the samples with a seed layer prepared by the magnetron sputtering. The differences in obtained morphology of SnO 2 nanorod arrays are shown in Figure 6. It is found that when the amount of acetic acid is reduced from 9.75 mL to 6 mL, SnO 2 nanorods tend to form increasing bundle sizes and lengths. Chen et al. [64] suggested that acetic acid can reduce the rate of hydrolysis due to the coordination of CH 3 COO − ions with Sn 4+ ions. Meanwhile, the acetic acid and its side product ethyl acetate, generated during solvothermal growth, are the organic ligands, which can promote the growth of well-defined tetragonal SnO 2 nanorods and prevent the fusion of SnO 2 nanorods into bundles. Our experimental results are consistent with the proposed mechanism of Chen et al. It is observed from the SEM images (Figure 6b,d) that the lengths of SnO 2 nanorods prepared at the condition with 6 or 6.5 mL acetic acid are significantly longer than the samples prepared at the condition using 9.75 mL acetic acid, which can be probably attributed to the higher hydrolysis rate during solvothermal growth because of using a lower amount of acetic acid (6 mL and 6.5 mL). On the other hand, when a lower amount of acetic acid was used (6 mL), SnO 2 nanorods were merged together to form larger bundles, as observed from Figure 6a,b. When the amount of acetic acid was increased from 6 to 6.5 mL, the sizes of SnO 2 nanorod bundles were significantly reduced with more free spaces generated among the nanostructures (Figure 6c,d). When the amount of acetic acid (9.75 mL) was further increased, SnO 2 nanorods became more distinguishable and well separated from each other, while their lengths were the shortest (80 nm ± 17 nm) among all conditions using different amounts of acetic acid. The lengths of SnO 2 nanostructures grown in different acetic acid concentrations are summarized in Table S5 (Supplementary Materials). Our results clearly demonstrate the role of acetic acid in controlling the nanomorphology of solvothermally grown SnO 2 nanostructures. It is noteworthy that using different amounts of acetic acid also varies the volume of precursor solution, which contributes to the change of growth pressure. However, in this case, it is not possible to decouple pressure changes by varying the amount of acetic acid. Nevertheless, the impact of pressure changes attributed to varying the volume of acetic acid from 6 mL to 9.75 mL on the morphology of SnO 2 nanostructures is relatively small compared to the conditions for preparing the samples, as shown in Figure 1. volume of precursor solution, which contributes to the change of growth pressure. However, in this case, it is not possible to decouple pressure changes by varying the amount of acetic acid. Nevertheless, the impact of pressure changes attributed to varying the volume of acetic acid from 6 mL to 9.75 mL on the morphology of SnO2 nanostructures is relatively small compared to the conditions for preparing the samples, as shown in Figure 1. Further investigation of growth time was performed based on the experimental condition for preparing samples, as shown in Figure 6 b,c. The growth temperature was set as 200 °C for different durations (6, 12, and 24 h). Besides, another growth condition was designed as a duration of 12 h followed by a replacement of the original precursor with the freshly prepared solution for an additional 12 h growth. The top-view and cross-sectional SEM images are shown in Figure 7. By comparing the SEM images among Figure 7a-f, different growth durations (6, 12, and 24 h) do not cause significant changes in terms of SnO2 nanostructures as well as the length of the nanorods, as observed from the topview and cross-sectional SEM images of the samples. However, if a two-step process is Further investigation of growth time was performed based on the experimental condition for preparing samples, as shown in Figure 6 b,c. The growth temperature was set as 200 • C for different durations (6, 12, and 24 h). Besides, another growth condition was designed as a duration of 12 h followed by a replacement of the original precursor with the freshly prepared solution for an additional 12 h growth. The top-view and cross-sectional SEM images are shown in Figure 7. By comparing the SEM images among Figure 7a-f, different growth durations (6,12, and 24 h) do not cause significant changes in terms of SnO 2 nanostructures as well as the length of the nanorods, as observed from the top-view and cross-sectional SEM images of the samples. However, if a two-step process is implemented by performing consecutive solvothermal growth two times with the use of freshly prepared precursor solutions, the obtained SnO 2 nanostructures become more compact with larger size of nanorod bundles. Meanwhile, the length of nanorods prepared by the two-step process exhibit an increase of~1.7 times (i.e., 219 nm ± 49 nm) compared to the length of SnO 2 nanorods prepared by the one-step solvothermal growth process. The summary of the dimensions of SnO 2 nanostructures grown with different growth duration is shown in Table S6 (Supplementary Materials). It is believed that the non-sensitivity of SnO 2 nanostructures towards changing of different growth durations in the one-step solvothermal growth process is due to the other limiting factors such as the total amount of different solutes in the precursor solution. Based on our experiments, it is challenging to vary the length of the nanorods as an independent parameter while keeping the desired density and dimension of SnO 2 nanorods/bundles unchanged. Further intensive research efforts should be placed on different combinations of experimental conditions. Nevertheless, our obtained results so far demonstrate the general trend of morphological changes in SnO 2 nanostructures caused by different solvothermal growth parameters. The SnO2 nanorod arrays can be used as the ETL in PSCs. The samples with the similar morphology of SnO2 nanorod arrays, as shown in Figure 7c,d, were used for device fabrication. The purity of SnO2 nanostructures used for device fabrication was confirmed by the X-ray diffractometer. The obtained XRD patterns of the SnO2 seed layer prepared by the magnetron sputtering and SnO2 nanorod arrays grown on the seed layer are shown The SnO 2 nanorod arrays can be used as the ETL in PSCs. The samples with the similar morphology of SnO 2 nanorod arrays, as shown in Figure 7c,d, were used for device fabrication. The purity of SnO 2 nanostructures used for device fabrication was confirmed by the X-ray diffractometer. The obtained XRD patterns of the SnO 2 seed layer prepared by the magnetron sputtering and SnO 2 nanorod arrays grown on the seed layer are shown in Figure 8. The detected XRD peaks were consistent to the reported work [67] showing the tetragonal SnO 2 material. The I-V characteristics of the fabricated PSCs were studied. The obtained results are plotted in Figure 9. It is found that the PSC with SnO 2 ETL treated by oxygen plasma exhibits significantly improved photovoltaic parameters compared to the device without this surface treatment on the ETL. The removal of the organic residuals from the surface of SnO 2 nanostructures is believed to be the reason for the enhancement of device performance. This finding demonstrates that a proper surface treatment on solvothermally grown SnO 2 nanostructures is important for the fabrication of PSCs. It is noteworthy that there is a strong interplay between the SnO 2 nanostructures and the perovskite layer deposited above. Therefore, each type of SnO 2 nanostructure requires individual optimization of perovskite layer as well as their material interface (ETL/Perovskite) in order to achieve the best processing condition for a certain combination of SnO 2 nanostructured ETLs and perovskite layers, resulting in optimized PSCs. However, these topics are out of the scope of this work. The obtained results are plotted in Figure 9. It is found that the PSC with SnO2 ETL treated by oxygen plasma exhibits significantly improved photovoltaic parameters compared to the device without this surface treatment on the ETL. The removal of the organic residuals from the surface of SnO2 nanostructures is believed to be the reason for the enhancement of device performance. This finding demonstrates that a proper surface treatment on solvothermally grown SnO2 nanostructures is important for the fabrication of PSCs. It is noteworthy that there is a strong interplay between the SnO2 nanostructures and the perovskite layer deposited above. Therefore, each type of SnO2 nanostructure requires individual optimization of perovskite layer as well as their material interface (ETL/Perovskite) in order to achieve the best processing condition for a certain combination of SnO2 nanostructured ETLs and perovskite layers, resulting in optimized PSCs. However, these topics are out of the scope of this work. The obtained results are plotted in Figure 9. It is found that the PSC with SnO2 ETL treated by oxygen plasma exhibits significantly improved photovoltaic parameters compared to the device without this surface treatment on the ETL. The removal of the organic residuals from the surface of SnO2 nanostructures is believed to be the reason for the enhancement of device performance. This finding demonstrates that a proper surface treatment on solvothermally grown SnO2 nanostructures is important for the fabrication of PSCs. It is noteworthy that there is a strong interplay between the SnO2 nanostructures and the perovskite layer deposited above. Therefore, each type of SnO2 nanostructure requires individual optimization of perovskite layer as well as their material interface (ETL/Perovskite) in order to achieve the best processing condition for a certain combination of SnO2 nanostructured ETLs and perovskite layers, resulting in optimized PSCs. However, these topics are out of the scope of this work. Conclusions The impacts of different growth parameters including growth pressure, substrate orientation, DI water-to-ethanol ratios, types of seed layer, amount of acetic acid, and growth time on the morphology of solvothermally grown SnO 2 nanostructures were systematically investigated. It was found that the volume of the precursor solution should be adjusted according to the size of the autoclave reactors to obtain the optimized growth pressure. The orientation of substrates can affect the growth direction of SnO 2 nanorods. The ternary solvent system composed of DI water, ethanol, and acetic acid should be optimized in terms of their volume ratio, which can affect the density of the SnO 2 nanorods and their bundle size. The presence of the seed layer on FTO significantly affects the morphology of SnO 2 nanostructures. The SnO 2 nanoparticle-based seed layers tend to induce more compact growth of SnO 2 nanorod arrays compared to the sample using the seed layer prepared by the magnetron sputtering. The length of SnO 2 nanorods can be increased by using a two-step solvothermal growth method while keeping other morphological properties of SnO 2 nanorod arrays constant, along with increasing thickness, is still challenging. For employing solvothermally grown SnO 2 nanostructures as an ETL in PSCs, a proper surface treatment such as oxygen plasma should be performed on the SnO 2 ETL in order to achieve improved photovoltaic performance of the devices. Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/nano12101686/s1, Table S1: The dimensions of SnO 2 nanostructures grown in different pressure by using different sizes of autoclave reactors; Table S2: The dimensions of SnO 2 nanostructures grown on substrates mounted at different orientations; Table S3: The dimensions of SnO 2 nanostructures grown in a ternary solvent system with different DI water to ethanol ratios; Table S4: The dimensions of SnO 2 nanostructures grown on different types of seed layers; Table S5: The dimensions of SnO 2 nanostructures grown in different acetic acid concentrations; Table S6: The dimensions of SnO 2 nanostructures grown with different growth durations; Figure S1: The topography obtained by the atomic force microscopy for the samples mounted in various orientations during solvothermal growth durations.	2022-05-17T15:07:29.732Z	2022-05-01T00:00:00.000	{ "year": 2022, "sha1": "882ce6bc2a9391fc9922d8eb8ff1c0528cb3f953", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2079-4991/12/10/1686/pdf?version=1652601771", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "b88a207aedfcffa61d278e8b125f519ad426e8ff", "s2fieldsofstudy": [ "Materials Science", "Engineering", "Physics" ], "extfieldsofstudy": [ "Medicine" ] }
259299056	pes2o/s2orc	v3-fos-license	Tetra-block: ultrasound femoral, lateral femoral-cutaneous, obturator, and sciatic nerve blocks in lower limb anesthesia: a case series Background The gold standard anesthesiologic procedure for urgent femur fracture surgery is Spinal Anesthesia. It is not always feasible because of patients' severe comorbidities and difficulties in optimizing drug therapy in the appropriate time frame such as discontinuation of anticoagulant drugs. The use of four peripheral nerve blocks (tetra-block) can be a winning weapon when all seems lost. Case presentation We present, in this case series, three Caucasian adult femur fractures (an 83-year-old woman, a 73-year-old man, and a 68-year-old woman) with different and major comorbidities (cardiac or circulatory disorders on anticoagulants therapy that were not discontinued on time; breast cancer and others) underwent the same anesthesiologic approach in the urgent setting. Ultrasound peripheral nerve blocks, that is femoral, lateral femoral cutaneous, obturator, and sciatic with parasacral approach were successfully performed in all patients who underwent intramedullary nailing for intertrochanteric fracture. We evaluated the adequacy of the anesthesia plane, postoperative pain control with the VAS scale, and the incidence of postoperative side effects. Conclusions Four peripheral nerve blocks (Tetra-block) can be alternative anesthesiologic management in urgent settings, in patients where drug therapy cannot be optimized, as in antiplatelet and anticoagulant therapy. Background Femur fracture (FF) represents an event with disabling outcomes in elderly patients, with a significant impact on the quality of life and public health in general. FF in the elderly is, in most cases, the result of an accidental fall or even minor trauma and is often associated with osteoporosis/low bone mass and other conditions [1] such as functional impairment of the lower limbs, Parkinson's disease, and visual impairment, [2] which can considerably increase the risk of falls. A Systematic Review of 72 studies carried out in 63 different countries revealed that Italy is among the countries with the highest incidence of hip fractures, annually recording an incidence of > 300 per 100,000 inhabitants for women and > 150 for men [3]. In addition to increased mortality in elderly patients, [4] FF sometimes has a devastating impact on quality of life, leading to the risk of reduced mobility with limitation or loss of autonomy and the inability to return to pre-trauma conditions [5]. Guidelines of Italian Society for Orthopedics and Traumatology (SIOT) recommend surgery for patients with FF on the day of arrival at the hospital (within 24 h of arrival) or, at the latest, the next day (within 48 h of arrival) [6]. The anesthetic approach to orthopedic surgery can be General Anesthesia (GA), Neuraxial Anesthesia (NA), and Loco-Regional Anesthesia (LRA). The literature does not definitively clarify the statistically significant difference between GA and Peripheral Nerve Blocks (PNB) for mortality and postoperative complications [7]. Nevertheless, some studies clearly show lower mortality after 30 days in patients undergoing Regional Anesthesia (RA) with PNB and highlight better pain control in the postoperative period, better hemodynamic stability, faster functional recovery, and a clear reduction in respiratory complications associated with GA [8]. Ultrasound femoral, lateral femoral-cutaneous, obturator, and sciatic (parasacral approach) nerve blocks (tetra-block) can be an efficient anesthesiologic approach when all seems lost in patients undergoing surgical femur fracture treatment. This anesthesiologic management showed several advantages including reduced hemodynamic impact because this technique had only affected the area of the body that undergoing the surgery, reduced the risk of post-operative fatigue and vomiting as well as improved post-operative pain management. Ultrasound peripherical nerve blocks are a safe approach in the hands of clinicians with considerable experience in this area. The proposed anesthetic approach may be particularly suitable for fragile patients or for those who could obtain greater benefits, given their comorbidities, from opioid-free anesthesia [9]. NA is not always feasible, both due to the lack of pharmacological optimization in patients who are often subject to anticoagulant therapy and the difficulty to position the patient himself, as well as the pain felt. The approach with LRA can be considered valid provided that it is performed by expert operators since the Local Anesthetic Systemic Toxicity (LAST) could be the greatest risk to encounter [10]. We propose the cases of three patients with multiple comorbidities, candidates for urgent surgery, for whom pharmacological optimization could not occur. Case presentation The patients agreed to the use of their data in the publication of this case series for scientific and clinical purposes. Ethics committee approval was not sought for this retrospective study since we analyzed data collected during routine clinical practice. Case 1 An 83-year-old Caucasian woman (Body Mass Index (BMI), 27.34 kg/m 2 ; weight, 70 kg; height, 160 cm) suffered from Type 2 Diabetes Mellitus, Chronic Obstructive Pulmonary Disease (COPD) with pleurogenic results, left pulmonary lobectomy, Chronic Kidney Disease III stage, persistent Atrial Fibrillation under treatment, Ejection Fraction (EF) 30% and a previous right femur fracture arrived to the emergency room of a small town following a fall during the night. The patient presented evident deformity of the left lower limb, so an x-ray was performed which showed AO/OTA 31A2 type left intertrochanteric fracture ( Fig. 1) [11,12]. In order to contain transport times, the patient was transported by helicopter to the base of the Cardarelli Hospital of Naples (Italy) and subsequently admitted to the Orthopedics and Traumatology Department of Federico II University Hospital of Naples (Italy). The patient's pharmacotherapy also included Novel Oral Anticoagulants (NAO), normally taken by the patient a few hours before the fall. The preoperative hemoglobin level was 9.0 g/dl. Given the orthopedic surgical treatment of intramedullary nailing (Fig. 2) that could cause major blood loss, Erythropoietin (450 UI/kg), and Ferinject (20 mg/ kg), dosed according to the patient's weight, were administered. PNB, the anesthetic technique choice, was performed with Mepivacaine 1% 200 mg, and Ropivacaine 0.375% 75 mg plus Dexamethasone 8 mg. Case 2 A 73-year-old Caucasian man (BMI, 30.64 kg/m2; weight, 96 kg; height, 177 cm), who suffered from Type 2 Diabetes Mellitus, Hypertension, severe Aortic Stenosis, and previous Covid19 infection, arrived at the emergency room of a local hospital due to chest and right hip pain after accidental fall caused by syncopal episode. After practicing the appropriate diagnostic tests NSTEMI was diagnosed. In addition, the x-ray of the painful hip showed AO/ OTA 31A1.2 type right intertrochanteric fracture ( Fig. 3) [11,12]. The patient was immediately transferred to the Cardiology, Hemodynamics, and Cardiac Intensive Care Unit (CICU) of Federico II University Hospital of Naples (Italy) where he underwent a revascularization procedure with the use of drug-eluting stents, heart failure after myocardial infarction (EF < 21%) The patient took pharmacotherapy for his pathologies and was on Dual Antiplatelet Therapy (DAPT) for the acute cardiac event. After a multidisciplinary meeting between the anesthesiologist, cardiologist, and orthopedic surgeon, it was decided to suspend Ticagrelor 4 days earlier and switch it with Tirofiban, which had to be discontinued 4 h before surgery. The preoperative hemoglobin level was 8.2 g/ dl. Given the orthopedic surgical treatment of intramedullary nailing of the right femur that could cause major blood loss, Ferinject (20 mg/kg) dosed-considering the patient's weight-was administered. Next, he underwent orthopedic surgical treatment of intramedullary nailing of the right femur (Fig. 4). PNB, the anesthetic technique choice, was performed with Mepivacaine 1% 200 mg, and Ropivacaine 0.375% 75 mg plus Dexamethasone 8 mg. Case 3 A 68-year-old Caucasian woman (BMI, 37.29 kg/m2; weight, 104 kg; height, 167 cm) suffered from Hypertension, Toxic Multinodular Goiter, Dyslipidemia, Obesity class II, Chronic Kidney Disease III stage and Invasive Ductal Carcinoma (IDC) with the convolution of metastasis to the vertebral column by D9 to L5, lungs, and bones. The patient presented evident deformity of the right lower limb, so an x-ray was performed. It showed AO/OTA 31A3.1 type right intertrochanteric fracture (Fig. 5) [11,12]. It was a pathological fracture; It was a pathological fracture, that is a spontaneous fracture not secondary to a traumatic event while she was admitted to our Oncology Department of Federico II University Hospital of Naples (Italy) for the treatment of the carcinoma, subsequently admitted to the Orthopedics and Traumatology Department to undergo surgical treatment of intramedullary nailing of the right femur (Fig. 6). In addition, the patient had extensive deep vein thrombosis of the medial twin vein and the left popliteal vein. The patient's pharmacotherapy also included Fondaparinux 10 mg subcutaneously once daily. PNB, the anesthetic technique choice, was performed with Mepivacaine 1% 200 mg, and Ropivacaine 0.375% 75 mg plus Dexamethasone 8 mg. Anesthesiologic management In the operating room, venous access was placed (16 or 18 Gauge) and antibiotic prophylaxis was administered (Cefazolin 1 or 2 gr. iv, or in case of allergy, Clindamycin 600 mg iv). Pantoprazole 40 mg iv was also administered. Pulse oximetry (SpO2), heart rate (HR), body temperature (C°), continuous invasive arterial (cIBP), and cerebral oximetry with ForeSight were monitored. Pre-load was performed with crystalloid 500 mL and Ondansetron 8 mg, and pre-procedural sedation was performed with Midazolam 0.01-0.03 mg/Kg. The patients underwent PNB and intra-operative sedation with Dexmedetomidine 0.7 gamma/Kg/h. All patients received O2-therapy with a nasal cannula with a flow rate of 2L/min. Ultrasound peripherical nerve block Ultrasound femoral nerve block (Fig. 7) The patient was positioned supine with the legs slightly abducted: the inguinal and thigh region were disinfected and sterile towels were placed to delimit the field. The ultrasound device was positioned contralaterally to the limb to be treated. The physician applied the probe to the patient's groin with the point of the probe conventionally placed on the right side of the patient. Using an ultrasound (US) transducer (Sonosite HLF38 × 13-6 MHz, Fujifilm Sonosite Europe, Amsterdam, Netherlands), the inguinal ligament was visualized as a hyperechoic structure; sliding in the caudal direction with the probe, the femoral vein, compressible, and the femoral artery, non-compressible and pulsatile, were identified. It was scrolled laterally until the femoral nerve was identified, which appeared like a hyperechoic triangle. From the lateral side of the transducer (lateral-to-medial orientation), an 85-mm long 21-Gauge 30° tip (Vygon Value Life, Italy) to ensure complete visualization of the needle tip, was inserted. The correct position of the needle tip was double-checked through the progressive injection of 3 ml of Saline Solution. An anesthetic solution of Ropivacaine 0.375% (dose 26.25 mg) plus Mepivacaine 1% (dose 70 mg) and Dexamethasone 4 mg, total volume 15 ml, was subsequently injected until a complete detachment of the iliac fascia. Ultrasound lateral femoral-cutaneous nerve block (Fig. 8) Once the Femoral Nerve block was completed, we proceeded with a lateral scrolling of the probe and we went to search for the compartment that was created between the Fascia Lata, the Tensor Muscle of the Fascia Lata and the Sartorio Muscle. There we could visualize the Lateral Femoral-Cutaneous Nerve with the characteristic "eye" appearance. From the lateral side of the transducer (lateral-to-medial orientation), an 85-mm long 21-Gauge 30° tip was inserted (Vygon Value Life, Italy) to ensure complete visualization of the needle tip. An anesthetic solution of Ropivacaine 0.375% (dose 1.875 mg) plus Mepivacaine 1% (dose 5 mg), total volume 1 ml, was subsequently injected. Ultrasound obturator nerve block (Fig. 9) Starting again from the patient's groin, after visualizing the femoral artery and vein, the probe was scrolled 2-4 cm caudally and later medially, until the triple layering of the Adductor Longus, Adductor Brevis and Adductor Grande Muscles were visualized. The visualization of the hyperechoic intermuscular septa allowed the branches of the obturator nerve to be shown as flat, oval structures. From the lateral side of the transducer (lateral-to-medial orientation), an 85-mm long 21-Gauge 30° tip (Vygon Value Life, Italy) was inserted to ensure complete visualization of the needle tip. The correct position of the needle tip was double-checked through the progressive injection of 3 ml of Saline Solution. An anesthetic solution of Ropivacaine 0.375% (dose 9.375 mg) plus Mepivacaine 1% (dose 25 mg), total volume 5 ml, was later injected. (Fig. 6). (Fig. 10-11) Lastly, we performed the Sciatic Nerve Block with a Parasacral approach to benefit from the analgesia provided by the blocks already performed. The patient was positioned in lateral decubitus, with the femur fracture limb positioned superiorly. The sacral region was disinfected and sterile towels were placed to delimit the field. A convex probe was used. A line, which connected the Greater Trochanter and the Postero-Superior Iliac Spine, was drawn. Along this line, with appropriate pressure, rotation, and tilting movements, the various structures could be identified, including the Piriformis Muscle, with a hypoechoic appearance, and the Sciatic Nerve, with a typical "honeycomb" appearance. Ultrasound sciatic nerve block From the lateral side of the transducer (lateral-tomedial orientation), a 100-mm long 21-Gauge 30° tip (Vygon Value Life, Italy) was inserted to ensure complete visualization of the needle tip. The correct position of the needle tip was double-checked through the progressive injection of 3 ml of Saline Solution. Electrical nerve stimulation (ENS) assisted technique was performed to confirm the target nerve. A stimulating current was used (0.4 mA), when the needle was near the target nerve and the lowest twitch response was seen or paresthesia was felt. An anesthetic solution of Ropivacaine 0.375% (dose 37.5 mg) plus Mepivacaine 1% (dose 100 mg) and Dexamethasone 4 mg, total volume 21 ml, were later injected. At the end of the execution of the ultrasound PNB, the motor and sensory blocks were tested, and evaluated respectively with the " Bromage scale" and with the " Hollmen scale". All the patients had an adequate anesthetic plane with a complete motor block (Bromage scale 1) and sensory block tested via pinprick and ice tests that showed sensation loss (Hollmen Scale 4). All patients were administered an interview at the end of surgery which showed they tolerated the anesthesiological procedure excellently without any pain or discomfort throughout the surgical phase. Post-operative management The patients were evaluated by clinicians every 6 h in each postoperative period (first 24 h) to determine: VAS; the presence of adverse effects such as nausea, vomiting, pruritus, shivering, and motor recovery time. Pharmacological therapy was based on the patient's response. In the postoperative period, VAS assessment was carried out with a 10 cm long line with verbal anchors at either extremities ("no pain" on the far left and "the most intense pain" on the far right). The patient marked a point on the line corresponding to the rating of pain intensity. After surgery, we administered intravenous Paracetamol 1 g 3 times a day. Oxycodone (up to 0.1 mg/Kg) was available as a rescue dose. Pain control was considered good in the case of a VAS score of less than 4. No patient required a rescue dose and no side effect was observed. In the 24 h following the operation, no patient required the use of rescue therapy. The median (range) times to full recovery from motor blockade was 12 (4-18) hours and 10 (4-12) hours for sensory blockade. Follow up On the first day after surgery, rehabilitation was initiated and training for standing was possible. Eventually, they were transferred to a rehabilitation hospital on the 15th day after surgery. These types of ultrasound-guided PNBs were appropriate for the procedure, showing fewer risks or side effects and this anesthesiologic management is replicable. Discussion The presence of numerous comorbidities and contraindications addressed us to the choice of anesthesiologic conduct with PNB. As far as we know, this is the first study that shows this anesthesiologic approach for surgical FF treatment. We define "Tetra-Block" as the four performed ultrasound PNBs. As to the first patient, GA appeared to be contraindicated due to the comorbidities. She presented a pulmonary condition that required suitable evaluations, both in terms of airway management and for any post-operative complications [13]. COPD is associated with increased morbidity, mortality, and length of stay [14]. GA appears to be an independent risk factor in the development and worsening of dementia [15]. The choice of GA was precluded for the second patient due to his categorical refusal. The third patient had lung metastasis and several randomized controlled trials were in progress to provide a better understanding of how volatile and intravenous hypnotics impact influenced cancer progression and to evaluate the effect of the anesthesia techniques on the immune system and tumor microenvironment [16]. Spinal anesthesia (SA) was the first choice for FF. It reduced the risk of postoperative complications such as hypoxia, myocardial infarction, deep vein thrombosis, pneumonia, pulmonary embolism, and delirium [17]. SA, in this case, was not possible due to some absolute contraindications: as to the first patient, a nephropathic subject, NAO therapy was not suspended, therefore blood coagulation framework was compromised; as regards the second patient, it was not possible to proceed with SA due to severe Aortic Stenosis. In the case of Aortic Stenosis, it has been demonstrated that it is still possible to carry out a NA using a Low-Dose Spinal [18]. This was not possible in our management due to technical difficulties of approach and surgical times. The possibility of operating with a loco-regional approach is supported by the literature, and the outcome of frail patients treated with PNB compared to those subjected to GA is better, both in terms of mortality and hospitalization time [19]. The literature testifies how hip and femur surgery can be performed with the execution of a Lumbar Plexus Block, associated or not with the Sciatic Nerve Block [20]. ESAIC/ESRA guidelines defined that deep nerve procedures, like Lumbar Plexus block, should be performed according to the recommendations for neuraxial procedures. If the INR is not below the minimum recommended level, regional anesthetic management should depend on the compressibility of the puncture site, the vicinity of (large) blood vessels, and/or neuraxial structures [21]. In such fragile patients, an opioid-free anesthetic technique can be used, considerably reducing the impact of adverse events and side effects of the use of opioids, but not eliminating adverse hemodynamic events [22]. You can adapt your anesthesiological management to every patient's needs. Our anesthesiologic approach to ultrasound PNB, named "Tetra-block", has some critical issues. First of all, multiple blocks are necessary to ensure an anesthesiologic plane for the procedure, with the need to have operators with considerable experience in the LRA. The execution by operators with little experience exposes us to the risk of LAST, 10 especially in patients with severe hypoalbuminemia or debilitated, or of inadequate anesthesia, with poor satisfaction by the patient and the surgical team. It is to be noted that ropivacaine 75 mg and mepivacaine 200 mg were the total dose. This dose is very far from the toxic dose for ropivacaine (3 mg/kg) and mepivacaine (6 mg/kg). Like all ultrasound nerve blocks, they are an alternative anesthesiologic strategy in many situations but do not forget that they need competence, ability, experience, and appropriateness. Last but not least, the cooperation of the surgical team; even the position of the patient can be difficult, the other lower limb is not anesthetized and patients often suffer from hip arthrosis. On the other hand, the execution of these PNBs does not involve the suspension of anticoagulant and antiplatelet drugs. Considering the various anesthetic alternatives, it was decided to proceed with the loco-regional approach. Conclusions Our report illustrates the use of PNB with minimally invasive hemodynamic monitoring as a valid alternative to spinal anesthesia in patients with FF who were undergone urgent surgery. The flexibility that can be achieved with LRA must prompt us to reflect on what the actual usage scenarios might be. It must be seen as an additional option in our anesthetic arsenal. This type of approach may be useful in patients where drug therapy cannot be optimized, as in antiplatelet and anticoagulant therapy. It is always necessary to carefully evaluate any limitations related to patient compliance, the risk of Systemic Intoxication from Local Anesthetics, and any hemodynamic repercussions. Study design We presented a series of three clinical cases in the urgent setting, in which we chose a locoregional anesthesia plan with four different peripheral nerve blocks that we called "Tetra-block". surgeries and follow-ups, furthermore, analyzed and interpreted the patient data regarding the vital sign and sensory block level. AC and CI performed the peripheral nerve blocks and were major contributors to writing the manuscript. GS and MV raised the initial idea and supervised all the perioperative management and patient care. AB, AC, and AI were the operating surgeons and provided the orthopedic material contained in the article. All authors read and approved the final manuscript. Funding No funding to declare. Availability of data and materials All data generated or analyzed during this study are included in this published article. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patients for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.	2023-07-01T13:33:40.563Z	2023-07-01T00:00:00.000	{ "year": 2023, "sha1": "5cb268e239ed46b1a704e4023d295b19e9d3ec0f", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Springer", "pdf_hash": "5cb268e239ed46b1a704e4023d295b19e9d3ec0f", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
248402092	pes2o/s2orc	v3-fos-license	Barriers and Facilitators Influencing Arab Muslim Immigrant and Refugee Women’s Breast Cancer Screening: A Narrative Review Introduction: This narrative review identifies barriers and facilitators influencing breast cancer screening among Arab Muslim immigrants and refugees. Low participation rates create health inequities in breast cancer screening among this population. Method: A systematic search of peer-reviewed empirical articles was performed. PRISMA, CASP, and MMAT checklists were used to appraise the studies. Results: Results include 3 themes: individual, health care system and health providers, and cultural factors. Discussion: Lack of fluency in the new country’s language, lack of knowledge, and poor exposure to breast cancer screening may contribute to the Arab Muslim women’s vulnerability to undiagnosed or delayed breast cancer diagnosis. Introduction Breast cancer represents the most frequent neoplasm affecting women globally . Breast cancer has now surpassed lung cancer in terms of incidence and mortality rates and one in every eight women faces breast cancer during their lifetime (World Health Organization [WHO], 2021). Although breast cancer is not preventable, risk factors can be decreased to reduce morbidity and mortality (Azubuike et al., 2018;Chetlen et al., 2016;Coleman, 2017). Women living in sub-Saharan African countries present the highest mortality rate of breast cancer globally (Karbakhsh, 2021). Breast Cancer Among Arab Muslim Immigrant and Refugee Women Socio-cultural factors influence Arab Muslim immigrant and refugee women's breast awareness and explain low participation in early screening programs (Elobaid et al., 2016;Gray et al., 2017;Racine et al., 2021aRacine et al., , 2021bVahabi et al., 2017;Vrinten et al., 2016). A recent study indicates that health literacy and language barriers are the most important factors limiting access to breast and cervical cancer prevention among Arab Muslim refugees in the United States (Siddiq et al., 2020). Women's gender roles and the lack of open discussion about breast cancer may generate shame and taboos among Arab Muslim women (Siddiq et al., 2020). For some women, getting cancer is a matter of destiny, which ultimately decreases breast screening rates (Banning & Hafeez, 2009;Salman, 2012;Shirazi et al., 2015;Zorogastua et al., 2017). Although Islam supports women's duty to take care of their health (Siddiq et al., 2020), religious beliefs may create fatalism, placing breast health beyond women's locus of control (Pinchas-Mizrachi et al., 2021). Beliefs in modesty, women's willingness or tolerance to being physically touched, and a male provider may decrease participation in breast cancer screening and create embarrassment or fear of clinical breast examination and mammography (Siddiq et al., 2020). The fear of pain from the mammogram and of the results negatively impact breast cancer screening practices. Lack of understanding of the receiving country health care system and lack of linguistic, geographic, and financial accessibility reduce breast cancer screening (Al-Amoudi et al., 2015;Elobaid et al., 2016;Islam et al., 2017;Kamaraju et al., 2019;Mellon et al., 2013;Raymond et al., 2014;Saadi et al., 2012Saadi et al., , 2015Wang et al., 2019;Zorogastua et al., 2017). Lack of health care providers' support and discrimination within the health care systems further decrease breast cancer screening among Arab Muslim immigrant and refugee women (Racine et al., 2021b;Tackett et al., 2018). Despite medical and technological advancement, health disparities related to breast cancer exist (Gray et al., 2017;Karbakhsh, 2021), and these inequities likely affect Arab Muslim refugee women due to disruptions in their home countries and lack of breast health care in refugee camps (Al-Smadi et al., 2017;Bakkal Temi et al., 2017;Bowser et al., 2017;Gammouh et al., 2015;Mansour et al., 2018;Samadi, 2017;Siddiq et al., 2020). Low rates of breast cancer screening in Muslim native and refugee women have been documented in Gulf countries, Jordan, Lebanon, Turkey, and other countries in the Arab world (Al-Amoudi et al., 2015;Donnelly & Hwang, 2015;El-Arnaout et al., 2019;Elobaid et al., 2016;Gammouh et al., 2015;Hashim et al., 2018;Kamimura et al., 2018;Mansour et al., 2018;Siddiq et al., 2020). Low survival rates and late-stage diagnosis have been reported in Libya and Malaysia (Coleman, 2017). More than half of women in the Middle East present lymph nodes indicating an advanced stage of the illness coupled with a low survival rate (Ghoncheh et al., 2015). Donnelly and Hwang (2015) mentioned that Qatari women are more likely to develop breast cancer at a younger age than Western women. Researchers suggest that Arab Muslim refugee women who migrated to Western Europe and North America are likely to present poorer health outcomes when it comes to breast cancer than women in the general population (Abolfotouh et al., 2015;Saadi et al., 2015;Salman, 2012;Shirazi et al., 2015;Tackett et al., 2018;Vahabi et al., 2017). Increased migration to Western nations may increase breast cancer morbidity as Arab Muslim refugee women present minimal exposure to breast cancer screening programs in their home countries coupled with a lack of understanding of preventive health care (Karbakhsh, 2021;Mansour et al., 2018;Siddiq et al., 2020). Studies in Turkey, Lebanon, and Jordan indicated that the increased influx of Syrian refugees added pressure on these countries' health care systems affecting access to preventive and curative services (El-Arnaout et al., 2019). Finally, resettlement challenges like learning a new language and seeking employment may impinge breast cancer screening as health promotion and prevention may not be seen as a priority compared to social and economic integration activities (Siddiq et al., 2020). Design This integrative review focused on examining the barriers and facilitators of breast cancer screening in Arab Muslim immigrant and refugee women. The review relies on a narrative approach to synthesize the data (Popay et al., 2006). The research question was: Research Question 1: What barriers and facilitators of breast cancer screening influence Arab Muslim immigrant and refugee women's access and utilization? Search Methods A comprehensive systematic search of English-language, original empirical peer-reviewed papers was performed from January 2000 to July 2021. We searched MEDLINE, CINAHL, Embase, Cochrane Library, Web of Sciences, and PubMed databases. Search terms included "Breast Cancer Screening" and "Refugee." Then, we added "immigrant," "Muslim," "women," "refugee," and "Islam" as keywords. Cross-sectional, descriptive, randomized controlled trials, qualitative, and mixed-methods design studies were retrieved. We excluded non-empirical and theoretical papers, editorials, gray literature, letters to the editor, and conference abstracts. The initial database searches yielded 231 records. The Medline search yielded 58 peer-reviewed articles that met the inclusion criteria. Out of these 58 publications, three were duplicates, and 18 did not meet the inclusion criteria. A total of 37 articles was retrieved from Medline. The CINAHL search retrieved 139 articles. Out of 139 articles, 11 were duplicates, 125 were excluded, and three were selected. The PubMed search produced 34 results, but 11 articles were duplicates, 21 were rejected, and two were retrieved. Three articles were retrieved through a manual search. Figure 1 illustrates the process of selection using Preferred Reporting Items for Systematic Reviews and Meta-Analysis diagram (PRISMA) (Moher et al., 2009) Appraisal of the Studies We used the Critical Appraisal Skills Program (CASP) checklists (Critical Appraisal Skills Programme, 2020) and the Mixed Methods Appraisal Tool (MMAT) (Hong et al., 2018) to appraise the quality of the studies. CASP provides various checklists to effectively assess a range of studies using a ranking system to ensure the quality and rigor of the studies. A positive answer to any question on the CASP checklists was worth one point, and a negative response or a zero score was given if the answer was unclear. The Mixed Methods Appraisal Tool (MMAT) Version 2018 (Hong et al., 2018) was used to appraise the quality of the three mixed-method studies. The MMAT starts by asking two screening questions: 1) Are there clear research questions; and 2) Do the collected data address the research questions? As these three studies met the screening questions, we assessed the five statements for mixed-method studies (Hong et al., 2018). A yes response was worth 1 point, while a no or cannot tell answer was assigned zero. This scoring scheme is similar to the one we applied with CASP checklists to maintain consistency. The two team members participated in the quality appraisal of the studies. Disagreements were solved through consensus. Data Extraction and Synthesis The review relies on a textual approach to data synthesis (Popay et al., 2006). The two team members read the articles and extracted the data. A synthesis table was used to organize and critique the extracted data into meaningful units of analysis (Pinch, 1995). Study title, author, journal, year of publication, purpose, research questions, methodology, data collection, statistical tests used, results, and conclusions were extracted (Table S1). We used thematic analysis to interpret the findings. Search Results Forty-five publications met the inclusion criteria for the review. The characteristics of the studies are presented in Table S1. Thirty-one studies (69%) were conducted in the United States of America, nine (20%) in Canada, four (9%) in the United Kingdom, and one study (2%) in Turkey. In terms of study designs, 13 (28.9%) were qualitative, 9 (20%) quasi-experimental, 8 (18 %) descriptive quantitative, 5 (11%) cross-sectional, 4 (8.9%) retrospective cohort design, 3 (6.6%) mixed methods, 2 (4.4%) systematic reviews, and 1 (2.2%) randomized controlled study. Nine studies guided by a quasi-experimental design were retrieved (Dunn et al., 2017;Gondek et al., 2015;Kamaraju et al., 2018Kamaraju et al., , 2019Padela et al., 2018Padela et al., , 2019Percac-Lima et al., 2012, 2013Pratt et al., 2020). These studies described interventions to promote breast cancer screening in Arab Muslim immigrant or refugee women. A mean score of 5.89 out of 10 was assigned to quasi-experimental studies indicating quality but with some limitations related to lack of reporting of effect sizes and experimental conditions and lack of generalizability. Thirteen qualitative studies were retrieved (Al-Amoudi et al., 2015;Ayash et al., 2011;Banning & Hafeez, 2010;Islam et al., 2017;Kawar, 2013;Mellon et al., 2013;Padela et al., 2016;Racine et al., 2021a;Raymond et al., 2014;Saadi et al., 2012Saadi et al., , 2015Shirazi et al., 2013Shirazi et al., , 2015. Qualitative studies investigated barriers and facilitators related to breast cancer screening, while others focused on Arab Muslim immigrant and refugee women's beliefs and attitudes about screening practices. Qualitative data were collected through focus groups, face-to-face and telephone interviews. A mean score of 9.54 was compiled for the qualitative studies indicating a very high quality of the selected studies. Three mixed-method designs were retrieved (Banning & Hafeez, 2009;Wang et al., 2019;Zorogastua et al., 2017). Each mixed-method study scored 5 out of 5 statements supporting the robustness of the methodology and the trustworthiness of the results. Thirteen descriptive and cross-sectional quantitative studies documented predictors of breast cancer screening among Arab Muslim women (Alatrash, 2020;Hasnain et al., 2014;Jaffee et al., 2021;Kamimura et al., 2018;Kobeissi et al., 2014;Lofters et al., 2018;Padela et al., 2015;Racine et al., 2021b;Rashidi & Rajaram, 2000;Salman, 2012;Schwartz et al., 2008;Shirazi et al., 2006;Szczepura et al., 2008). Quantitative data were collected through written or telephone surveys and follow-up questionnaires. Quantitative descriptive and cross-sectional studies received a mean score of 8.77. This score points to the high quality of the selected studies and confidence in the validity of the results. A randomized controlled study (Erenoglu & Sozbir, 2020) scored 5 out of 11. This study was included as it met the criteria of a straightforward research question, randomization, and the value and impact of the study. The effect size, generalizability, and costs/benefits ratio were not mentioned, decreasing the score. Two systematic reviews (Bowser et al., 2017;Schoueri-Mychasiw et al., 2013) were retrieved and scored 9 out of 10 points, indicating a very high quality of the findings. Four studies involving a retrospective cohort design (Lofters et al., 2019;Price et al., 2010;Vahabi et al., 2016Vahabi et al., , 2017 were selected and scored 9 out of 10 points on the CASP cohort checklist. This score indicated a very high quality of the research and confidence in the results and the findings' applicability in practice. The number of subjects included in the samples varied from 12 (Islam et al., 2017) to 238 (Vahabi et al., 2017). Thematic analysis was used to analyze and synthesize the data (Braun & Clarke, 2006). Thematic analysis generated three themes: (a) Individual factors, (b) health care system and health providers' factors, and (c) cultural factors. Theme 1: Individual Factors Individual factors include personal barriers and facilitators affecting the uptake of breast cancer screening practices of Arab Muslim immigrant and refugee women. Language and education, time constraints, breast cancer beliefs, and contextual facilitators are included as subthemes. Time Constraints Issues of time, geographic accessibility of the clinics, lack of transportation, and family and child care commitments constrained Arab Muslim women's access and participation in breast cancer screening practices (Kamaraju et al., 2018;Saadi et al., 2015;Zorogastua et al., 2017). More urgent priorities like searching for employment, learning the language, and child care may affect breast cancer screening practices (Wang et al., 2019). Breast Cancer Beliefs Several studies indicated that Arab Muslim women's health beliefs affected breast cancer screening behaviors. For instance, self-efficacy, perception of risk (susceptibility to breast cancer), perceived importance of breast self-examination (BSE), perceived importance of mammography, and intention to be screened positively relate to active participation in screening activities (Hasnain et al., 2014;Islam et al., 2017;Price et al., 2010;Zorogastua et al., 2017). Studies using the Health Belief Model reported that perceived seriousness of breast cancer and health motivation were associated with increased screening behaviors among Arab American women (Schwartz et al., 2008;Zorogastua et al., 2017). Perceived benefits of prevention, perceived seriousness of the illness, health motivation, higher level of education, confidence, and skills in performing breast selfexamination (BSE) correlated with increased awareness of breast cancer risks (Shirazi et al., 2006). Contextual Factors Contextual factors have a significant impact on women's mammography use. Having a higher knowledge about breast cancer impacts Arab Muslim women's mammogram uptake (Kobeissi et al., 2014), the same as having a relative who presented breast cancer (Padela et al., 2015). Racine et al. (2021b) found that Muslim Arab refugee women with family responsibilities such as looking after children and other relatives at home had a higher mammography rate. In addition, Arab Muslim women who realized the importance of mammography had more screening (Hasnain et al., 2014). Media, magazines, and newspapers indirectly increased women's knowledge on breast cancer by increasing awareness of the issue and may lead to increased participation in screening programs (Banning & Hafeez, 2009, 2010. Health Care System Breast cancer screening practices involve interactions with the receiving country's health care system and health care professionals. Years of residence in the receiving country, degree of acculturation, and citizenship influence Arab Muslim women's access to breast preventive services (Hasnain et al., 2014;Islam et al. 2017;Kamaraju et al., 2019;Kawar et al., 2013;Lofters et al., 2019;Padela et al., 2015;Wang et al., 2019;Zorogastua et al. 2017). Exposure and participation in preventive breast cancer programs in the home country were likely to affect post-resettlement experiences with the receiving country's health care system. Mellon et al. (2013) indicated that family experiences of discrimination with the receiving country's health care system affected Arab American women's breast cancer behaviors. Religious discrimination within the health care system constituted another obstacle to clinical breast examination and mammograms (Padela et al., 2015). Health Care Providers However, guidance and support of health care providers facilitated access and utilization of breast cancer preventive care (Kamaraju et al., 2018). Jaffee et al. (2021) underlined that a lack of providers' support negatively affects Arab Muslim women's BC screening behaviors and increases women's mistrust of health care professionals. Al-Amoudi et al. (2015) indicated that Somali immigrant women who did not receive any recommendation for BCS from their physician negatively affected breast cancer behaviors. Bowser et al. (2017) reported statistically significant relationships between physician gender and breast cancer screening practices among Arab Muslim women from the Middle East and North African countries. Ability to speak Arabic and preferences for health care providers of similar religious backgrounds or female physicians or nurses play a crucial role among Arab Muslim immigrant and refugee women (Banning & Hafeez, 2009;Racine, 2021a;Rashidi & Rajaram, 2000;Raymond et al., 2014;Zorogastua et al., 2017). Health Care System and Health Care Providers' Facilitators Access to publicly funded health care systems facilitated breast cancer screening practices (Bowser et al., 2017;Islam et al., 2017). Having health insurance and access to a primary physician increased women's access to mammography (Padela et al., 2015). Trusting relations and communications with health care providers were fundamental in supporting Arab Muslim immigrant and refugee women's use of mammograms (Bowser et al., 2017). Studies showed that Arab Muslim immigrant and refugee women preferred female physicians from the same ethnocultural groups and a provider able to speak Arabic (Kamimura et al., 2018;Lofters et al., 2018;Padela et al., 2016;Saadi et al., 2012;Vahabi et al., 2016Vahabi et al., , 2017. Appointment reminders and personal contact from health providers facilitated breast cancer screening (Banning & Hafeez, 2009;Rashidi & Rajaram, 200;Raymond et al., 2014;Saadi et al., 2015). Theme 3: Cultural Factors This theme includes cultural beliefs and facilitators that may affect Arab Muslim immigrant and refugee women's breast screening practices (Alatrash, 2020;Racine et al., 2021a;Zorogastua et al., 2017). The embarrassment of revealing breasts represented obstacles among Arab Muslim women in Michigan (Alatrash, 2020). Arab Muslim women may present shyness, modesty, and embarrassment in their home countries and abroad due to the belief that breasts must be hidden (Abolfotouh et al., 2015;Banning & Hafeez, 2009;Elobaid et al., 2016). Stigma about cancer creates barriers affecting Arab Muslim women's breast awareness and participation in breast cancer screening activities (Racine et al., 2021b;Zorogastua et al., 2017). Some traditional beliefs affected women's views toward breast cancer screening (Al-Amoudi et al., 2015;Islam et al., 2017;Padela et al., 2016Padela et al., , 2019Shirazi et al., 2013). Kawar et al. (2013) reported that embarrassment, family relationships, and traditional healers' consultations decreased motivation to seek breast cancer preventive care. Religious beliefs might decrease motivation to screening. A few women thought that breast cancer came from divine intervention and that prayers can help cope with illness (Hashim et al., 2018). Fatalistic attitudes were among the most frequently mentioned cultural barriers in Arab Muslim immigrant and refugee women (Al-Amoudi et al., 2015). Cultural Facilitators The literature indicates that gender, language, and religion facilitate breast cancer screening education and uptake among Arab Muslim immigrant and refugee women. Eight quasi-experimental intervention studies focused on promoting breast cancer screening in Arab Muslim immigrant or refugee women (Ayash et al., 2011;Dunn et al., 2017;Erenoglu & Sozbir, 2020;Gondek et al., 2015;Kamaraju et al., 2018;Padela et al., 2018Padela et al., , 2019Percac-Lima et al., 2012, 2013Pratt et al., 2020). Educational interventions consisted of education programs or programs combined with navigation assistance, group discussions and messages, culturally appropriate educational materials, and video presentations. Individual interventions with specific attention to gender and language significantly decreased breast cancer inequities among Arab American immigrant women in New York City (Ayash et al., 2011) and in Bosnian and Somali Muslim refugee women in Boston (Percac-Lima et al., 2013). Culturally or religiously congruent interventions used bilingual or bicultural health educators and were conducted in mosques and other socio-cultural organizations. Discussion Our results underline that knowledge, language, gender, traditional beliefs, and health care providers' characteristics can act as barriers and facilitators to Arab Muslim immigrant and refugee women's breast cancer screening practices. Lack of knowledge on breast cancer, minimal formal education, fatalism, and stigma are significant individual barriers decreasing breast cancer screening. Lack of time due to resettlement also negatively impacts breast cancer screening. Lack of fluency in English and a poor understanding of the receiving's country health care systems affect access to care and programs. Health care providers' lack of cultural competency negatively affects Arab Muslim immigrants' and refugees' attitudes and behaviors toward breast cancer screening. A geographically and financially accessible health care system, the presence of culturally and religiously tailored educational strategies, and health care professionals' support are likely to influence Arab Muslim immigrant and refugee women's breast cancer screening practices positively. Our results indicate the need to develop and implement culturally competent health promotion interventions. As reported in recent studies, education programs must target non-Muslim health care providers to decrease Arab Muslim women's mistrust and perceptions of ethnocentrism and racism (Jaffee et al., 2021;Racine et al., 2021b). Education programs positively impacted Syrian refugee women's breast awareness (Erenoglu & Sozbir, 2020) and intentions to screen (Pratt et al., 2020). Religiously tailored mosque-based interventions targeting beliefs impinging mammography uptake improved mammography rates (Padela et al., 2018(Padela et al., , 2019. Culturally tailored and language-concordant navigator programs increased mammography screening rates among Muslim refugee women (Percac-Lima et al., 2012). Utilizing a health educator of the same ethnic group reduces cultural, linguistic, and health care barriers (Milenkov et al., 2020). Implications for Nursing Practice and Research Our review highlights the need to specifically address gender, language, and cultural and religious beliefs (Saadi et al., 2012(Saadi et al., , 2015. Breast cancer awareness and education must occur early in the resettlement process to develop or maintain health promotion behaviors. Arab Muslim immigrant and refugee women present knowledge and language barriers and time constraints related to post-resettlement activities, collaborations with religious or community leaders, and settlement agencies represent the cornerstone of the breast cancer prevention programs. This review indicates the urgent need to tailor preventive interventions to improve breast cancer screening among Arab Muslim immigrant or refugee women. Interventions should enhance women's knowledge, skills, and confidence, toward breast screening while addressing religious and cultural factors such as fears, fatalism, stigma, and taboos. Nurses should not hesitate to rely on interpreters and be aware of the resettlement issues affecting Arab Muslim women's access to breast cancer screening activities. Nurses must be knowledgeable about time constraints and understand that breast cancer early screening may not be a post-resettlement priority. Future research should design and test culturally sensitive interventions for this population of women who may be vulnerable due to preand post-resettlement conditions. Limitations Though we conducted an extensive literature search, we only retrieved articles in English. Most studies used self-reports to assess women's breast cancer screening practices introducing potential biases. Several studies collapsed Arab Muslim immigrants and refugees into one category, obfuscating the specific needs of refugee women. The strength resides in capturing empirical studies beyond North America to include Turkey, Lebanon, and Jordan. Conclusion Addressing Arab Muslim refugee women's lack of knowledge on breast care and increasing early exposure to breast cancer screening is recommended in the early stage of resettlement. Culturally adapted and religious-based interventions delivered in Arabic represent promising avenues to increasing breast cancer screening and avoiding late-stage diagnosis. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by funding provided by Sigma Theta Tau International and the Canadian Nurses' Foundation to the first author. The second author was funded by The Scientific and Technological Research Council of Turkey (TUBITAK). Supplemental Material Supplemental material for this article is available online.	2022-04-28T06:23:24.364Z	2022-04-26T00:00:00.000	{ "year": 2022, "sha1": "67d3bc4255802c424c36774ddac24725d1d95700", "oa_license": "CCBYNC", "oa_url": "https://journals.sagepub.com/doi/pdf/10.1177/10436596221085301", "oa_status": "HYBRID", "pdf_src": "Sage", "pdf_hash": "54dfdbc573b0ba777186af01a8f29419bdf0b8fe", "s2fieldsofstudy": [ "Medicine", "Political Science" ], "extfieldsofstudy": [ "Medicine" ] }
269715505	pes2o/s2orc	v3-fos-license	Quantitative characterization of microstructure and research on spatial variation characteristics of loess of different strata in Luochuan, Shaanxi, China The complete sequence of loess strata in Luochuan has become a typical section in loess strata, and is the main focus of research for many scholars studying loess. We were based on the theory of aeolian loess and established a set of quantitative index parameters for loess microstructure through our previous research, such as equivalent diameter, sphericity, morphology ratio, orientation angle Phi, orientation angle Theta, pore Eq-Radius, throat Eq-Radius and throat channelLength. Through the quantitative characterization of various index parameters of the Luochuan loess, we found that the probability density of each index parameter meets a specific distribution well, and in terms of spatial dimension, it shows that as the depth of the strata increases, the average particle size and the mode of pore Eq-Radius, throat Eq-Radius and throat channelLength generally increase, while the mode of particle morphology ratio generally decreases. In addition, loess particles in deeper strata are less prone to vertical sedimentation and tend to deposit gently or horizontally. Most particles in different strata are distributed in a northwest or southwest direction. During the formation period of strata, the main cause for spatial differences is the material carrying force. We conducted a statistical analysis on the correlation between the macroscopic physical properties of loess and its microstructure index parameters. Specifically, we found a positive correlation between loess density and the average particle size and the mode of particle equivalent diameter, Additionally, we found a negative correlation between loess liquid limit and plastic limit, and the mode of particle morphology ratio. Furthermore, there was a negative correlation between permeability coefficient and the mode of pore Eq-Radius, throat Eq-Radius, and throat channelLength. Introduction From the 19th century to the early 20th century, scholars from various countries conducted in-depth research on the causes of loess and proposed many different theories.Lyell believed that the causes of loess were alluvial, floodplain, lake sedimentation, and marine sedimentation.Richthofen proposed the eolian loess theory, and later, other researchers suggested the theory of pluvial and slope accumulation.The controversy surrounding the genesis of loess persists to this day.Among them, the eolian loess theory is the most recognized viewpoint among many scholars in various countries at present. Based on the aeolian Loess theory, many scholars have conducted a series of studies on loess.However, in the early studies, their focus was primarily on the macro physical properties of loess.With the innovation of technology and the deepening of research, scholars have gradually noticed that the microstructure characteristics of loess are the most fundamental factors controlling the structural properties of loess.These characteristics directly influence the macro physical properties and engineering properties of loess.The relevant research findings on the microstructure of loess are crucial for addressing loess engineering challenges.Chinese scholars Haizhi Zhu [1,2], Zonghu, Zhang [3], and others have pioneered the research of loess microstructure in China.Foreign scholars have also made significant contributions to loess microstructure [4][5][6][7] during the same period. With the continuous development of observation techniques and image processing technology, such as the application of optical microscope, scanning electron microscope (SEM), mercury intrusion method, nuclear magnetic resonance (NMR), and computed tomography (CT), scholars have conducted more specific and detailed research on the microstructure characteristics of loess including particle morphology, pore distribution, particle contact, connection, and arrangement, and have achieved remarkable results [8][9][10][11][12]. In recent years, some domestic scholars have dedicated themselves to exploring the relationship between the microstructure of loess and its macroscopic physical and mechanical properties.Then have conducted research the combines microscopic experiments with macroscopic physical and mechanical experiments, leading to fruitful results.These results could provide a reference basis for multi-scale analysis of loess disasters [13][14][15][16][17][18][19][20][21][22][23][24].In his research on the permeability properties of unsaturated loess, Wang Haiman assessed the permeability of densely packed loess and examined the microstructure changes in loess before and after rainfall using SEM and NMR methods.Additionally, a model was developed to estimate the permeability of unsaturated densely packed loess [25].Modified loess is of great significance in enhancing the engineering properties of loess.The improved loess has demonstrated enhancements in both microstructure and macroscopic strength, making it a research hotspot in recent years [26][27][28].Despite these achievements, there are still controversies about the causes and mechanisms of loess geological disasters.At the same time, research results mainly focus on the two-dimensional microstructural characteristics of loess, and few researchers have conducted studies on the three-dimensional microstructural characteristics and quantification of loess. The three-dimensional microstructure of loess can be obtained through high-resolution scanning of an optical microscope combined with continuous sectioning and three-dimensional image reconstruction technology.NMR is more suitable for detecting pore water in loess.While scanning electron microscopy and CT scanning can provide high-resolution images, they are costly and difficult to access for most researchers.Therefore, considering the economic and technical feasibility, this paper utilizes the optical microscope, which has a higher level of technological maturity, as the primary tool for this study. The Luochuan loess profile is situated on the Luochuan loess plateau in the central region of the Loess Plateau, with a thickness ranging from 138 to 150 meters.Due to its complete stratigraphic sequence and good exposure, it has become a typical profile in the loess strata and has been favored by many researchers in loess.Therefore, there are many research results on the Luochuan loess [29][30][31][32][33].However, most of these studies focus on the macroscopic physical properties and engineering characteristics of loess, with limited research on its microstructure.Therefore, we selected eight typical strata of the Luochuan loess, including L1, S1, L2, S2, L6, L7, L8, and L9 for sampling.Through field and laboratory experiments, we obtained the physical and mechanical parameters of the loess.We conducted research on the threedimensional microstructural characteristics of the loess in the above strata based using optical microscopy and continuous slicing techniques. Experiment of physical property In the process of sampling loess from different strata in Luochuan, the destruction of the sections increased the sampling difficulty and risk.To ensure the representativeness of the samples, loess samples were collected from the L1, S1, L2, S2, L6, L7, L8, and L9 strata.Indoor physical and mechanical tests were conducted on loess samples from various strata to determine their physical and mechanical properties.The density, liquid limit, plastic limit, and shear strength of loess samples in each stratum are presented in Table 1.The cohesion (C) and internal friction Angle (φ) were determined through direct shear experiments, while the collapsibility coefficient was obtained through double-line experiments. Research method of three-dimensional microstructure of loess The continuous slicing method based on an optical microscope is used to study the threedimensional microstructure of loess.The experiment and image reconstruction process (Fig 1) mainly include the following steps.(1) Sample preparation.We cut the loess sample into a cylinder with a diameter of 1 cm and a height of 1.5 cm.Subsequently, we saturated it with a soaking solution in a vacuum environment [34], The soaking solution consists of epoxy resin, acetone, ethylenediamine, and dibutyl phthalate in a volume ratio of 100:50:2:1.After a period of time, the sample completely hardened, and we prepared the hardened sample for modeling. (2) Image acquisition.We utilized a Multiprep device to grind and polish the sample.During this process, we employed a vertical displacement laser monitor and adjusted the polishing duration to control the sample thickness, ensuring a reduction of approximately 2 microns after each polishing cycle.We used the Leica DM6000M with a planar image resolution of 4micron per pixel to observe and capture photos of the polished sample, in order to obtain a two-dimensional image of the sample.Repeat the aforementioned grinding, polishing, observation, and photography process to ultimately acquire a series of 2-micron interval twodimensional images.(3) Three-dimensional (3D) structure reconstruction.In AVZIO software, we begin by manually aligning images using the Align feature and then utilize the Slices command to automatically align slices.After that, we use the Thresholding tool to determine the appropriate threshold for segmenting particles and pores by adjusting the grayscale of the image.After completing threshold segmentation, we extract two-dimensional microstructures such as particles and pores, and then proceed to creating the morphology of three-dimensional particles and pores.(4) Quantitative analysis [35].Based on the results of the experiment and image reconstruction, the 3D microstructure of loess particles and the three-dimensional pore network were constructed.Various important parameters related to volume, morphology, orientation, etc., were utilized to quantitatively depict the characteristics of loess particles.The volume characteristics could be presented using the equivalent diameter (Eq-D), which can be defined as Eq-D ¼ ffi ffi ffi ffi ffi ffi , where the parameter V represents the volume of a loess particle.The morphological characteristics could be depicted using sphericity and morphology ratio.The sphericity is defined as the ratio of the surface area of spheres with the same volume as loess particles to the surface area of loess particles.The particle morphology ratio is defined as the ratio of the major axis (L) to the minor axis (W) of a particle.The orientation characteristics could be depicted using the parameters Phi and Theta, which represent the dip angle and strike angle of a particle, respectively.The definition of the characterization parameters for loess particle is presented in Fig 2 .The pore network is divided into a large number of pores by throats.The pore Eq-Radius is defined as the radius of a sphere which is equal to the pore volume, Throat is defined as a narrow channel connecting adjacent pores.The throat channelLength is defined as the distance between the centers of adjacent pores.Definition of characteristic parameters for loess pore is presented in Fig 3. Particle characteristics Particle equivalent diameter.There are differences in the equivalent diameter of loess in different strata.However, the probability density of the equivalent diameter of loess meets a Rational distribution, Figs 4 and 5 show the experimental results, fitting curve, and cumulative probability curve of the particle equivalent diameter probability density distribution in the L1 stratum and various strata respectively.The probability density function is as follows: Where x is the particle equivalent diameter, and a, b, c, d, e are the fitting parameters.According to the fitting situation, and the R value of goodness of fit can exceed 0.95.Through the analysis and statistics of the modes of the Eq-D and average particle sizes of loess in different strata, it is evident that there is a significant difference in particle size with the changes in loess strata depth.The modes of Eq-D and the average particle sizes of loess in L1, S1, L2, and S2 strata are significantly smaller than those in L6, L7, L8, and L9 strata.This suggests that the particle size of deeper loess strata is generally larger than that of shallow loess strata. Particle morphology.According to the distribution characteristics of particle sphericity probability density, the sphericity experiment data of loess particles in L1-L9 strata of Luochuan can be well fitted using the Beta distribution (Eq 2).Figs 6 and 7 show the distribution of particle sphericity probability density and fitting curve in the L1 stratum and various strata.The probability density fitting function is as follows: Where x is the particle sphericity, and α and β are fitting parameters.According to the fitting situation of the experimental results, the R value of goodness of fit can exceed 0.95.Comparing the fitting distribution curves in Fig 7, it can be concluded that the overall particle sphericity shows the characteristics of gradual decrease with the increase of strata depth. The particle sphericity of various strata is distributed between 0.2 and 1.0, but mainly concentrated between 0.6 and 0.85.Except for the L6 and L7 strata, the quantity percentage of particles with a particle sphericity greater than 0.7 in loess particles in other strata exceeds 50% of the total.According to the classification of particle sphericity characteristics [35], the loess particles could be mainly classified as thin-slice, long-strip, sub-prismatic granule, multiangled granule, sub-globularity, and globularity, respectively.L6 and L7 strata are mainly composed of multi-angled granule and sub-globularity particles, while remaining strata are primarily composed of sub-globularity particles.Through the analysis and statistics of the modes of particle sphericity in various strata, it was observed that the particle sphericity of L1, S1, L2, and S2 strata gradually increases with the increase of loess strata depth, while the particle sphericity of L6, L7, L8, and L9 strata gradually decrease with the increase of loess strata depth. Figs 8 and 9 show the distribution of particle morphology ratio probability density and fitting curve in the L1 stratum and various strata.Its distribution meets the Gamma distribution (Eq 3), and the R value of goodness of fit can exceed 0.90.The probability density function is as follows: Where x is the particle morphology ratio, and a and b are the fitting parameters.It can be seen that the mode of particle morphology ratio generally shows a de-creasing trend with the increase of loess strata depth, as depicted in Fig 9. Particle orientation.Fig 10 shows the percentage distribution curve of particle Phi angle in L1-L9 strata of Luochuan.The changes in particle Phi angles in L1-L9 strata show similar in-crease and decrease characteristics.When comparing the quantity percentage of Phi angles ranging from 0 to 50 degrees, it is evident that the deeper strata consistently exhibit a higher percentage than the shallow strata.This suggests that particles in the deeper layers of loess are less likely to be deposited vertically and tend to settle more gently or horizontally. Pore and throat characteristics According to the experimental results, the pores were extracted from the image using AVZIO software, and the three-dimensional pore model (Fig 3) was established.The quantitative analysis and description of loess pores in different strata in Luochuan are achieved using three quantitative index parameters: pore Eq-Radius, throat Eq-Radius and throat channelLength. Figs 12 and 13 show the probability density distribution of the pore Eq-Radius in the L1 stratum and various strata.From the distribution characteristics, it can be seen that the fitting function of the probability density distribution of the pore Eq-Radius also meets the gamma distribution, and the R value of goodness of fitting can exceed 0.90. According to the pore size classification standard proposed by Lei Xiangyi [36], the quantity percentage of micropores and small pores in strata L1-L9 is extremely low.The quantity percentage of meso-pores are 52.1%,49.9%, 35.0%, 40.0%, 32.5%, 32.4%, 40.0%, and 32.5%, respectively.The quantity percentage of macro-pores are 47.5%, 50.0%, 65.0%, 60.0%, 67.5%, 60.0%, and 67.5%, respectively.It shows that there are almost no micro-pores and small pores in each stratum of loess, which are mainly meso-pores and macro-pores, and most of them are macro-pores; On the whole, the quantity percentage of meso-pores decreases with the increase of loess strata depth, on the contrary, the quantity percentage of macro-pores increases with the increase of loess strata depth.It can be concluded that the mode of pore Eq-Radius generally increases with the increase of loess strata depth through the analysis and statistics of the mode of pore Eq-Radius in each stratum. Throat size is the main factor determining the permeability and pore connectivity of loess.Based on the results of microstructure experiments and data analysis, the probability density distribution of the throat Eq-Radius was obtained.From the distribution characteristics, it can be seen that the probability density distribution fitting function of the throat Eq-Radius meets the Gaussian distribution (Eq 4).The probability density function is as follows: Where x is the throat Eq-Radius, and a, b, and c are the fitting parameters.Figs 14 and 15 show the probability density distribution of the throat Eq-Radius in L1 stratum and various strata, and R value of goodness of fitting can be close to 1. The throat Eq-Radius in each stratum is primarily distributed between 0 micron and 35microns, and the mode of throat Eq-Radius is mainly distributed between 7 microns and 10microns.The number of throats with a radius between 0 micron and 10microns in each stratum can reach approximately 50%.According to statistical analysis, the modes of throat Eq-Radius in L1-L9 strata are 7.8microns, 7.5microns, 9.5microns, 8.7microns, 9.0microns, 9.1microns, 9.1 microns, and 9.0 microns, respectively.The mode of throat Eq-Radius shows a gradually increasing characteristic with the increase of loess strata depth. Based on the experimental results, the probability density distribution characteristics of the throat channelLength in L1-L9 strata of Luochuan loess were obtained.Figs 16 and 17 show the probability density distribution and fitting curves of the throat channelLength, respectively, in the L1 stratum and various strata.It can be seen that the probability density distribution fitting function of the throat channelLength meets the Gamma distribution well. According to statistical analysis, the modes of throat channelLength in L1-L9 strata are 41microns, 40.5microns, 44.5microns, 43.5microns, 44.5microns, 45.5microns, 41.0microns, and 46.5 microns, respectively.It shows the mode of the throat channelLength in L1-L9 strata of Luochuan loess has an gradually increasing characteristic with the increase of loess strata depth. By comparing the distribution of pore Eq-Radius and throat Eq-Radius in different strata, it can be observed that the sizes of pores and throats in deeper loess strata are larger.It is inferred that the reason for this phenomenon is related to the particle sizes of loess in different strata.The loess in deeper strata has larger particles, smaller sphericity, and is prone to the formation of macropores, mostly in the form of particle aerial structures [36].The loess in shallower strata has smaller particles, larger sphericity, and more surface-to-surface contact between particles, making it prone to the formation of mesopores and small pore. Analysis of differences of loess strata in Luochuan During the lengthy deposition process of loess, the loess particle sizes in different strata exhibit certain variations.Based on the micro-scale study of loess particles in the above strata, it can be observed that the loess particles tend to become larger with the increase of loess strata depth as shown in Fig 5 .The change of loess particle sizes in different strata is not only associated with the particle size of the original loess parent material and the material handling force, but also with the change of climate conditions and weathering during the formation of strata to some extent.The primary material source of the Loess Plateau comes from the desert of Inner Mongolia next to Mongolia in the northwest.Based on this, it can be inferred that the material source of each stratum is essentially the same.Therefore, the material carrying force has become the main reason for the difference of loess particle size within the aforementioned strata.The particle sizes of L6, L7, L8, and L9 strata are large, suggesting that there was strong carrying wind and dry and cold climate during that time, but L8 and L9 strata underwent significant physical weathering, resulting in slightly smaller particle sizes compared to L6 and L7 strata.The particle sizes of L1, S1, L2, and S2 strata are small, suggesting that the carrying wind force was relatively weakened during that time and suffered different degrees of weathering during carrying process.It can be inferred that most of the loess materials forming L6, L7, L8, and L9 strata, such as silts, were suspended in the air during the carrying of strong wind force, resulting in relatively few instances of abrasion between them.At the same time, medium sand, fine sand, and other materials were swiftly transported to the vicinity of loess accumulation remotely by leaps and bounds under the carrying of strong wind force.The loess materials forming L1, S1, L2, and S2 strata, such as silts, were suspended close to the surface in the air and carried by the carrying of weak wind force, causing constant abrasion of the materials.At the same time, some medium sands and fine sands carried from a long distance and near the loess accumulation were carried twice or repeatedly under the carrying of weak wind force, and finally loess particles with high roundness and poor sorting were formed. We have conducted preliminary explorations and inferences on the genesis of the Luochuan loess based on microstructure quantitative index parameters.We have clarified that the material sources of the loess of various strata in Luochuan were basically the same during the formation period.The material carrying force was the main cause of the differences in the properties of the various strata.This research result is essentially consistent with the overall environment of the formation of the Loess Plateau.It is also consistent with the current mainstream research results [37][38][39][40].However, we need to conduct more research on the Luochuan loess from multiple perspectives to confirm our hypothesis.In the next research, we will try to combine stratum magnetic susceptibility with microscopic structural parameters to conduct further research, revealing the paleoclimatic environment during the formation period of the loess of various strata in Luochuan.In the study, we found that the quantity percentage of micro-pores and small pores in the loess of various strata in Luochuan is extremely low.This feature may be influenced by the experimental method used, and a large number of pores less than 2 microns are ignored due to continuous slicing. The correlation between physical properties and microstucture index parameters of loess A highlight of our research results is that we revealed the intrinsic relationship between loess microstructure and physical properties through quantitative characterization of microstructural indicators.This approach differs from previous methods that relied loess microstructure to investigate physical properties [41][42][43][44][45], offering new research ideas for subsequent scholars.In our research results, the density of loess is significantly correlated with the average particle size and the mode of particle equivalent diameter overall.Additionally, the liquid limit and plastic limit are significantly correlated with the mode of morphology ratio overall.The permeability coefficient is significantly correlated with the mode of pore Eq-Radius and throat Eq-Radius on the whole.However, its shear strength parameters and collapsibility coefficient do not demonstrate a correlation with the microstructural index parameters.This lack of correlation may be influenced by the limited sample data available. The physical properties of loess are closely associated with the characteristics of loess particles, pores, and throats.Among these factors, the density, average particle diameter, and the mode of particle equivalent diameter of L1-L9 strata increase overall with the increase of strata depth as depicted in Fig 18, indicating that the density of loess shows a positive correlation with the average particle size and the mode of particle equivalent diameter.The increase in density of Luochuan loess is related to the increase in particle size to some extent. Fig 19 shows the correlation between the plastic limit, liquid limit and the mode of particle sphericity and particle morphology ratio.The plastic limit and liquid limit exhibit a significant negative correlation with the mode of morphology ratio, while the plastic limit and liquid limit show a weak negative correlation with the mode of particle sphericity.With the increase of loess strata depth, the density and average particle size of loess also increase, and the particle morphology ratio of loess overall decreases and the quantity percentage of thin-slice and longstrip particles decreases, and the quantity percentage of sub-prismatic granule, multi-angled granule, and sub-globularity particles relatively increases.This leads to an increase in the specific surface area of loess, subsequently increasing the liquid limit and plastic limit.Fig 20 illustrates the correlation between permeability coefficient and the mode of pore Eq-Radius and throat Eq-Radius.The mode of pore Eq-Radius and throat Eq-Radius increase with the increase of strata depth, while the permeability coefficient decreases significantly with the increase of strata depth, indicating a negative correlation between the permeability coefficient and their mode.Fig 21 illustrates the correlation between the permeability coefficient and the mode of throat channelLength, and it shows a negative correlation between the permeability coefficient and the mode of throat channelLength.Based on this, it can be inferred that the permeability behavior of Luochuan loess primarily depends on the quantity variations of pores and throats unit volume, and there is little correlation with the size of pores and throats. Conclusions The probability density of the parameters of loess particles and pores microstructure meet a specific functional distribution well. The overall characteristics of the loess in Luochuan show that with the increase of loess strata depth, the size of particle, pore and throat, and throat channelLength gradually increase, and the particle morphology ratio gradually decreases.L1, S1, L2, and S2 strata were formed during the weak wind period in the ancient climate, while the L6, L7, L8, and L9 strata were formed during the strong wind period in the ancient climate.The variation trends of particle Phi angle and Theta angle are basically the same, and loess particles in deeper strata are not easy to deposit vertically and more tend to deposit gently or horizontally.Most particles in different strata are distributed in a northwest or southwest direction.The pores distribution of loess in various strata is mainly dominated by mesopores and macropores, but in deeper strata, macropores account for a larger proportion and are mostly single particle stacked pore structure. The physical properties of loess are closely related to its microstructure parameters.The density of loess has a positive correlation with the average particle size and the mode of particle equivalent diameter.The plastic limit and liquid limit have a negative correlation with the mode of morphology ratio.The permeability coefficient has a negative correlation with the mode of pore Eq-Radius, throat Eq-Radius and throat channelLength, inferring that the permeability behavior of Luochuan loess primarily depends on the quantity rather than size variations within unit volume for both pores and throats.	2024-05-12T05:05:40.939Z	2024-05-10T00:00:00.000	{ "year": 2024, "sha1": "d1a31a6efbd5425560d41a96b3c4c969e586df12", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "d1a31a6efbd5425560d41a96b3c4c969e586df12", "s2fieldsofstudy": [ "Geology" ], "extfieldsofstudy": [ "Medicine" ] }
269764325	pes2o/s2orc	v3-fos-license	Dynamic soil properties for soil health (DSP4SH) database 1.0 – Phase 1 and 2 datasets The dynamic soil properties for soil health (DSP4SH) is a Science of Soil Health Initiative that was designed to collect, process, and publicize scientifically rigorous datasets that inform sound indicators and interpretations. The Soil and Plant Science Division of the United States Department of Agriculture - Natural Resources Conservation Service (USDA-NRCS) and university cooperators collected a suite of standardized soil health metrics across eight states (Oregon, Washington, Kansas, Minnesota, Illinois, Connecticut, North Carolina, and Texas) within five soil survey regions (Northwest, North Central, Northeast, Southeast, and South Central). The DSP4SH database provides a substantial dataset of soil health metrics assessed. The dataset is composed of dynamic soil properties (DSP) data collected from each management system or ecological state represented by one to three independent plot replicates. Each plot has a minimum of three pedons. Nine groups from the DSP4SH monitoring network provided datasets used in developing the database. The submitted data includes 37 laboratory measured parameters, 60 variables of layer/horizon descriptions, 41 variables for laboratory analysis conducted at the Kellogg Soil Survey laboratory, and 12 variables for the management systems. An additional 31 variables were developed for site or plot description. Additional variables were developed to normalize the dataset. In preparation for DSP assessment, all tables (except for dataset from KSSL lab) were categorized by management system or ecological state. The categories were business as usual (BAU), the reference condition (Ref) and the soil health management (SHM). The overarching goal of DSP4SH phase 1 and 2 dataset publication is to promote increased accessibility, further analysis of the data, and overall understanding of the benefits of surveying dynamic soil properties for soil health. a b s t r a c t The dynamic soil properties for soil health (DSP4SH) is a Science of Soil Health Initiative that was designed to collect, process, and publicize scientifically rigorous datasets that inform sound indicators and interpretations.The Soil and Plant Science Division of the United States Department of Agriculture -Natural Resources Conservation Service (USDA-NRCS) and university cooperators collected a suite of standardized soil health metrics across eight states (Oregon, Washington, Kansas, Minnesota, Illinois, Connecticut, North Carolina, and Texas) within five soil survey regions (Northwest, North Central, Northeast, Southeast, and South Central).The DSP4SH database provides a substantial dataset of soil health metrics assessed.The dataset is composed of dynamic soil properties (DSP) data collected from each management system or ecological state represented by one to three independent plot replicates.Each plot has a minimum of three pedons.Nine groups from the DSP4SH monitoring network provided datasets used in developing the database.The submitted data includes 37 laboratory measured parameters, 60 variables of layer/horizon descriptions, 41 variables for laboratory analysis conducted at the Kellogg Soil Survey laboratory, and 12 variables for the management systems.An additional 31 variables were developed for site or plot description.Additional variables were developed to normalize the dataset.In preparation for DSP assessment, all tables (except for dataset from KSSL lab) were categorized by management system or ecological state.The categories were business as usual (BAU), the reference condition (Ref) and the soil health management (SHM).The overarching goal of DSP4SH phase 1 and 2 dataset publication is to promote increased accessibility, further analysis of the data, and overall understanding of the benefits of surveying dynamic soil properties for soil health. © Value of the Data • The data for dynamic soil properties (aggregate stability, soil organic carbon, permanganate oxidizable carbon, autoclaved-citrate extractable protein, and soil respiration) can be used as a reference and baseline values for soil health indicators that can be linked to soil inherent properties and climate to enhance soil information.• The data can also be used to assess the relationship and effectiveness of land use and management systems or conservation agriculture on overall soil health metrics.The data provides reference values that could be beneficial for state and transition models and other forms of related data modeling.• These data provide a collection of dynamic soil properties and soil health indicators that can be used to standardize and update soil survey information.• This dataset can be used to understand the influence of dynamic soil properties on soil health.These data would be beneficial to answering questions regarding the selection of land management and conservation practices that may be posed by researchers, soil scientist, ecologist, rangeland, and soil health specialists. Background Soil health is a foundational concept that describes the sustainability and productivity of a given soil.It is accepted that overall soil health is dependent on physical, biological, and chemical properties that are often dynamic in nature.Dynamic Soil Properties (DSPs) refer to the characteristics of soil that change on human timescale in response to both natural occurrences and human-induced activities, that includes agricultural and wildland management.These properties serve as markers for understanding how soil functions and undergoes transformations over time.The Science of Soil Health Initiative is a United States Department of Agriculture -Natural Resources Conservation Service (USDA-NRCS) effort to better understand soil health by assessing the variability of soil health indicators across the multiple soil types and land management practices.The USDA-NRCS has identified the following potential indicators that may provide key information to aid in the selection of best practices for conservation: bulk density, water stable aggregates, soil organic carbon (SOC), total nitrogen, texture, gravimetric moisture content, effervescence, electrical conductivity, pH (1-to-1, soil to water ratio), soil respiration, soil enzymes activities ( β-glucosidase, N-acetyl-β-d-glucosaminidase, phosphomonoesterases (acid/alkaline phosphatase), and arylsulfatase), permanganate oxidizable carbon (POXC), and autoclaved-citrate extractable protein content.The datasets from the measurement of these indicators were collected in phases as these projects were completed.The phase 1 and 2 datasets were used for the development of first version of the database referred to as 'dynamic soil properties for soil health database version 1.0'. Data Description The provided database includes ten datasets and four metadata tables [ 1 ].There are five levels ( Fig. 1 ) of the database.The project level allows for information and details on each project and management systems.The project level (with parent table named as sourceproject ) is linked to the design level (with child table named as projectdesign ) which provides details about the description of the project and linked to the site level (with table named as plotoverview ).The site level has plot and site information that is linked to the pedon level.The pedon (referring to a 3-dimensional, representative sample of the soil, encompassing all its horizons and unique characteristics) level has the management system table (named as dspplotmgt ) and pedon table (named as pedon ) while the site table is linked to the weather table (named as weather ).The pedon level is linked to the description level which is made up of pedon description (with table named as layerdescription ) and laboratory analyses (with table named as cooplabmst ) tables.An intermediate table designed to serve between the description and laboratory analyses is referred to as the layerdesignation table.The horizon is a section within the soil profile that has undergone distinct changes in its physical, chemical, or biological properties due to soil-forming processes, resulting in a recognizable and unique composition compared to other sections.The layer is defined as a fixed depth increment of the pedon. The laboratory analyses tables provide details about the analyses conducted at the cooperator's laboratory and the Kellogg Soil Survey Laboratory -KSSL (with table named as kssllabmst ). Experimental Design, Materials and Methods The acquisition of data began with the identification of sites based on various management differences (like tillage intensity, cover crops application and crop residue integration) and ecological states (like reference or cropland) that are relevant to soil health.The collection of information on ecological state and management history for each project ( Fig. 2 ) was then carried out.This was followed by the identification and collection of the nearest climate station information for each site and plot information was also collected for each site.After locating a plot representing the management system, a characterization pedon and two satellite pedons were described per plot.For each pedon, each layer or pedogenetic horizon was described according to standard soil survey nomenclature using the Field Book for Describing and Sampling Soils [ 2 ].Field description and observations included horizon nomenclature, field texture, soil color, redoximorphic features, concentrations, rock fragment volume, consistence, structure, roots, and pores information [ 2 ].Field measurements included infiltration rates using single ring Fig. 1.Generalized design of data collection.The flowchart above shows the five levels of the database, with data collected at the project level (dark green rectangle) linked to the dataset project design (dark orange rectangle) and linked to the dataset collected at the plot and site level (depicted as darkish purple color).The plot rectangle is linked to the management system and pedon level dataset (dark magenta color) while the site level dataset is linked to the weather dataset (dark magenta color).The pedon is linked to datasets of layer level laboratory analyses and descriptions (green rectangle) at the layer and horizon levels.The inset picture is a typical 2D representation of a pedon showing horizon and layers over 120 cm depth (from top to bottom).At least five fractions or peds were collected from each layer or horizon and consolidated as a single sample.Analyses were carried out for the layers (0-5 cm and 5-10 cm) and genetic horizons of all classification pedon.The number of pedons that were described and analyzed is 292.The number of horizons and layers analyzed is 1685.infiltrometer according to Ogden et al. [ 3 , 4 ].Samples were collected from each layer at a fixed depth (0-5 cm and 5-10 cm) and thereafter by genetic horizon to 100 cm or bedrock.The collected samples were analyzed in the laboratory for biophysicochemical properties including bulk density, water stable aggregates, soil organic carbon (SOC), total nitrogen, texture, gravimetric moisture content, effervescence, electrical conductivity, pH (1-to-1, soil-to-water ratio), soil respiration, soil enzymes activities ( β-glucosidase, N-acetyl-β-d-glucosaminidase, phosphomonoesterases (acid/alkaline phosphatase), and arylsulfatase), permanganate oxidizable carbon (POXC), and autoclaved-citrate extractable protein content [ 2 ]. Details of all laboratory methods used can be found in Standardized Methods for Selected Laboratory Procedures to Assess Soil Health.In brief, SOC was determined using the dry combustion method according to Nelson and Sommers (1996, Laboratory Method 1) [ 5 ].Air dried soil sample was sieved to fine-earth fraction ( < 2 mm) and homogenized.The homogenized sample undergo further processing from < 2 mm to < 180 μm fraction using planetary ball mill and 80 mesh ( ≈180 μm).An aliquot of < 180 μm sample was packed into tin foil that was weighed and analyzed for total carbon, nitrogen, and sulfur using an elemental analyzer.The soil sample was treated with hydrochloric acid and the evolved carbon dioxide (CO 2 ) was measured manometrically as percentage of calcium carbonate (CaCO 3 ).The SOC was calculated as the difference between total carbon and inorganic carbon.The mean weight diameter (MWD), an index used to relate aggregate stability to water stable aggregates [ 6 ] was determined using the Yoder wet sieving method according to Mikha and Rice (2004, Laboratory Method 2) [ 7 ].For Yoder wet sieving method, the undisturbed soil clod subsample was sieved to < 8 mm and air-dried.The air-dried sample was then placed in the top sieve of the nested sieves, prewetted for 10 mins before agitating for 5mins at 30 osc/min to disaggregate the soil through slaking.The mean weight diameter was then calculated as the sum of the fractional-mean weight diameter retained on each sieve.For water stable aggregates determined by KSSL wet aggregate stability method (Laboratory Method 3) [ 8 ], the undisturbed soil clod subsample processed to a < 8 mm fraction and air-dried was gently crushed to pass through 2 mm sieve and collected on a 1 mm sieve.The retention of the 2-to 1-mm air-dried sample was then measured on a 0.5 mm sieve after sample was submerged in distilled water overnight and thereafter agitated.The soil respiration was measured by conducting a 4-day incubation of the sample, with CO 2 output determined by sealed chamber alkali trap respirometry (Laboratory Method 4) [ 9 ]. Soil samples were also quantified for five soil enzyme activities.The soil enzyme assays were quantified for β-glucosidase (BG) according to Eivazi and Tabatabai [ 10 ]; N-acetyl-βd-glucosaminidase (NAG) according to Eivazi and Tabatabai [ 10 ]; phosphomonoesterases (acid phosphatase (ACP) and alkaline phosphatase (ALP)) according to Eivazi and Tabatabai [ 11 ]; and arylsulfatase (AS) according to Tabatabai and Bremner [ 12 ].Briefly, 1 gram of soil sample was treated with modified universal buffer and p-nitrophenyl derivate substrate for each enzyme and then incubated for 1 hour at 37 °C.Calcium chloride (CaCl 2 ) and THAM (pH 12) or NaOH solution was then added after incubation.Colorimetric determination of p-nitrophenol in an aliquot of the filtered solution was carried out and reported as potential enzyme activities (Laboratory Method 5).POXC was determined by quantifying reduction in the violet color intensity of 0.02 M KMnO 4 (potassium permanganate) because of reaction with oxidizable C in soil according to Weil et al. (2003, Laboratory Method 6) [ 13 ].Autoclaved-citrate extractable protein content was determined by extracting protein from samples using a neutral sodium citrate buffer solution. After clarification of the extract, the protein content was quantified using a bicinchoninic acid protein assay according to Walker (Laboratory Method 7) [ 14 ].Soil texture was determined using the particle-size distribution analysis method (Gee and Or.2002, Laboratory Method 8) [ 15 ].The bulk density was determined using the field-state soil core method according to Grossman and Reinsch (2002, Laboratory Method 9) [ 16 ].A metal sampler of known volume was driven into the soil.The metal cylinder was then removed from the soil extracting a known volume of the sample.Excess soil was removed from the edge of the cylinder and moist sample weight was recorded.The sample was then oven-dried at 110 °C to a constant weight and the bulk density calculated.The water content was determined using the gravimetric moisture content method (Laboratory Method 10) [ 8 ].The effervescence was determined using the CaCO 3 effervescence method (Laboratory Method 12) [ 8 ].The electrical conductivity was measured with an electric conductivity meter (Laboratory Method 13).The pH was measured using a pH meter in a soil to water (1:1 w/v) solution (Laboratory Method 14) [ 8 ]. The pipeline for processing the collected DSP4SH data started with cooperator submission of data template (DSP4SHDT) and metadata to the DSP4SH project management team ( Fig. 3 ).The DSP4SHDT, a template used for data transfer, is a spreadsheet that has 10 tabs.The tabs in the DSP4SHDT are "Instructions" -used to provide general information on how to use the template, "TablesOverview" -serves as a data dictionary (with brief detail on table name, column name, method and explanation, unit of measurement, and data type) that is used to provide the explanation of tables and fields within each table, "Project_design" -provides the space to enter description of the project design, "Project-Plot_overview" -provides the space to enter data on project at the plot level, "Map_photo_schematic" -provides description and details about where to populate plot maps, "Weather" -the space to enter details about the climate station closest to each site, "Pedon_entry" -the spreadsheet where details on each pedon can be added, "Layer_description" -spreadsheet where all data collected for all layers/horizon sampled can be added, "Layer_lab_msmt" -spreadsheet where all laboratory data can be added, and "ChoiceLists" -is a metadata for all lists inputted within previous tabs. The submitted data template was then processed for format QA/QC and data QA/QC while iterations were performed in conjunction with the cooperators to resolve any data quality issues before stratification of the data.The database schema, entity relationship model and entity relationship diagram were then developed for the database.The data were imported from the DSP4SHDT, the database was created and uploaded with the data using custom-built R script.The SQLite database was then published as a data product that is available to the end users.This open access database is available to scientists, researchers, policy makers, and the public. The database can be accessed, and tables combined to analyze the soil health metrics and ecological or management systems provided in the database.The visualization of the distribution of the SOC versus land management faceted by the project provided in Fig. 4 shows one of many possibilities of harnessing what is available in the database.The SOC is available in the 'cooplabmst' entity which was then joined with the 'pedon' entity and then 'dspplotmgt' entity.The generation of these entities was used to generate a plot of SOC vs land management in Fig. 4 . The combination of query and data manipulation can allow for visual exploration of the data.For example, the variables like total nitrogen (available as 'TN_pct') and water aggregate stability (available as 'KSSL_WSA') from the cooperators' laboratory measurement entity, 'cooplabmst' and KSSL measurement entity 'kssllabmst' were queried from the database.The two variables were then filtered and joined to create a plot of total nitrogen vs water stable aggregates as shown in Fig. 5 . The number of observations within the 'cooplabmst' was also explored and shown in Fig. 6 .This was achieved by a combination of the 'dspplotmgt' entity which was combined with the 'pedon' entity and then 'cooplabmst' entity.The OregonState project had 294 observations, Texas A&M project had 246 observations, UnivOfTexasRGV project had 198 observations, Washington-State project had 181 observations, KansasState project had 174 observations, UnivOfConn project had 167 observations, UnivOfMinnesota project had 164 observations, UnivOfIllinois project had 160 observations, and NCState project had 101 observations.This gave rise to a total of 1685 observations across all cooperators' lab measurements. Limitations There are limited data that combine assessment of dynamic soil properties and ecological state or management systems in the US.Despite the database accomplishing this distinct achievement, the restricted availability of these data across different regions of the US results in significant data gaps essential for developing soil health models under varying ecological conditions and management strategies.Subsequently, enhancing data collection effort s in underrepresented states across each region, focusing on a variety of soil types, will effectively bridge these gaps. As previously noted, the database amalgamates soil characterization and management data, that can serve as a reference value for future studies.However, this data is not designed for use as a long-term soil monitoring data.In the absence of additional data, this limitation hampers the understanding of temporal soil changes under similar environmental conditions. The combination of the data from this database with climate data for each location will enhance the modelling and analyses for ecological system functions. Fig. 2 . Fig. 2. Distribution of DSP4SH projects.The DSP4SH projects were distributed across 9 different locations within US.Project locations are indicated by filled circles; different fill colors indicate different projects. Fig. 3 . Fig.3.Generalized pipeline of processing DSP4SH data.The cooperators submitted the data template (DSP4SHDT) accompanied by the metadata to the DSP4SH data management team.This was followed by the template formatting for QA/QC, data QA/QC and then categorization of management systems.The database and schema were then designed.Then, data was entered, and database was made available to end users.The green color depicts the cooperators' activities, the blue color depicts the DSP4SH data management team activities, the brown color depicts the DSP4SH database development activities, and the red color signifies activities of the end users.The solid lines indicate progression in the direction of the arrows while the dotted lines indicate progressions with iteration. Fig. 4 . Fig. 4. Distribution of the SOC vs. land management faceted by projects.The figure shows comparison of SOC and land management system across project and within project.Also shown in the figure is the count (n) of each classification (BAU, Ref or SHM) and measurement of SOC.The violin plot shows the spread of the SOC observations while the boxplot inset shows the central tendency and interquartile range of the SOC measurement by mass.Each observation of SOC represents SOC measurement for each layer by depth (cm). Fig. 5 . Fig. 5. Plot of total nitrogen (percent) vs water stable aggregates (percent) by laboratory measurements with marginal density plot for each variable.	2024-05-15T15:04:26.682Z	2024-05-01T00:00:00.000	{ "year": 2024, "sha1": "efc9c9197a71e1e21860c4507f28fce950079fd7", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "e4bbed2f19b9033491532450994f8628766486b0", "s2fieldsofstudy": [ "Environmental Science" ], "extfieldsofstudy": [ "Medicine" ] }
244480009	pes2o/s2orc	v3-fos-license	Therapists’ emotional state after sessions in which suicidality is addressed: need for improved management of suicidal tendencies in patients with borderline personality pathology Introduction Patient suicidality is a frequently experienced topic for psychotherapists. Especially adolescents with borderline personality pathology (BPP) often exhibit suicidal tendencies. Previous research which examined therapists’ countertransference towards suicidal patients suggested that therapists are negatively affected and distressed by them. We hypothesize that this emotional response of the therapists is related to specific sessions in which suicidality came up as a topic. Accordingly, the objective of this study consists in examining therapists’ emotional state on a session level of analysis. Methods The sample consisted of N = 21 adolescents (age 13–19 years) with BPD or subthreshold BPD. Therapists’ emotional states were measured in n = 418 sessions using the Session Evaluation Questionnaire. Principal component analysis was used to reduce dimensionality of the therapist response. The emotional states were compared depending on whether suicidality has been addressed in the session (SS) or not (NSS). Results Two components could be identified. Firstly, therapists were more aroused, excited, afraid, angry and uncertain after SS than after NSS. Secondly, therapists were more aroused, excited, definite and pleased after SS than after NSS. Discussion: Suicidality does not always have to be a burden for therapists: Both a “distress” and an “eustress” component occur in this context from which the latter is supposed to help clinicians master a difficult situation. Since countertransference feelings are often not fully conscious, it is necessary to do research on therapists’ emotional states after sessions in which suicidality is addressed. This is crucial to both prevent the therapeutic process from being endangered and preserve clinicians’ mental health. Clinical implications and limitations are discussed. Introduction Suicidality is defined as "the risk of suicide, usually indicated by suicidal ideation or intent, especially as evident in the presence of a well-elaborated suicidal plan" [1]. 50 to 95% of the clinicians report having experienced patients with some form of suicidal ideation or behavior [2][3][4]. Moreover, suicidal statements [5] and suicidal ideation [6] were considered as one of the most stressful aspect of the therapeutic work. The current study investigates the impact of suicidality addressed in psychotherapeutic sessions on therapists' post-session emotional state. Suicidality plays an important role in Borderline Personality Disorder (BPD). 60 to 70% of patients with BPD commit a suicide attempt at some point in their life and the rate of completed suicide is 50 times higher in patients with BPD than in the general population [7,8]. Despite the controversy regarding the diagnosis of BPD in underage patients, there is convincing evidence that diagnosing BPD in adolescents is as reliable and valid as in adults [9]. The elimination of the age limit of previous versions for the diagnosis of personality disorders in the International Classification of Diseases 11th revision (ICD-11 [10];) and the consequent possibility of early diagnosis provides the opportunity for effective early intervention [11]. Thus, a chronification and longlasting impact of the personality disorder could be prevented. The term countertransference is of great historical importance for therapeutic work since it has heightened therapists' awareness of their own feelings and has been coined in different ways. In the last decades, it has received more attention since it was acknowledged having great impact on treatment process and outcome [2,12]. The totalistic view defines countertransference as the total emotional reaction of the psychotherapist to the patient in the treatment situation [13]. The term countertransference is important in that it provides a historical framework for this question, but for the sake of simplicity and conceptual clarification, the following study will refer to "therapist's emotional response" or the operationalized form of it, "emotional state". In association with suicidal patients, the term "countertransference hate" was introduced [14], which can arise in psychotherapists when working with these clients. Its management through full awareness is essential for therapy success, since therapists' emotional response is assumed to predict patients' suicide outcome [15]. Moreover, the therapist's fear of losing a patient to suicide impedes an adequate assessment of the patient's inner experience, leads to countertransference reactions and thus to alliance ruptures [16,17]. The fear that a client could commit suicide represents a widespread therapeutic feeling [18] and interferes with their ability to work [19]. Therefore, it is crucial to conduct research on therapists' emotional states after psychotherapy sessions in which suicidality is an issue. Scientific evidence about therapist' countertransference towards suicidal patients is broadly consistent and suggest that it involves negative emotions such as anxiety, anger and higher self-reported distress [3,5,6,14]. Therapist stress is related to increased feelings of anxiety, tension, hopelessness, fear or embarrassment [20]. Others assume a wider spectrum of reactions ranging from hopelessness or sense of failure to the desire to intrusively nurture the patient [21]. A recent systematic review which included ten quantitative studies examined health care professionals' countertransference toward suicidal patients [22]. Results showed that "suicidal patients elicit disinterest, anxiety, confusion, overwhelming, entrapment, rejection, inadequacy, helplessness or distressbut also engangement and fulfillmentamong healthcare professionals, which suggests a specific and mostly adverse suicidal-related countertransference." (p. 10). Moreover, the authors draw the conclusion that current suicidal ideation seems to be involved in eliciting such countertransference [12,23], whereas the evidence on current or past suicidal behavior is inconclusive [12,[24][25][26]. Nevertheless, it should be stated that study designs, settings, measurement methods of countertransference and suicidality varied widely across the included studies, making any quantitative synthesis of findings difficult. Since previous research used a between-subject design comparing therapist responses towards suicidal versus towards non-suicidal patients, the question remains whether specific suicidal statements or general personality structure mediated the correlation with therapist response. Moreover, research has mostly focused on therapist traits rather than states such as affect. Therefore, it is important to examine within-therapist variables to further the understanding of therapist factors [27]. The current study meets the need for a prospective study design that can reveal differences between sessions within the patient-therapist dyad. In this way, it can be investigated what influence actual suicidal expressions have on clinicians. Our hypothesis is that suicidality addressed in a psychotherapy session burdens the therapist in terms of more negative emotions and distress measured after the session. The aim of this study is to discover a pattern of therapist' emotional state after sessions in which suicidality was addressed. This awareness of the emotional state after specific sessions allows for a concrete intervention and prevention to ensure both therapists' and patients' mental health. Methods Some of the method passages are adopted from Zimmermann et al. [28,29] where the study design has been described in more detail. Ethics approval and consent to participate All methods were carried out in accordance with relevant guidelines and regulations. All experimental protocols were approved by the 'Ethikkommission Nordwestund Zentralschweiz'. Ethical approval was obtained from the local ethics committee 'Ethikkommission Nordwestund Zentralschweiz'. All adolescents, their parents and the therapists provided their informed consent. Patient sample This study is part of the multi-center study 'Evaluation of Adolescent Identity Treatment' [28,30] that has been registered at clinicaltrials.gov (NCT02518906). The study aimed at showing non-inferiority of Adolescent Identity Treatment vs Dialectic Behavior Therapy as the outcome research part but, additionally, aims at answering a number of psychotherapy-process questions (see [28]for details). The current analyses are based on the entire available data collected at one participating center (Psychiatric Hospitals of the University of Basel, UPK). The patients were recruited between September 2015 and September 2019 from a specialized consultation for patients with personality disorders which is part of the UPK. A total of N = 23 adolescents (N = 21 female, N = 2 male) participated in this study. For each psychotherapy 25 sessions of Adolescent Identity Treatment were planned (AIT; [31]). To increase sample homogeneity, male participants were excluded from data analysis: In woman, deliberate self-harm is often used to communicate distress or to modify other peoples' behavior whereas in males deliberate self-harm is associated with greater suicidal intent [32]. Moreover, suicidal ideation is reported far more often by woman than by men. For this reason, the two male patients could be outliers. The following inclusion criteria were applied for the patients: age 13-19 years; three or more BPD criteria (Structured Clinical Interview for DSM-IV Axis II Personality Disorders [SCID-II] [33];); and identity diffusion according to the Assessment of Identity Development in Adolescence (AIDA; total t score > 60 [34,35];). The mean age of the remaining 21 female patients was 16.3 (SD = 1.6) years. Fifteen patients presented with BPD and six with subthreshold-BPD (three or four fulfilled BPD criteria in SKID-II). Six patients dropped out of treatment but were included in this study. Nine recordings of therapeutic sessions were missing due to technical difficulties or human failure. Figure 1 shows the available sessions for each patient. Psychotherapy Eight clinicians were involved in the current study of whom six were female (75%). All therapists were psychologists or psychiatrists who underwent or were currently undergoing psychotherapy training to obtain the Swiss specialist degree "Fachpsychologe/in für Psychotherapie FSP" or "Facharzt für Psychiatrie und Psychotherapie FMH". They were either focused on psychodynamic or systemic psychotherapeutic approaches for children and adolescent. Additionally, the therapists underwent training and supervision in Adolescent Identity Treatment which is a manualized therapeutic approach (AIT [31];). "AIT is a psychodynamic method for the treatment of personality disorders in adolescents and integrates modified elements of Transference-Focused Psychotherapy (TFP [36];) with behavior-oriented home-plans, psychoeducation and a stronger focus on working with parents. The main techniques consist of clarification, confrontation and interpretation focusing on affects in the here and now and on dominant object-relationship dyads [31]. All psychotherapists received specific training for AIT. Adherence and competence of the psychotherapists was checked by one of the authors of the AIT manual. Suicidality sessions (SS) After each session, the clinicians had to note information on essential moments in the last session and on suicidality ("Was current suicidality addressed?"). Sessions are treated as "sessions in which suicidality was addressed" (SS) if suicidal tendency had either been mentioned as an essential moment or if the question on current suicidality had been answered with yes. All the other sessions were labeled as "sessions in which suicidality was not addressed" (NSS). The term suicidality comprises current suicidal ideation, a concrete suicide plan and suicide attempt and excludes non-suicidal selfinjury behavior. Figure 1 shows SS and NSS for each patient. Session evaluation questionnaire (SEQ) Therapist response was measured using the Session Evaluation Questionnaire (SEQ; [37][38][39]), Form 5. It comprises 21 seven-point bipolar adjective scales on which therapists and patients are instructed to circle the appropriate number to show how they perceived the previous session. The items are divided into two sectors: Session evaluation ("This session was …" ) and postsession mood ("Right now I feel..."). Post-session mood included: happy-sad, angry-pleased, moving-still, uncertain-definite, calm-excited, confident-afraid, friendly-unfriendly, slow-fast, energetic-peaceful and quiet-aroused. Every item can either be subsumed under the subscale "positivity" or "arousal". Internal consistency, measured by coefficient alpha, was high for all SEQ indexes across different settings and conditions [39]. Of particular interest for this study were the items calm-excited and quiet-aroused from the subscale "arousal", and the items confident-afraid, angry-pleased and uncertain-definite from the subscale "positivity". The German version (SEQ-D) was used which presents with good psychometric properties [40]. Statistical analyses Data analysis was performed with R [41]. Principal component analysis (PCA) was used to reduce the dimensionality of the five therapist response variables (angrypleased, calm-exciting, confident-afraid, quiet-aroused, uncertain-definite). PCA performs this task by creating new uncorrelated variables that successively maximize variance. The result is defined by the data at hand and not a priori [29,42]. A scree plot was used to select relevant principal components. For hypothesis testing, a linear mixed-effect model with random intercept was used (R package 'nlme' [43]) to predict each relevant principal component of the therapist responses in the Session Evaluation Questionnaire (SEQ). Data about SS vs NSS was used as the dependent variable. The patient-therapist dyad was used as the grouping factor for the random effects. For statistical significance, alpha was set to 0.05. Model assumptions were verified using Q-Qand residual-plots. 69%. Five patients (46%) mentioned suicidality twice during psychotherapy, two patients (18%) three times, one patient once, one patient four times, one patient ten and one eleven times (M = 3.82, SD = 3.24, Mdn = 2). Principal component analysis This step was used to extract principal components from the SEQ data as a preparatory step to later correlate the extracted components with observed suicidality (see section "Hypothesis testing" below). The five principal components (PC1, PC2, PC3, PC4, and PC5) extracted from the SEQ-Items (calmexcited, quietaroused, confidentafraid, uncertaindefinite and angrypleased) explained 53, 23,9,8, and 7% of the variance in the suicidality data. In Table 1, the correlation of the original variables with the principal components is shown. According to the Scree-Test [44] and the Kaiser-Guttman rule [45,46], two components (PC1 and PC2) were extracted. PC1 explained more than half of the variance (53%). It correlated positively with the items calmexcited, confidentafraid and quietaroused, and negatively with the items uncertaindefinite and angrypleased. High PC1 scores involve therapist uncertainty, anger, excitement, fear and arousal. PC2 explained 23% of the variance and correlated positively with the items quietaroused, angrypleased, calmexcited, and uncertaindefinite and negatively but negligibly with the item confidentafraid. Higher PC2 scores go along with higher therapist arousal, satisfaction, excitement and definiteness. Description of therapist response according to session evaluation questionnaire (SEQ) Descriptive data of therapist response measured by the SEQ items is shown in Table 2. The lower a score is the more it corresponds to the left adjective in the bipolar scale, the higher a score is the more it corresponds to the right one. The average score over all items was M = 3.37 in SS and M = 2.82 in NSS which indicates that on average therapists felt more angry, excited, afraid, uncertain and aroused after SS. PC1-and PC2-scores condense these results. Discussion The aim of the current study was to investigate therapists' emotional responses toward young female adolescents with borderline pathology talking about their suicidal ideations or about a concrete suicide plan. We hypothesized that therapists' post-session emotion would differ depending on whether suicidality came up as a topic or not (SS vs NSS). We expected that therapists are more afraid, excited, aroused, uncertain and angry after SS than after NSS. Therapist response was measured by Session Evaluation Questionnaire (SEQ) which had to be filled out by all therapists after each session. Results showed that SS are accompanied by higher therapist arousal. PC1 reflects that SS are associated with higher clinician excitement, anger, fear, arousal and uncertainty than NSS. Moreover, PC2 explained 23% of the variance in the therapist response items predicted by suicidality. This pattern is also characterized by higher arousal and excitement but demonstrates as well satisfaction and definiteness. Previous research on this topic has especially dealt with therapists' emotional responses towards "suicidal patients". For example, a study with suicidal adolescents has shown that due to therapists' countertransference and due to trait-like suicidal ideation, therapists are unwilling to treat those patients at risk for suicide [47]. In patients with BPD it was suggested that the history of suicide attempts was associated with a worse general level of personality organization [48]. In contrast, the current study underlines the importance of significant psychotherapy moments with regard to therapists' emotional response and shows that specific episodes and sessions affect therapists. Shifting the focus from "difficult patients" to identifying subjectively difficult or even precarious moments in therapy allows clinicians to seek help on a more tangible level. PC1 reflects therapists' negative affect and arousal and is to be summarized in the term "distress". This is consistent with qualitative research [5,6] and seems highly associated with patients' suicidal statements since this component explains more than half of the variance in the SEQ data. Distress is to be understood as the "negative stress response, often involving negative affect and physiological reactivity: a type of stress that results from being overwhelmed by demands, losses or perceived threats. It has a detrimental effect by generating physical and psychological maladaptation and posing serious health risks for individuals. This generally is the intended meaning of the word stress" [1]. The concept of distress [49] emphasizes the importance of feeling able to meet the challenge by appropriate coping strategies. Therapists are often confronted with difficult situations in therapy and have a lot of responsibility. To ensure therapists' mental health, which is the foundation of their work [50], it is important to strengthen therapists' coping strategies. In contrast to this, PC2 seems to point towards a positive kind of arousal whereby therapists feel definite and pleased. PC2 shows similarities with the concept of "eustress" and will be referred to as 'eustress-component'. In the APA Dictionary of Psychology, eustress is understood as a "positive stress response, involving optimal levels of stimulation: a type of stress that results from challenging but attainable and enjoyable […] tasks […]. It has a beneficial effect by generating a sense of fulfillment or achievement and facilitating growth, development, mastery, and high levels of performance" [1]. This state is not a threat but a challenge and leads to positive feelings if the therapist thinks the situation was well managed. The correlation of SS with the eustresscomponent is of importance for clinical practice: Patient suicidality does not automatically have to be burdening for therapists and they can even feel more pleased and definite after sessions in which suicidality was addressed compared to sessions in which it was not. Addressing suicidality in psychotherapy can be a window of opportunity. Of course, causality cannot be inferred from this correlation of patient suicidality and therapists' emotional state after session in which suicidality is addressed. For example, an alternative hypothesis could be stated that therapists' emotional state during the session has an influence on the patients in whether or not they raise the issue of suicidality. Another variant could be that the mood of the therapists has nothing to do with the session per se and thus represents a finding by chance. Moreover, Krause & Lutz [51] point out that therapist, patient, and context variables cannot be separated and affect each other. In this respect, it can be assumed that therapists respond to patients, but patients also respond to therapists, which makes any conclusive statement impossible. Thus, the therapeutic situation is a result of responsiveness both of the part of the therapist as of the patient. Clinical implications Although countertransference is a complex concept with a meaningful historical background, it offers a good possibility to do justice to the multifaceted feelings therapists are confronted with during psychotherapyboth consciously and unconsciously. Since working with suicidal patients can be a very challenging and anxietyprovoking task for psychotherapists, their awareness of concrete emotional patterns in sessions where patients seem at risk for suicide is essential and should discourage the temptation of acting out unconscious feelings or conflicts [52]. Avoiding therapist errors emanating from negative countertransference improves the therapeutic relationship and finally also the patient's suicidal tendency [53]. Thus, supervision and notes on therapy sessions should sensitize clinicians to their emotional reaction for specific sessions where suicidality has come up as a topic. Secondly, we proposed that addressing suicidal ideation during psychotherapy is associated with both distress and eustress in therapists. The question remains of what distinguishes the therapeutic "distress" reaction from the "eustress" reaction and how the former can be transformed into the latter. In this regard, our results suggest that the main difference between these two emotional responses is that the "eustress" response includes a sense of definiteness and satisfaction and the absence of anxiety, anger and uncertainty which in turn facilitates development and the mastery of a concrete situation. Therefore, it is crucial that therapists feel definite, self-effective and act in the feeling that they are part of the therapy process. By being aware of possible conflictive emotions, a stable therapeutic relationship can be established which in turn leads to a promising outcome. A focus for how therapists can be supported to feel comfortable in challenging situations and a clinical culture that allows for errors and uncertainties should be established. Again, supervision can be helpful for clinicians to gain understanding of a difficult moment in the therapeutic process. Thirdly, it should be kept in mind that suicidal ideation is not always reason for an alarm and that it can overshadow other important issues that should be discussed in psychotherapy [54]. Accordingly, it is suggested thatalthough suicidal ideation should always be taken seriouslythe amount of focusing on suicidality should be skillfully balanced with the focus on other meaningful topics. Limitations Despite the importance of this topic, some limitations of the current study need to be considered. First of all, the study includes a relatively small sample size of N = 21 dyads. Only 11 of these dyads discussed the topic of suicidality. However, in terms of investigated sessions, the sample size was 418 sessions of which 211 sessions stemmed from dyads who discussed suicidality. The sample size in the current study is based on a power analysis which was aimed to sufficiently power an outcome comparison (see study design in [28]). It was difficult to estimate how many of the patients would discuss suicidality during the psychotherapy. In this sense, the current study has an explorative character that can serve as a basis for future studies. A replication of the current results is necessary to confirm the observed effect which was significant despite the small sample size. Secondly, the patients' suicidality was not documented with a validated scale. The therapists documented whether suicidality was addressed or not after each session. Our rationale is that the discussion of suicidality is a salient event and determining whether the event occurred or not should be simple and straightforward. However, a validated scale might be helpful in getting more differentiated results to determine, for instance, the perceived intensity of addressing the topic or the topic's actual acuteness for the patient. Thirdly, therapist response was measured with the Session Evaluation Questionnaire (SEQ) which is a self-rating questionnaire that is not constructed as a stress-test. Although this seems adequate for a subjective description of postsession mood, it should be complemented by a validated stress questionnaire and by psychophysiological measures such as saliva cortisol level or electrodermal skin response to measure the reaction at the level of the autonomous nervous system [55][56][57]. Conclusion Two different therapist responses were identified after sessions in which young female patients with BPD or subthreshold BPD talked about their suicidal ideation or about a more concrete suicide plan. Therapists showed an emotional response pattern consisting firstly of arousal and negative affect ("distress") and secondly of arousal and positive affect ("eustress"). These two types of arousal are distinguished by being angry, uncertain and afraid versus feeling pleased and definite. First of all, this implies that patient suicidality does not always have to be emotionally burdening for therapists and moreover, that clinicians need to feel definite and selfeffective which in turn enables the mastery of a challenging situation. If therapists become aware of their "suicidality-countertransference" and their anxiety about patient suicide contained, this can on one hand enable a good therapy process and outcome (including prevent discontinuation of the therapeutic process) and on the other hand lead to clinicians' well-being. Conducting further research on how therapists' mental health can be preserved seems necessary in order to ensure appropriate and good care for patients with suicidal tendencies.	2021-11-23T14:56:11.915Z	2021-11-23T00:00:00.000	{ "year": 2021, "sha1": "325c1c8fe722fd9fb83764218f774b29e0a2a37f", "oa_license": "CCBY", "oa_url": "https://bmcpsychiatry.biomedcentral.com/track/pdf/10.1186/s12888-021-03549-9", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "325c1c8fe722fd9fb83764218f774b29e0a2a37f", "s2fieldsofstudy": [ "Psychology" ], "extfieldsofstudy": [ "Medicine" ] }
2518658	pes2o/s2orc	v3-fos-license	The free space optical interference channel Semiclassical models for multiple-user optical communication cannot assess the ultimate limits on reliable communication as permitted by the laws of physics. In all optical communications settings that have been analyzed within a quantum framework so far, the gaps between the quantum limit to the capacity and the Shannon limit for structured receivers become most significant in the low photon-number regime. Here, we present a quantum treatment of a multiple-transmitter multiple-receiver multi-spatial-mode free-space interference channel with diffraction-limited loss and a thermal background. We consider the performance of a laser-light (coherent state) encoding in conjunction with various detection strategies such as homodyne, heterodyne, and joint detection. Joint detection outperforms both homodyne and heterodyne detection whenever the channel exhibits"very strong"interference. We determine the capacity region for homodyne or heterodyne detection when the channel has"strong"interference, and we conjecture the existence of a joint detection strategy that outperforms the former two strategies in this case. Finally, we determine the Han-Kobayashi achievable rate regions for both homodyne and heterodyne detection and compare them to a region achievable by a conjectured joint detection strategy. In these latter cases, we determine achievable rate regions if the receivers employ a recently discovered min-entropy quantum simultaneous decoder. I. INTRODUCTION The principal goals of information theory are to determine the ultimate limits on reliable communication and to find ways of approaching these limits in practice. Pointto-point optical communication using laser-light modulation in conjunction with direct-detection and coherent-detection receivers has been studied in detail using the semiclassical theory of photodetection [1]. This approach treats light as a classical electromagnetic field, and the fundamental noise encountered in photodetection is the shot noise associated with the discreteness of the electron charge. These semiclassical treatments for systems that exploit classical-light modulation and conventional receivers (direct, homodyne, or heterodyne) have had some success, but we should recall that electromagnetic waves are quantized, and the correct assessment of systems that use non-classical light sources and/or general optical measurements requires a full quantum-mechanical framework [2]. Consider several recent theoretical studies on the point-to-point [3], [4], broadcast [5] and multiple-access [6] bosonic channels with linear loss and a thermal background. These studies have shown that achievable communication rates surpass what can be obtained with conventional receivers. Prior work has established that Holevo information rates are achievable for information transmission on general quantum point-to-point [7], [8], broadcast [9] and multiple-access [10] channels. In each case, the Holevo information rates are an upper bound to the Shannon rates computed for any specific transmitter-modulation receivermeasurement pair. For the general quantum channel, attaining Holevo information rates may require collective measurements (a joint detection) across the channel outputs. The next level of complexity beyond the point-to-point, broadcast, or multiple-access channels is arguably captured by the interference channel [11], [12], [13]. This channel, in general, can have M senders and M receivers (M ≥ 2), where each sender would like to communicate only with a partner receiver, 1 but most research focuses on the special case of two senders and two receivers. The Gaussian interference channel has been analyzed in depth in the classical information-theory literature, and this model readily applies for an optical interference channel with coherent-state inputs and coherent detection. Calculating the capacity of this channel in the general case has been an open problem for some time, but several researchers have found it in the special cases of "very strong" [11] and "strong" interference [12], [13]. Also, Han and Kobayashi determined the best inner bound on the channel's capacity, by having each receiver partially decode the message of the other sender along with a full decoding of the message of the partner sender [13]. However, all of these strategies assume that the underlying channel is classical, and we would expect a quantum strategy with a power-constrained encoding and collective measurement at the receivers to outperform such strategies. In this paper, we consider a pure-loss thermal-noise bosonic interference channel, particularly in the context of freespace (wireless) terrestrial optical communications. We assume a coherent-state encoding throughout this paper. We find achievable rate regions for the two-sender two-receiver channel with coherent-detection receivers, for the "very strong" and "strong" interference regimes. We find the rate region achievable with a joint-detection receiver (JDR) under "very strong" interference. We also determine a "min-entropy" JDRachievable rate region for the "strong" interference regime by exploiting a recent result from Ref. [14], and we find a conjectured JDR-achievable region were a conjecture from Ref. [14] regarding quantum simultaneous decoding true (c.f, page 4-15 of Ref. [15] for a classical simultaneous decoder). Next, we evaluate the Han-Kobayashi rate region for homodyne and heterodyne detection in the general case, and we show an achievable rate region using a "min-entropy" quantum simultaneous decoder from Ref. [14]. Finally, we conjecture a Han-Kobayashi-like achievable rate region using Conjecture 2 from Ref. [14]. Our results here differ from those in Ref. [14]-there some of us considered the general quantum interference channel whereas here we consider specifically a bosonic interference channel. II. A FREE-SPACE OPTICAL INTERFERENCE CHANNEL Consider a range-L line-of-sight free-space optical channel with hard circular transmit and receive apertures of areas A t and A r respectively. Assume λ-center-wavelength quasimonochromatic transmission. In the near-field propagation regime (Fresnel number product, D f ≡ A t A r /(λL) 2 1), a normal-mode decomposition of the free-space optical channel yields M ≈ 2D f orthogonal spatio-polarization transmitterto-receiver modes (D f spatial modes, each of two orthogonal polarizations) with near-unity transmitter-to-receiver power transmissivities (η m ≈ 1). In the far-field propagation regime (D f 1), only two orthogonal spatial modes (one of each orthogonal polarization) have appreciable power transmissivity (η ≈ D f for each mode). Sender m modulates her information on the m th transmitterpupil spatial mode, and Receiver m separates and demodulates information from the corresponding receiver-pupil spatial mode. With perfect spatial-mode control at the transmitter and perfect mode separation at the receiver, the orthogonal spatial modes can be thought of as independent parallel channels with no cross talk. However, imperfect (slightly non-orthogonal) mode generation or imperfect mode separation can result in cross talk (interference) between the M channels. We take our interference channel model as a passive linear mixing of the input modes along with the possibility of a thermal environment adding zero-mean, isotropic Gaussian noise. Although the results here apply to cyclic interference channels with M senders and M receivers, for simplicity, we limit ourselves to the M = 2 case, in which case the channel model reduces tô where η 11 , η 12 , η 21 , η 22 ,η 1 ,η 2 ∈ R + , √ η 11 η 12 = √ η 21 η 22 , The following conditions ensure that the network is passive: We constrain the mean photon number of the transmittersâ 1 andâ 2 to be N S1 and N S2 photons per mode, respectively, the environment modesν 1 andν 2 are in statistically independent zero-mean thermal states [2] with respective mean photon numbers N B1 and N B2 per mode. Note that for a coherent-state encoding and coherentdetection at both receivers, the above model is a special case of a complex Gaussian interference channel, and we can study its capacity regions in various settings by applying the results from Refs. [11], [12], [13]. If the senders prepare their inputs in coherent states \|α 1 and \|α 2 , with α 1 , α 2 ∈ R, and both receivers perform real-quadrature homodyne detection on their respective modes, the result is a classical Gaussian interference channel [2], where Receivers 1 and 2 obtain respective conditional Gaussian random variables Y 1 and Y 2 distributed as where the "+1" term in the noise variances arises physically from the zero-point fluctuations of the vacuum. Suppose that the senders again encode their signals as coherent states \|α 1 and \|α 2 , but this time with α 1 , α 2 ∈ C, and that the receivers both perform heterodyne detection. This results in a classical complex Gaussian interference channel [2], where Receivers 1 and 2 detect respective conditional complex Gaussian random variables Z 1 and Z 2 , whose real parts are distributed as III. VERY STRONG INTERFERENCE Carleial determined the capacity region of a classical Gaussian interference channel in the case of "very strong" interference [11]. Suppose that X 1 and X 2 are the input random variables for Senders 1 and 2 and that B 1 and B 2 represent the outputs for Receivers 1 and 2, respectively. Then the most general way to state the condition for very strong interference is that the following information inequalities should hold for all input distributions p X1 (x 1 ) and p X2 (x 2 ) [15]: Carleial proved the surprising result [11] that the capacity region of the interference channel under this setting is the convex closure of positive rate pairs (R 1 , R 2 ) such that for some input distributions p X1 (x 1 ) and p X2 (x 2 ). The coding strategy is simply for each receiver to decode first what the other sender is transmitting, remove this signal, and then decode the message intended for him. The conditions in (3)(4) translate to the following ones for the case of coherent-state encoding and coherent detection: and the capacity region becomes where i = 1 for homodyne detection and i = 0 for heterodyne detection. We can also consider the case when the senders employ coherent-state encodings and the receivers employ a jointdetection strategy on all of their respective channel outputs. The conditions in (3)(4) readily translate to this quantum setting where we now consider B 1 and B 2 to be quantum systems, and the information quantities in (3)(4) and (5) now become Holevo informations [14]. Theorem 6 in Ref. [14] provides a simple proof that Carleial's result applies in the quantum domain to these Holevo informations. So, the conditions in (3)(4) when restricting to coherent-state encodings translate to where g (N ) is the entropy of a thermal state with mean photon number N : An achievable rate region is then These rates are achievable using a coherent-state encoding, but not necessarily optimal (though they would be optimal if the minimum-output entropy conjecture from Refs. [3], [16] were true). Nevertheless, these rates always beat the rates from homodyne and heterodyne detection, and Figure 1 displays examples of the capacity (and achievable rate) regions in the low-and high-power regimes. IV. STRONG INTERFERENCE Sato [12] and Han-Kobayashi [13] independently determined the capacity of a classical Gaussian interference channel under "strong" interference. A channel has "strong" interference if the following information inequalities hold for all p X1 (x 1 ) and p X2 (x 2 ) [15]: High Power Hom. Het. Joint R R Fig. 1. Capacity regions for coherent-state encodings and coherent detection, and achievable rate regions for coherent-state encodings and joint-detection receivers-both with η 11 = η 22 = 1/16 and η 12 = η 21 = 1/2 ("very strong" interference for coherent detection and such that (8-9) hold). The LHS displays these regions in a low-power regime with N S 1 = N S 2 = 1 and N B 1 = N B 2 = 1, and the RHS displays these regions in a highpower regime where N S 1 = N S 2 = 100. Homodyne detection outperforms heterodyne detection in the low-power regime because it has a reduced detection noise, while heterodyne detection outperforms homodyne detection in the high-power regime because its has an increased bandwidth. The capacity region of the classical interference channel under this setting is the convex closure of positive rate pairs (R 1 , R 2 ) such that [12], [13], [15]: The conditions in (10)(11) translate to the following ones for coherent-state encoding and coherent detection: and the capacity region has the two inequalities in (6-7) and an additional bound on the sum rate: where again i = 1 for homodyne detection and i = 0 for heterodyne detection. The situation for a joint-detection strategy over all of the channel outputs becomes more complicated for the case of "strong" interference, because we require a quantum simultaneous decoder [14] in order to achieve the information rates in (12)(13) with B 1 and B 2 becoming quantum systems. Such a simultaneous decoder is analogous to a classical simultaneous decoder (e.g., see page 4-15 of Ref. [15]), but we have not yet been able to prove the existence of it in the quantum case (see Conjecture 2 of Ref. [14]). Yet, we do have an achievable simultaneous decoding strategy expressed in terms of minentropies (see Theorem 4 of Ref. [14]), where the min-entropy of a probability distribution is the negative logarithm of the probability of its mode [17], and, as a simple extension of this idea, the min-entropy of a density operator is the negative logarithm of its maximum eigenvalue. For a thermal state with average photon number N B , its min-entropy is ln (N B + 1), Fig. 2. The parameters of the free-space channel are N B 1 = N B 2 = 1, η 11 = η 22 = 0.3, η 21 = η 12 = 0.6, N S 1 = N S 2 = 2 for the low-power regime, and N S 1 = N S 2 = 100 for the high-power regime. The LHS figure depicts the "strong" interference capacity regions in the low-power regime for homodyne and heterodyne detection, the achievable rate region with the conjectured joint detection, and the convex hull over regions arising from different min-entropy decoding strategies. Homodyne detection outperforms heterodyne detection in this regime, and the min-entropy decoders perform the worst. The RHS figure displays the same regions in the high-power regime. Heterodyne detection outperforms homodyne detection, and interestingly, the min-entropy decoders outperform heterodyne detection. and this result allows us to determine an achievable rate region with a simultaneous decoding strategy (similar to the simultaneous decoding inner bound on page 6-7 of Ref. [15]): This is one particular variation of an achievable strategy in which we have the bounds on the individual rates expressed in terms of min-entropies and the bound on the sum rate expressed with von Neumann entropies, though note that there are other variations we could consider in light of Theorem 4 of Ref. [14]. We can then take the convex hull of the achievable rate regions for these different strategies to get an achievable rate region for a min-entropy quantum simultaneous decoder (the RHS of Figure 2 displays an interesting example of such a "min-entropy" region). If Conjecture 2 of Ref. [14] regarding the existence of a quantum simultaneous decoder were true, then the rate region in (12-13) would be achievable under under the conditions of (10-11) (with Holevo information rates replacing Shannon rates). Figure 2 displays the different capacity and achievable rate regions when a free-space interference channel exhibits "strong" interference. V. HAN-KOBAYASHI RATE REGIONS The Han-Kobayashi region is the best known achievable rate region for the classical interference channel [13]. The coding strategy to achieve this region is for each receiver to decode partially the other sender's message while fully decoding the partner sender's message. With this strategy, the four parties can choose to take advantage of channel interference while achieving the task of paired sender-receiver communication. A compact description of the Han-Kobayashi region comes from its reduction with a Fourier-Motzkin elimination algorithm [18]. It is the convex closure of all positive rate pairs (R 1 , R 2 ) satisfying the following inequalities and the inequalities obtained from the ones below by swapping the indices 1 and 2: In the above, U m is the "personal" random variable of Sender m, and W m is her "common" random variable. The Han-Kobayashi coding strategy readily translates into a strategy for coherent-state encoding and coherent detection. Sender m shares the total photon number N Sm between her personal message and her common message. Let λ m be the fraction of signal power that Sender m devotes to her personal message, and letλ m denote the other fraction of signal power that Sender m devotes to her common message. The inequalities above become the following ones for the case of coherent-state encoding along with coherent detection: R 1 ≤ γ η 11 λ 1 N S1 η 21 λ 2 N S2 + N 1 + γ η 12λ1 N S1 η 12 λ 1 N S1 + N 2 , R 1 + R 2 ≤ γ η 11 λ 1 N S1 η 21 λ 2 N S2 + N 1 + γ η 12λ1 N S1 + η 22 N S2 η 12 λ 1 N S1 + N 2 , where γ (x) = ln (1 + x) /2 i , N m = 2 iη m N Bm + 1 /4 i , i = 1 for homodyne detection, i = 0 for heterodyne detection, and m ∈ {1, 2}. We also conjecture a Han-Kobayashi achievable rate region if the senders employ coherent-state encodings and the receivers exploit joint-detection receivers (this again follows from Conjecture 2 of Ref. [14] regarding the existence of a quantum simultaneous decoder). The inequalities for the region are similar to those in (14)(15)(16)(17)(18) It also depicts what we can achieve by employing 49 variations of the min-entropy decoding strategy from Ref. [14] and taking the convex hull of these achievable rate regions. Finally, it displays the conjectured region if a joint detection decoder were to exist (see Conjecture 2 of Ref. [14]). All of these regions are with respect to a 10%-personal, 90%-common Han-Kobayashi power split. "swapped" inequalities, with the exception that Holevo informations replace mutual informations. We can obtain an achievable rate region by exploiting the quantum simultaneous decoder from Theorem 4 of Ref. [14] that gives rates which are a difference of a min-entropy and a von Neumann entropy. The region's characterization is in terms of the Han-Kobayashi (HK) characterization with 14 inequalities [13], corresponding to two different multiple access channels (MACs) induced to each receiver by the HK coding strategy (seven inequalities for each MAC). There are 49 variations of these min-entropy decoders-one of the information rates in the seven inequalities for each MAC is von Neumann and the other six are min-entropy rates. Then taking the convex hull of these 49 different achievable rate regions gives an achievable rate region for a min-entropy decoding strategy. Figure 3 plots the regions achievable with coherent detection, the min-entropy decoder, and the conjectured joint detector for a particular HK power split. VI. CONCLUSION The semiclassical models for free-space optical communication are not sufficient to understand the ultimate limits on reliable communication rates, for both point-to-point and multiple-sender-receiver channels. We presented a quantummechanical model for the free-space optical interference channel and determined achievable rate regions using both structured and unstructured receivers. Interestingly, the minentropy decoder from Ref. [14] can achieve rates that are unachievable by both homodyne and heterodyne detection when the channel exhibits "strong interference." Finally, we determined the Han-Kobayashi inner bound for homodyne and heterodyne detection, and we conjectured a rate region of this form if a quantum simultaneous decoder were to exist. Several open problems remain for this line of inquiry. Perhaps the biggest open question is to prove Conjecture 2 from Ref. [14] concerning the existence of a quantum simultaneous decoder for a general quantum interference channel. Also, we do not know if a coherent-state encoding is in fact optimal for the free-space interference channel-it might be that squeezed state transmitters could achieve higher communication rates as in Ref. [6]. One could also evaluate the ergodic and outage capacity regions based on the statistics of η ij , which could be derived from the spatial coherence functions of the stochastic mode patterns under atmospheric turbulence.	2011-05-20T11:04:15.000Z	2011-02-13T00:00:00.000	{ "year": 2011, "sha1": "b4643ec494cdd8e95602806f9f853b8e6273a12e", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1102.2627", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "40badfd149cfc494796aa285666ab238de9bfdd5", "s2fieldsofstudy": [ "Physics", "Computer Science" ], "extfieldsofstudy": [ "Computer Science", "Physics", "Mathematics" ] }
18814438	pes2o/s2orc	v3-fos-license	The Quark Form Factor at Higher Orders We study the electromagnetic on-shell form factor of quarks in massless perturbative QCD. We derive the complete pole part in dimensional regularization at three loops, and extend the resummation of the form factor to the next-to-next-to-leading contributions. These results are employed to evaluate the infrared finite absolute ratio of the time-like and space-like form factors up to the fourth order in the strong coupling constant. Besides for the pole structure of higher-loop QCD amplitudes, our new contributions to the form factor are also relevant for the high-energy limit of massive gauge theories like QED. The highest-transcendentality component of our results confirms a result recently obtained in N=4 Super-Yang-Mills theory. Introduction The electromagnetic form factor of quarks is a quantity of considerable interest in Quantum Chromodynamics (QCD) and in gauge theories in general. At high photon virtualities Q 2 this quantity receives double logarithmic corrections of infrared and collinear origin [1], which take the form of double poles in dimensional regularization for the case of massless on-shell quarks studied in the present article. These contributions can be resummed by evolution equations in Q 2 based on universal factorization properties of the amplitude in the relevant kinematic limit, resulting in the well-known exponentiation of the form factor [2][3][4]. So far perturbative calculations have been performed up to two loops for both the massless on-shell case [5,6] and heavy quarks [7]. Accordingly, the exponentiation has been studied up to the next-to-leading (NL) contributions [8][9][10]. Higher-order corrections to the quark form factor are not only of general interest in quantum field theory, but also relevant for practical applications, as this quantity contributes to phenomenologically important processes. Research in the past years has yielded dramatic progress in next-to-next-to-leading order (NNLO) perturbative calculations, see, for example, Ref. [11] and numerous references therein. This progress also led to further investigations of the general structure of amplitudes and cross section at higher loop-orders, which in turn further stimulated the interest in all-order resummations. Consequently, the intimate connection between resummation and perturbative results at multiple loops has become much more prominent [12][13][14]. Very recently, we have presented the first complete calculation of the third-order corrections to a hard-scattering observable depending on a dimensionless variable, the structure function F 2 in photon-exchange deep-inelastic scattering [15]. After exploring the consequences of that result for the soft-gluon threshold resummation in Ref. [16], we here present its implications on the quark form factor (from now on always referring to the massless on-shell case, if not explicitly indicated otherwise) and its resummation. We are able, after extending the two-loop form factor beyond the previous order ε 0 in dimensional regularization, to derive the complete series of poles, ε −6 . . . ε −1 , at three loops. These terms in turn provide the coefficients required to extend the exponentiation of the form factor to the next-to-next-to-leading (NNL) contributions which we work out explicitly. This article is organized as follows: In Section 2 we address the resummation of the quark form factor. We briefly recall the evolution equation and its solution, and present the explicit expansion up to four loops in terms of two perturbative functions, A(α s ) (known up to three loops from the NNLO splitting functions [17,18]) and G(α s , ε). In Section 3 we extend the two-loop form factor to order ε 2 and extract the pole terms at three loops from our structure-function calculation [15]. These results are employed to extend the first-and second-order parts of the resummation function G to higher orders in ε, and to derive the leading-ε term at the third order in the strong coupling. Some first implications of these results are discussed in Section 4. Here we extend the ratio of the time-like and space-like form factors [9] to the fourth order in α s , compare to a recent result for The resummation of the quark form factor The subject of our study are the QCD corrections to the γ * qq (or γ * qq) vertex, where γ * denotes a space-like (or time-like) photon with virtuality Q 2 , and q/q a massless external quark / antiquark. Until Section 4 we will focus on the space-like case, thus the relevant amplitude is where the scalar function F on the right-hand side is the space-like quark form factor. This quantity can be calculated order by order in the strong coupling constant α s and, as mentioned above, is so far known to two loops [5,6]. F is gauge invariant, but divergent. As usual we work in dimensional regularization with D = 4 − 2ε, thus these divergences show up as poles ε −k in the present article. The exponentiation of the form factor, which extends beyond the resummation of renormalization group logarithms, is achieved by solving the well-known evolution equations [3,4,[8][9][10] Here µ represents the renormalization scale, and the functions G and K are subject to the renormalization group equations [3] All infrared singularities are collected by the scale-independent function K, which in the MS scheme consists of a series of poles in ε. The function G, on the other hand, is finite for ε → 0 and includes all dependence on the scale Q 2 . The renormalization properties of G and K are both governed by the same anomalous dimension A, because the sum of G and K is a renormalization group invariant. This quantity is given by a power expansion in the strong coupling, for which we use the convention (also employed for all other expansions in α s throughout this article) In fact, the anomalous dimension A also occurs in many other circumstances, for instance as the coefficient of the 1/(1 − x) + contribution to the Altarelli-Parisi quark-quark splitting function and as the anomalous dimension of a Wilson line with a cusp [22]. As already indicated by the argument of the beta function, the solution of Eqs. (2.3) and (2.4) requires the running coupling in D dimensions. Following Refs. [10,23] we defineā(λ, a s , ε), where λ is a dimensionless ratio of scales like λ = Q 2 /µ 2 . The resummation of the NNL contributions to the form factor requires the scale dependence ofā to NNLO accuracy [24,25] (see the discussion at the end of this section), obtained by solving with the boundary conditionā(1, a s , ε) = a s . Extending the result of Ref. [23] by one order, this solution is given bȳ where we have used the abbreviations With Eq. (2.7) for the running coupling, Eq. (2.3) can now be solved to the required accuracy, The perturbative expansion of the boundary condition G(1,ā, ε) can be derived by comparison to the fixed-order results for the form factor. After recursively determining (see, e.g., Ref. [10] for details) the scale-independent counterterm function K from Eq. (2.4), the resummed quark form factor reads ln F α s , with the boundary condition F (α s , 0, ε) = 1 [9]. After expanding the D-dimensional coupling according to Eq. (2.7), ln F exhibits double logarithms of Q 2 /µ 2 and double poles in ε, which are generated by the two integrations. In addition the integral over the anomalous dimension A leads to terms which are independent of the outer integration variable ξ. These logarithmic singularities at ξ = 0 are canceled by the function K order by order in the perturbative expansion. The well-known relation (2.10) can be employed either for a direct evaluation of the form factor due to the analyticity in D dimensions [10] or, by means of finite-order expansions and matching, for predictions of perturbative results at higher orders. Here we will focus on the latter issue. In particular, we will derive explicit results at three and four loops. This is done by performing the integrations in Eq. (2.10) after inserting the perturbative expansions of all quantities. The resulting integrals can be evaluated using algorithms for the evaluations of nested sums [26,27]. Some technical details for this step are given in Appendix A, where Eqs. (A.4)-(A.7) represent sample types of relevant integrals. Further details may also be found in Ref. [10]. It is convenient to express the loop-expanded form factor in terms of the bare (unrenormalized) coupling α b s instead of the renormalized coupling α s as in Eq. (2.10). The couplings α b s and α s are related by with the renormalization constant Z α s in the MS scheme given by and also the bare expansion parameter normalized as a b s = α b s /(4π) . The perturbative expansion of the bare (unrenormalized) quark form factor then reads (2.13) In terms of the i-th order parameters A i in Eq. (2.5) and the corresponding functions G i (ε), the expansion coefficients up to four loops read 14) The The ε 0 term of G 1 , together with β 0 and the lowest-order anomalous dimension A 1 , specify the two most singular terms ε −2n and ε −2n+1 to all orders α n s . Likewise, if (besides two more contributions to G 1 ) also the leading term of G 2 and the NLO quantities β 1 and A 2 are known, the resummation fixes the first four leading poles at each order. This has been the status up to now, referred to as the next-to-leading (NL) contributions in Section 1. In the next section, we will present the leading term of G 3 and the corresponding higher coefficients in the ε-expansions of G 2 and G 3 . Together with β 2 (as indicated before Eq. (2.6)) and our recent result for A 3 [17], these results provide the NNL terms at all orders, especially fixing the ε −4 and ε −3 poles in Eq. (2.17). Fixed-order results and resummation coefficients We now turn to the extraction of the quark form factor up to order α 3 s from our third-order computation of the deep-inelastic structure functions [15]. As also discussed in Refs. [17,18], the calculation has been performed via forward Compton amplitudes and the optical theorem. The cuts of the corresponding diagrams always include real-emission contributions, thus the purely virtual form-factor part cannot be directly read off at this level. Nevertheless we can reconstruct the form factor from our results, except (as explained below) at the highest power of ε which was consistently kept in the calculations. Consequently, we can derive all 1/ε pole terms at order α 3 s , since the forward Compton amplitudes have been computed to order ε 0 for Ref. [15]. Our starting point for the determination of the form factor is the unrenormalized (and unfactorized) partonic structure function F b for γ * q →X in the limit x → 1, where x denotes the partonic Bjorken variable. Using the end-point properties of the harmonic polylogarithms [28] in which these results are expressed, we remove all regular contributions and only retain the singular pieces proportional to δ(1 − x) and the +-distributions at order α n s , The resulting expressions are then compared to the general structure of the n-th order contribution F b n in terms of the l-loop form factors F l and the corresponding pure real-emission parts S l , (3. 2) The x-dependence of the factors S k is given by the D-dimensional +-distributions f kε defined by The α n s contributions F n and S n in Eq. (3.2) exhibit poles in ε up to order ε −2n . The corresponding bare structure function F b n , on the other hand, only include terms up to ε −n , as the higher divergences on the right-hand sides cancel for these inclusive quantities due to the Kinoshita-Lee-Nauenberg theorem [29,30]. In fact, the complete cancellation already occurs at the level of the individual diagrams of Ref. [15] for the forward Compton amplitude. Once the products of lower-order quantities in Eq. (3.2) have been subtracted from F b n , the contribution of the n-loop form factor F n can be extracted by performing the substitution does not work at the highest order of ε kept in the calculation of F b n . Hence, as stated above, the determination of the n-loop form factor F n to order ε k in this approach requires the calculation of the bare partonic structure function F b n to order ε k+1 . In addition, the subtraction of the lower-order contributions F l and S l with l < n in Eq. (3.2) requires the extension of these quantities to higher orders in ε. Specifically, the first-and secondorder quantities are required to order ε 3 and ε 1 , respectively, for the extraction of the pole terms of the three-loop form factor. These functions have been determined from the calculation of F b 1 to order ε 4 and F b 2 to order ε 2 . In fact, anticipating a future extension to the finite parts of the three-loop form factor F 3 , we have extended these calculations to one more power of ε, making use of the fact that the one-and two-loop integrals for the calculation of the structure functions were evaluated to order ε 5 and ε 3 anyway, see Table 3 of Ref. [15]. As a check of these new two-loop results (the one-loop quantities are known to all orders in ε anyway), a separate calculation of F 2 has been performed to order ε 2 in the approach of Refs. [5,6]. The results for the corresponding seven diagrams are listed in Appendix B. To the accuracy in ε just discussed, the unrenormalized quark form factor reads, up to three loops in the notation of Eq. (2.13), Here n f stands for the number of effectively massless quark flavours, C F and C A are the usual QCD colour factors, C F = 4/3 and C A = 3, and the values of Riemann's zeta function are denoted by ζ n . Eq. (3.7) and the ε 1 and ε 2 parts of Eq. (3.6) are new results of this article. The four highest 1/ε poles of the three-loop form factor F 3 provide the first complete verification of the resummation of the next-to-leading contributions. With the anomalous dimensions (2.5) known up to A 3 , the remaining two poles are sufficient to fix the NNL contributions to the function G in Eq. (2.9). Especially, we can derive the first (ε = 0) term of the third-order function G 3 (ε). Before we turn to these results we recall, for completeness, the known coefficients of the cusp anomalous dimension A(a s ). The results for A 1 and A 2 , have been known for a long time [31]. The recently completed expression for A 3 reads [17] See Refs. [32][33][34] for previous partial results on the n f -contributions. Very recently the ζ 2 2 term in Eq. (3.9) has been confirmed in Ref. [19], see the discussion at the end of Section 4. Inserting Eqs. (3.8) and (3.9) into the resummation relations (2.14) -(2.16) and comparing to the explicit results (3.5) -(3.7), we obtain the following perturbative expansion of Eq. (2.9) at µ 2 = Q 2 : As discussed at the end of Section 2, these results are sufficient to fix the next-to-next-to-leading contributions, i.e., the six highest poles in ε, to all orders in the strong coupling. In fact, in view of a future extension of G 3 to order ε, the first-and second-order results (3.10) and (3.11) already transcend this accuracy by one power in ε. We close this section by a brief discussion of our three-loop results (3.7) and (3.12). The former result for the 1/ε poles of the quark form factor in massless QCD is not directly applicable to any physical process. For use in cross section calculations such as e + e − → 2 jets at the next-to-nextto-next-to-leading order (N 3 LO), one would need the finite contribution to F 3 as well. However, the resulting leading term (3.12) of G 3 is of immediate interest for predictions of the pole structure of QCD amplitudes at higher orders [13,14] generalizing Catani's NNLO formula [12]. For the four-quark amplitude at N 3 LO, qq → qq, for instance, an explicit prediction has been derived in Ref. [13], for which Eqs. (3.7) and (3.12) now provide the last missing piece of information. The time-like case and non-QCD applications So far, our discussion has been restricted to space-like photon momenta, q 2 = −Q 2 < 0 . The modifications for the time-like case q 2 > 0 are obtained by analytic continuation. For the resummed quark form factor in Eq. (2.10) this continuation has been discussed in Ref. [9], while the finiteorder expansions (2.13) are transferred to q 2 > 0 according to [5] −q 2 Of particular interest is the absolute ratio \|F (q 2 )/F (−q 2 )\| of the renormalized time-like and space-like form factors. This quantity is infrared finite and directly enters the cross section for Drell-Yan lepton pair production in hadronic collisions. Transforming Eqs. (2.14) -(2.17) back to the renormalized quantities using Eqs. (2.11) and (2.12), and then employing the analytic continuation (4.1) we obtain the expansion in terms of the couplings a s (q 2 ) = a s (−q 2 ) = a s . Note that, since this ratio is infrared finite, only the ε = 0 parts of the coefficients G i enter Eq. (4.2). Consequently, all terms contributing at the fourth order are now known, with the exception of the four-loop cusp anomalous dimension A 4 of which only the small n 3 f contribution has been derived so far [32]. The effect of A 4 is expected to be small, therefore we can nevertheless evaluate the ratio (4.2) also numerically up to the fourth order, employing the [ This result does not look like a nicely converging expansion, but so far does not exhibit a clear factorial growth of the higher-order coefficients either. As already pointed out in Ref. [9], the only genuine l-loop contribution at order α l s is given by the anomalous dimension A l , which in Eq. (4.4) contributes 24%, 7% and (2 ± 1)% of the total coefficient at the second, third and fourth order, respectively. On the other hand, the contributions of the quantities G l−1 at order α l s are large, amounting to 37%, 41%, 50% at l = 2, 3, 4. Consequently, the higher-order ( l ≥ 5 ) terms in Eq. (4.4) cannot be predicted quantitatively at this point. Exponentiations like Eq. (2.10) for the form factor F have also been studied for electroweak interactions [20], where a fermion or gauge-boson mass m acts as a regulator for collinear or infrared singularities. Of course, both the counter-term function K in Eq. (2.4) and the lower integration limit in Eq. (2.10) are modified in this case, as they depend on the infrared sector of the theory. However, the leading (ε = 0) term of the function G in Eq. (2.3) is independent of the regulator at each order in the coupling constant. This contribution entirely originates in the so-called hard region in an expansion of the loop integrals in different regions [35]. In this region all loop momenta are of order Q, effectively leading to the massless case considered in Eqs. (2.16) and (3.7). This 'universality' implies, for instance, that Eq. (3.12) provides a prediction for the coefficient of ln(Q 2 /m 2 ) in the three-loop quantity F 3 for an Abelian gauge theory with fermion masses like Quantum Electrodynamics (QED) [36] after the usual identification of the colour factors. For QED, e.g., one has C F = 1, C A = 0 and T f = 1 instead of our QCD convention T f n f = n f /2. Another interesting implication of Eq. (3.7) arises for maximally supersymmetric Yang-Mills theory (MSYM), i.e., Yang-Mills theory with N = 4 supersymmetry in four dimensions. QCD results may be carried over to this theory using the inspired observation [37] that the MSYM results can be obtained from the contributions of leading transcendentality in QCD. This procedure has been applied to the QCD results for the three-loop anomalous dimensions of spin N of leadingtwist operators [17,18], which were employed to extract corresponding quantities in MSYM [37]. Strikingly enough, the resulting MSYM anomalous dimensions completely agree with predictions based on integrability for the planar three-loop contribution to the dilatation operator [38]. This agreement has been checked up to spin N = 8 in Ref. [39] and is now established up to N = 70 [40] (for a review see also Ref. [41]). Although no formal proof exists for the procedure of Ref. [37], it has recently been used in reverse, namely to predict terms of highest transcendentality in the QCD form factor. Based on studies of planar amplitudes in MSYM at three loops [19], where an interesting pattern of iteration for the four-point amplitude has been found, both the coefficients A l \| MSYM and the leading contribution to G l \| MSYM have been determined for l ≤ 3. Our new result for the three-loop form factor F 3 in Eq. (3.7) and for coefficient G 3 in Eq. (3.12) puts us in a position to check this part of Ref. [19] and thereby provide further evidence on the procedure of Ref. [37]. The only transcendental numbers entering the results for the form factor are the values ζ n of Riemann's zeta function. Hence the procedure of Ref. [37] implies that, as each order in α s , one keeps only the highest terms ζ n and ζ i ζ j with i + j = n. After the SYM identification C A = C F = n c (terms with n f do not contribute at the highest transcendentality), Eqs. Both relations agree with the results of Ref. [19], and hence with the prescription of Ref. [37]. Summary We have derived new higher-order QCD results for the electromagnetic form factor of on-shell massless quarks. Specifically, we have extracted all third-order 1/ε pole terms in dimensional regularization from our recent computation of the three-loop coefficient functions for inclusive deep-inelastic scattering [15], supplemented by a higher-ε extension of the two-loop contributions. These results, together with our extension of the resummation of the form factor to the next-tonext-to-leading contributions, fix the six highest 1/ε poles to all orders. As an example, we have provided the explicit expression for the coefficients of ε −8 . . . ε −3 at four loops. While the pole terms of the form factor alone are not sufficient for use in other three-loop calculations like e + e − → 2 jets, they do have immediate theoretical applications both for the infrared structure of higher-order QCD amplitudes and for other gauge theories such as QED and N = 4 Super-Yang-Mills theory, where our results confirm a recent corresponding calculation in Ref. [19]. Moreover, our present results are sufficient (up to a numerically irrelevant uncertainty due to the unknown four-loop cusp anomalous dimension) for extending the finite absolute ratio of the time-like and space-like form factors, which directly enters the description of the Drell-Yan process, to the fourth order in α s . We close by noting that the computation of the finite part of the three-loop quark form factor F 3 by an extension of the techniques employed in this article is feasible. Note that both the normalization in Eq. (2.13), where we have pulled out the factor (Q 2 /µ 2 ) −2ε , and our convention for ε are different from those in Ref. [6].	2014-10-01T00:00:00.000Z	2005-07-04T00:00:00.000	{ "year": 2005, "sha1": "0768f203b40fe581ac9ade2e8688c8db158aa65d", "oa_license": null, "oa_url": "http://arxiv.org/pdf/hep-ph/0507039", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "766e8f44843581ff56308ae23c042d0ed5db58ef", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
268345476	pes2o/s2orc	v3-fos-license	NAVIGATING THE COMMODITY MARKET: AN OVERVIEW FOR INVESTORS AND TRADERS The paper provides an in-depth overview of commodity markets, covering traded goods, key stakeholders, financial instruments such as swaps, options Authorities Regulating Regulatory agencies like the Commodity Futures Trading Commission (CFTC) and Financial Conduct Authority (FCA) are crucial in monitoring and controlling commodities markets to maintain equity and market integrity.Market regulators often work with commodity exchanges to ensure adherence to exchange policies.Understanding the responsibilities of market participants, commodity exchanges, and regulatory frameworks is crucial for efficient trading in the commodities market. REVIEW OF LITERATURE Gilbert (2010) aptly asserts the increasingly intertwined relationship between commodities and the stock market, a sentiment echoed by scholars such as Vivian & Wohar (2012), emphasizing the need to understand this dynamic for the benefit of financial market investors.Gorton and Rouwenhorst (2004) discovered a significant negative correlation between commodities and stock market indices due to differing business cycle behaviors.Understanding the correlation between the Indian stock market and commodities is crucial for both domestic and global investors, as highlighted by Stoll and Whaley (2010), given the rising inclusion of commodities in investment portfolios.Research by Hillier, Draper, and Faff (2006) emphasizes the advantageous performance of precious metals in high volatility markets.Lucey, Tully, and Poti (2006) identified a negative effect on gold and futures markets, particularly examining their seasonality.Tansuchat, McAleer, and Chang (2009) emphasized how the direction of stock prices depends on whether a company is an oil producer or consumer, affecting stock market performance accordingly.Alquist, Kilian, and Vigfusson (2011) focused on the predictability of oil prices concerning macroeconomic factors.Hamilton (2009) and Kilian (2009) attributed the increase in commodity prices to the rapid growth of emerging economies, noting minimal chances of a strong relationship between commodities and stock markets.Beck (1993) highlighted the strong, albeit short, relationship between commodity spot and futures prices, indicating potentially high returns.Maghayereh's (2003) study analyzed the long-term relationship between Jordanian stock prices and macroeconomic variables, revealing insights through integration analysis and monthly time series data from 1987 to 2000. RESEARCH METHODOLOGY The research for this study draws from a variety of secondary sources, such as online resources, websites, magazines, research articles from both national and international journals, as well as pertinent books and other references. Financial tools used in commodity trading 1. Contracts for Futures Futures contracts are standardized agreements for purchasing or selling a good at a fixed price, used for hedging against price swings, speculation, and price discovery.They are traded on commodities exchanges and offer a transparent method for determining market expectations. Options Contracts An option contract allows a holder to buy or sell a commodity at a predetermined price within a specific time frame.It serves as a risk management tool, a low-risk strategy for speculators, and a foundation for advanced trading techniques like spreads and straddles. 3. Swaps: Commodity swaps involve exchanging cash flows based on a commodity's price fluctuation, offering risk management, customization, and less volatility.Understanding these financial instruments is crucial for traders and investors to tailor strategies, control risk, and capitalize on market opportunities. Technical and Fundamental Analysis Trading commodities is a difficult undertaking that necessitates a thorough comprehension of price patterns and market dynamics.This section discusses the importance of technical and fundamental analysis in commodity trading and gives an outline of the methods used in each. The Significance of Technical and Fundamental Analysis Fundamental and technical analysis are essential strategies for traders and investors in the commodities market.Fundamental analysis examines underlying factors affecting prices, providing a comprehensive view for long-term investors and hedgers.Technical analysis uses past data to identify entry and exit points, trends, and patterns, benefiting speculators and short-term traders. Doing Fundamental Analysis Fundamental analysis is a systematic examination of various factors influencing commodity prices, including supply and demand analysis, economic indicators like GDP growth and inflation rates, government policies like trade agreements and subsidies, and geopolitical events like trade disputes or hostilities.It helps understand the overall economic state and its impact on commodity prices. Performing Technical Analysis Technical analysis predicts future price changes by analyzing previous data, chart patterns, and candlestick patterns.It aids commodity traders in making informed decisions, utilizing risk management techniques and resources to navigate the market. Hazards Associated with Trading Commodities Price risk is the most common risk in commodities trading, resulting from weather, geopolitical events, and supplydemand imbalances.Operational risk involves daily hazards in commerce and logistics, causing supply chain disruptions, financial losses, and reputational harm.Geopolitical risk involves uncertainty from political developments, such as trade disputes and wars. Risk Management Techniques and Resources Effective risk management is crucial in trading commodities.Techniques include diversification, hedging, and risk tolerance assessment.Diversification involves trading various commodities to spread risk, while hedging uses derivative contracts to protect against price changes.Risk tolerance assessment helps determine one's readiness and ability to take on risk, enabling appropriate trading strategies.Risk monitoring and analytics can help track exposure to hazards in real-time, while compliance with industry standards and regulations can reduce operational risks.Geopolitical analysis can help manage supply disruptions and create backup plans.Scenario analysis can determine potential market changes' impact on portfolios, enabling response plans.Insurance and risk transfer are crucial for preventing unforeseen events like weather damage or supply chain disruptions.Advanced risk management techniques and strategies are essential for traders and investors to protect their interests and make informed decisions in the commodities market. Strategies for Trading Commodity traders use various strategies to manage risk and capitalize on market opportunities.Trend Following involves following established patterns in commodity prices, using technical indicators to determine trend strength and entry/exit points.Mean Reversion involves predicting deviations from historical averages, often using statistical techniques to trade against the prevailing trend.Arbitrage involves exploiting pricing differences across marketplaces or commodities.Scalping is a short-term trading technique, aiming to profit from minimal market fluctuations.Day trading protects against overnight market risk by opening and closing positions within the same trading day.Swing trading focuses on intermediate-term market patterns. The Value of Trading Plans and Risk Tolerance Choosing the appropriate approach is only one aspect of successful trading; other factors include knowing your risk tolerance and following a planned trading plan: ➢ Risk Tolerance: The amount of risk that a trader is ready and able to take on is known as their risk tolerance.Knowing your level of risk tolerance is crucial before you trade since it affects how big of a position you can take and how you manage your risk.Higher risk-takers might employ more aggressive trading tactics, whereas lesser risk-takers might favor more cautious methods.➢ Trading Plan: Your trading objectives, techniques, risk management guidelines, and trading timetable are all detailed in your trading plan.It supports you in keeping your trading activity consistent and disciplined.A well-considered plan has rules for modifying strategies in response to shifting market conditions, as well as entry and exit criteria and risk-reward ratios.➢ Emotional Discipline: Adhering to your trading plan and properly managing risk require emotional discipline.Your approach may be undermined by spontaneous decisions resulting from emotional reactions.Stick to your plan and keep emotional control during trading.➢ A trader's approach to the commodity market is comprised of interconnected factors such as trading methods, trading plans, and risk tolerance.The key to long-term success in commodity trading is the proper strategy selection, paired with a strong risk management framework and strict adherence to a trading plan.Diversification of Portfolio and Commodities ➢ With their own risk-return characteristics and benefits for diversification, commodities can be a worthwhile addition to investment portfolios.This section will cover the benefits of diversifying a portfolio with commodities and how commodities may be included into investing portfolios. Integration of Commodities into Investment Portfolios There are several ways to include commodities into investment portfolios, and each one provides a unique level of exposure to the commodity asset class: ➢ Direct Investment: Investing in physical commodities is one option available to investors, as are funds dedicated to certain commodities or exchange-traded products (ETPs) that monitor the performance of particular commodities.Direct exposure to changes in commodity prices is offered by this strategy.➢ Commodity Futures: By purchasing commodity futures contracts, investors can access commodity markets without having to hold actual commodities.Commodity futures offer an opportunity to profit from both growing and declining prices.➢ Commodity Stocks: Purchasing stock in enterprises engaged in the production, exploration, or distribution of commodities such as mining, oil, or agricultural companies indirectly relates the profitability of an investment portfolio to the performance of commodities. Advantages of Using Commodities to Diversify a Portfolio Commodity-based investment portfolio diversification has a number of benefits.➢ Increased Diversification: There is little correlation between traditional asset classes like equities and bonds and commodities.By providing benefits of diversification, they can lower the total risk of a portfolio when introduced.Commodities may do well during difficult economic times, offsetting losses in other asset groups.➢ Inflation Hedging: Precious metals, such as gold, are particularly effective commodities to use as inflation hedges.Commodity prices may rise in response to rising inflation, protecting an investment portfolio's real value.➢ Portfolio Stability: By counteracting the volatility of stocks and bonds, commodities can improve portfolio stability.During times of financial instability, commodities can serve as a stabilizing factor.➢ High Return Potential: Commodity markets have the potential to see large price swings, which presents chances for profitable investments.In addition, rising global demand for resources and raw materials can be advantageous for commodities. ➢ Risk-Adjusted Returns: Better risk-adjusted returns could come from a diversified portfolio that holds commodities.Commodity investments can improve the overall risk-return profile of a portfolio because of their low correlation with other assets.➢ Alternative Investment Opportunities: Investors seeking alternatives to conventional equities and bonds may find that commodities investments provide exposure to real assets and international markets. Commodities can enhance diversification but also pose risks like supply and demand dynamics and price volatility.Hence, portfolio distribution should consider investor's time horizon, objectives, and risk tolerance.Including commodities in portfolios can reduce risk, diversify, and expose investors to asset classes with weak correlations to equities and bonds. The successful commodity trade is the gold trade.The trade type is long (buy) gold futures.➢ Geopolitical unpredictability: Investors frequently turn to gold as a safe-haven asset during periods of political unrest or economic unpredictability.➢ Inflation hedge: Demand for gold rises in tandem with expectations of inflation since gold is thought to be an inflation hedge.Negative or low interest rates increase the appeal of gold and other non-interest bearing assets.➢ Result: A lucrative trade as a result of increased gold prices The unsuccessful commodity trade is the oil trade.The trade type is long (buy) oil futures.The failure of the oil industry was attributed to an oversupply, a global economic slowdown, and a sharp decline in oil prices. REVIEW OF THE FACTORS Commodity exchanges depend on supply-demand equilibrium, influenced by geopolitical events, economic conditions, and weather.Oversupply can lead to profitable trades, while disruptions can occur.Factors like technological advancements, government policies, market sentiment, exchange rates, and environmental factors affect agricultural goods prices.Proficient traders evaluate these aspects and handle risk effectively.New developments in commodities trading and make some predictions about where the commodity markets may go. ➢ An Examination of Sustainability and ESG Factors:In commodity trading, environmental, social, and governance (ESG) factors are becoming more and more significant.Investors that prioritize social responsibility and environmental sustainability are being sought after by traders.ESG factors consist of: ➢ Renewable Energy: To encourage a move toward greener energy sources, there is growing interest in trading renewable energy certificates, such as carbon credits and RECs.➢ Sustainable Agriculture: Traders are concentrating on trading commodities made with environmentally friendly processes and sustainable farming practices.➢ Ethical Mining: Responsible and ethical sourcing of minerals and metals is a concern for commodity merchants.➢ Carbon Markets: Emissions-intensive industries are trading carbon permits, and the establishment of carbon markets-where carbon emissions can be bought and sold-is gathering momentum. Using technology and digitization In order to improve transparency, decrease fraud, and streamline trade procedures, the commodity trading sector is implementing cutting-edge technology like blockchain and artificial intelligence (AI).Blockchain technology and smart contracts are used to automate and transparently verify transactions. Volatility and Risk Management Rising geopolitical tensions and unpredictability in the world economy could keep influencing commodity price volatility.To successfully negotiate these obstacles, traders will require sophisticated risk management techniques.➢ Alternative Financial Products: As an alternative asset class, commodity trading continues to draw in investors looking for chances for hedging and diversification.➢ Impact of Geopolitics: Trade disputes and other geopolitical developments have the potential to upset supply chains and have an effect on commodity prices.This pattern is anticipated to continue. ➢ Impact of Climate Change: Commodity markets will be impacted by climate change's effects, which include extreme weather occurrences and altered farming practices.It's possible that insurance products linked to climate derivatives and weather patterns may proliferate.➢ Ecological Supply Networks: Sustainable supply chains are becoming more and more important to businesses, which will have an effect on commodity procurement and trade.➢ Modifications to Rules: New laws that governments are putting into place to address financial stability, environmental concerns, and transparency in commodities trade will have an impact on market dynamics.➢ Future Course of Commodity Markets: The following trends are expected to have a significant impact on commodity trade in the future: ➢ Transition to Sustainable Commodities: ESG-compliant commodities will become more popular, and trading and investment decisions will be influenced by sustainability considerations.➢ Greater Technological Integration: As trade becomes more efficient and transparent, risks can be managed and commodity flows can be tracked more easily thanks to digitalization and technology.➢ Renewable Energy: Trading in renewable energy commodities, such as solar and wind power, is anticipated to increase as the globe shifts to greener energy sources.➢ Global Uncertainty: The commodity markets will be impacted by geopolitical conflicts and environmental changes, which may result in price swings and supply disruptions.➢ Regulatory Environment: Changing rules, such as those pertaining to carbon pricing, will have a greater impact on commodity trading tactics.➢ Sustainable supply chains: Businesses will give priority to these, which will have an impact on sourcing and trading. In conclusion, a major emphasis on sustainability, technological integration, and adaptation to a shifting geopolitical and regulatory environment will characterize commodity trading in the future.To prosper in the changing economy, investors and commodity traders will need to understand these patterns.In summary, the ability to successfully traverse the complexities of the market, make well-informed judgments, and manage risks are critical skills in the realm of commodity trading.Important ideas and new developments that traders and investors should be aware of when they enter or continue their trip in the commodities market have been clarified by this overview.4. Take a Long-Term View: Although commodities can be included in a diverse investment portfolio, it's crucial to take a long-term view when investing in them.Although short-term volatility is typical, commodities have demonstrated the ability to yield strong returns over time. 5. Constant Learning: Since the commodities market is ever-changing, it's critical to stay current on news, trends, and innovations.To improve your abilities and expertise, think about pursuing additional education and training.6. Keep an Eye on World Events: Pay attention to changes in the environment, economy, and geopolitics.The price of commodities and trading possibilities may be significantly impacted by these occurrences.7. Environmental and Ethical Considerations: Sustainable development and moral business conduct are growing trends.These elements can affect market mood and the direction of commodities, so take them into account when making trading decisions. Promotion of Ongoing Education and Risk Awareness: Commodity trading offers lucrative opportunities but requires ongoing education and risk understanding.Successful traders navigate the market with confidence, embracing evolving trends and risk management.Embracing sustainability and prudence is crucial for long-term success in this lucrative venture. Keep Up with New Developments in Trends: By nature, commodities are erratic.Effective risk management techniques, like position sizing and diversification, are essential for safeguarding your cash.3. Accept how the commodity trading industry is changing, taking into account new rules, technology developments, sustainability, and ESG issues.Opportunities for success may arise from adjusting to these developments. 1. Important Takeaways: Know the Basics: It's critical to have a firm understanding of the mechanics of supply and demand, as well as the effects of geopolitics on the economy.2.Risk Control Is Essential:	2024-03-12T16:12:19.511Z	2024-03-06T00:00:00.000	{ "year": 2024, "sha1": "4e4c2716788c828dd8600a747c8398dd0e1787c8", "oa_license": null, "oa_url": "https://eprajournals.com/IJHS/article/12451/download", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "f66b93f7c773595f1e2d1e454fcea58f8df8ea54", "s2fieldsofstudy": [ "Business", "Economics", "Environmental Science", "Education" ], "extfieldsofstudy": [] }
253719632	pes2o/s2orc	v3-fos-license	An optimal reinsurance problem in the Cramér–Lundberg model In this article we consider the surplus process of an insurance company within the Cramér–Lundberg framework with the intention of controlling its performance by means of dynamic reinsurance. Our aim is to find a general dynamic reinsurance strategy that maximizes the expected discounted surplus level integrated over time. Using analytical methods we identify the value function as a particular solution to the associated Hamilton–Jacobi–Bellman equation. This approach leads to an implementable numerical method for approximating the value function and optimal reinsurance strategy. Furthermore we give some examples illustrating the applicability of this method for proportional and XL-reinsurance treaties. Introduction The determination of optimal insurance contracts is a classical topic in insurance mathematics. The first results are stated in a static utility theoretic framework and concern the relation between a risk facing individual and the insurer. The goal is the construction of an optimal insurance arrangement for the first party with a certain constraint stemming from the second party. Classical contributions in this context are Kenneth (1973), Raviv (1979) and Borch (1974), where one finds a collection of pioneering articles. A more recent paper by Guerra and Centeno (2008) studies this problem for exponential utility and provides the link to the maximization of the so-called adjustment coefficient which is the decay rate of the ruin probability for increasing initial capital. The idea of using reinsurance for maximizing the adjustment coefficient was introduced by Waters (1983), further studied by Centeno (1986Centeno ( , 2002 and Schmidli and Hald (2004), and can be considered as the motivation for studying optimal reinsurance. The first paper to study dynamic optimal reinsurance in the classical risk model for the minimization of the ruin probability is Schmidli (2001), who dealt with the case of proportional reinsurance treaties. This approach was extended to excess of loss contracts by Hipp and Vogt (2003). A general presentation on ruin probability minimization by means of reinsurance in the classical and diffusion risk model can be found in Schmidli (2008). Furthermore, this reference provides some asymptotic studies of the behaviour of optimal strategies, which in certain situations coincide with the ones maximizing the adjustment coefficient. Some additional results with a focus on non-proportional reinsurance contracts are given in Hipp and Taksar (2010). Using a different criterion to assess the performance of an insurance portfolio, Eisenberg (2010) thoroughly covers a variety of capital injection minimization problems under both the classical risk model and its diffusion approximation where the insurer has the possibility to dynamically reinsure its risk. The incorporation of dynamic reinsurance to the classical problem of maximizing the dividend payouts of an insurance company prior to ruin in a compound Poisson framework was treated by Azcue and Muler (2005) for general reinsurance schemes and by Mnif and Sulem (2005) for excess of loss reinsurance. In a diffusion setting, the corresponding problem was studied by Højgaard and Taksar (1999) in the case of proportional reinsurance. Combining dividend pay-outs maximization with proportional risk exposure reduction, Schäl (1998) formulated a piecewise deterministic Markov model where only jumps but not the deterministic flow can be controlled. In contrast to the aforementioned references which deal with optimal reinsurance for continuous time risk processes, Schäl (2004) investigates a discrete time insurance model controlled by reinsurance and investments in a financial market with the intention to either maximize the expected exponential utility or minimize the ruin probability. An analogous problem was treated by Irgens and Paulsen (2004), where the authors examine the purpose of maximizing the expected utility of terminal wealth by use of optimal investment and reinsurance. Finally, we would like to mention a new approach linking ruin theoretical concepts with the framework of worst-case optimization theory explored by Korn et al. (2012). Embedded in a differential game setup, the authors applied a worst-case scenario approach to maximize the expected utility of the surplus of an insurance company at some given deterministic terminal time by dynamic proportional reinsurance. In this contribution, we will study the use of dynamic reinsurance for maximizing a particular economic performance measure which for a diffusion risk model was introduced by Højgaard and Taksar (1998a, b). For its definition, let X u = (X u t ) t≥0 be a surplus process comprising a reinsurance strategy u. The performance measure of this particular strategy is defined by where δ > 0 denotes a discount or preference rate and τ u is the time of ruin of X u . In Taksar (2000) this measure is motivated by the following arguments: the surplus of the insurance company is kept on a bank account and interest gains are immediately distributed as dividends, thus maximizing expected discounted dividend payments is equivalent to maximizing (1). Another way to motivate this value function in a Markovian environment is to introduce a random life time S ∼ E x p(δ) which is independent of all other model ingredients. Then one observes which tells that the performance measure is proportional to the expected surplus at a random exponential time S. This means that a dynamic reinsurance strategy is used for maximizing the surplus at some exogenous point in time. Cost functions of the form (1), or more generally involving a running costs function l(X t ), are also studied by Cai et al. (2009) in an uncontrolled piecewise-deterministic compound Poisson environment. The structure of the manuscript is as follows. In Sect. 2, we give a precise mathematical formulation of the problem, introducing the controlled surplus process and the value function. The analytical characterization of the value function is presented in Sect. 3. It starts with a collection of basic properties and employs the dynamic programming approach for achieving a final statement. Section 4 includes some comments on the numerical procedure obtained from the analytical results and two illustrative examples. Finally, a conclusion is stated in Sect. 5. Problem statement In the sequel, we will always work on a probability space (Ω, F, P) which carries all stochastic quantities to be defined in the following. In the Cramér-Lundberg model (also known as compound Poisson model or classical risk model), the surplus process X = (X t ) t≥0 of a homogeneous insurance portfolio is modeled as Starting with an initial deterministic surplus X 0 = x ≥ 0, the surplus process increases linearly due to premiums that are collected continuously over time at a constant rate c > 0. On the other hand, it decreases due to claims happening at the arrival times of a homogeneous Poisson process N = (N t ) t≥0 with intensity λ > 0. The claims {Y i } i∈N constitute a sequence of positive independently and identically distributed random variables with a density function f Y (·) and finite mean μ. Later on we will use Y as a representative random variable from this distribution. In addition, the sequence {Y i } i∈N and N are assumed to be independent. The flow of information is given by the filtration {F t } t≥0 which is generated by the surplus process X . In the remainder of the manuscript, we will use the symbol E for the expectation with respect to the probability measure P, for the conditional expectation E(· \| X 0 = x) we will use the expression E x . Fundamental quantities in this framework are the time of ruin and the probability of ruin for initial capital x ≥ 0. In some of the proofs below we will compare pathwise, i.e., we fix an ω ∈ Ω, processes starting at different initial values x and y. Therefore it will be necessary to add the initial value in the definition of the time of ruin, for example E x (X τ y ) denotes the expected value of the surplus started at x stopped at the time of ruin as if the surplus would have started in y (x > y) (thinking along the same path). Certainly, we have, using θ = inf{t ≥ 0 \| X t < x − y}, E x (X τ y ) = E x (X θ ), but we believe that out of the context our notation will be more intuitive. It is well known, that for avoiding almost sure ruin, it is necessary to choose a premium intensity fulfilling the net-profit condition c > λμ. Therefore, based on the expected value premium principle we set c = (1 + η)λμ with a safety loading η > 0. For further details on classical problems in risk theory and related topics we refer to Asmussen and Albrecher (2010). Assume now that in order to reduce the risk exposure of the portfolio, the insurer (cedent) has the possibility to take reinsurance in a dynamic way. Namely, at each time t, the insurer transfers a portion of the premium income to a reinsurer, who in turn commits to cover a part of the occurred claims. The dynamic reinsurance setup we are going to use follows the presentation from Schmidli (2008). Formally, a reinsurance scheme is given by a monotone increasing function r : [0, ∞) → [0, ∞) which fulfills 0 ≤ r (y) ≤ y. Then r is the retention function with the meaning that for a claim of size Y , the amount r (Y ) is paid by the insurer and Y − r (Y ) is taken by the reinsurer. For introducing a control possibility a family of available schemata R is parameterized by a control parameter u from a compact set U . This means that for u ∈ U the chosen reinsurance contract is given by r (·, u) ∈ R, where r : [0, ∞) × U → R + with 0 ≤ r (y, u) ≤ y. In addition we assume that r (y, u) is continuous in both arguments. After fixing the family R, the set of available reinsurance schemes is given by R = {r (·, u) ∈ R \|u ∈ U , 0 ≤ r (y, u) ≤ y, r continuous, and increasing in y}. For later use we denote by ρ(y, u) the generalized inverse of r (y, u) in the y−component, which due to monotonicity exists. Naturally, when employing reinsurance there are premiums to be paid. We assume that the reinsurer uses a deterministic premium function π : L 1 (Ω, P) → [0, ∞), such that when fixing u ∈ U the premium is based on π(Y − r (Y, u)). From an aggregated risk perspective, if the insurer chooses reinsurance u ∈ U at time t, the premium at rate λ π(Y − r (Y, u)) is paid to the reinsurer. Consequently the premium income of the insurer reduces to c(u) = c − λ π(Y − r (Y, u)). In the sequel, we shall always assume that c(u) is continuous and that full reinsurance leads to a negative premium income, i.e., c < λπ(Y ). The premium function π may be based on the expected value principle, where θ > η denotes the safety loading of the reinsurer, or on the variance principle, Possible concrete choices for R and U are the classical situations of proportional reinsurance and excess-of-loss reinsurance. In the first case we have r (y, u) = uy and u ∈ U = [0, 1], in the second case r (y, u) = min(y, u) and u ∈ U = [0, ∞]. Notice, that in the latter case, an infinite retention level is equivalent to no reinsurance. In the following we will restrict the set of control parameters to the set U = {u ∈ U \| c(u) ≥ 0} for avoiding a negative premium rate. Since U is supposed to be compact and c(·) is continuous we have that U is compact. Remark 1 The idea of a dynamic reinsurance strategy can be explained as follows. At each time instant t, the insurer chooses a control parameter u = u t ∈ U which specifies a reinsurance scheme r (·, u) from an available set of schemes. The choice of u simultaneously determines the extent to which the insurer wants to reduce its risk exposure and the additional cost this protection incurs, taking the form of a reinsurance premium. Namely, if a claim occurs at time t, the insurer pays r (Y, u t ) and the reinsurer pays the rest, i.e. Y − r (Y, u t ). In exchange of this risk transfer, the insurer pays to the reinsurer a reinsurance premium at a rate λπ (Y − r (Y, u t )). Let u = (u t ) t≥0 be a U-valued stochastic process which is {F t } t≥0 previsible and called a reinsurance strategy. Then the dynamics of the controlled surplus process X u = (X u t ) t≥0 are described by Remark 2 From Rogers and Williams (1994, p.182) we can deduce that the previsibility of u induces the fact that it is progressively measurable and thus also measurable as a function in time. Since the premium rate c(·) is assumed to be continuous and bounded by c, the integral t 0 c(u s ) ds exists at least in the Lebesgue sense. Because jumps of the process X u occur according to the fundamental Poisson process and behaves continuously between jump times, the process X u is right continuous with existing limits from the left, i.e., cádlág. Consequently, X u is progressively measurable as well and for fixed ω, X u (ω) is measurable in t. Again, integrals of the form t 0 X s ds certainly do exist in the Lebesgue sense. The time of ruin τ u x denotes the time the controlled surplus process X u first becomes negative, From now one we call a stochastic process u = {u t } t≥0 admissible reinsurance strategy if it fulfills all the previously made assumptions. In this context the previsibility is crucial. That is, at claim time T i , the reinsurance parameter is chosen based on the information up to time T i −. The previsibility of the reinsurance strategy is a natural assumption in this setting, otherwise the insurer could change the reinsurance parameter to full reinsurance at the claim occurrence time. The reinsurer would then pay all claims while all premiums would be collected by the insurer. Let U denote the set of admissible reinsurance strategies. Associated to an admissible reinsurance strategy u and an initial reserve x ≥ 0, we define its performance criterion as the expected cumulative discounted surplus process until ruin, with δ > 0 a discount or preference rate. In the sequel, we will refer to V u (x) as the return function. The optimization problem then consists of finding the optimal return function, or value function, defined as and an optimal admissible reinsurance strategy u leading to the value function, i.e. a strategy which delivers the maximal return function (5). Main results In this section, we first derive some elementary bounds, which allow for a rough characterization of the value function. In a next step, we are able to prove the existence of a solution to an integro-differential equation which is closely related to the problem's Hamilton-Jacobi-Bellman equation. Finally, a verification argument provides the bridge between these analytical results and the stochastic optimization problem of interest. Some elementary bounds Proposition 1 For x ≥ 0, the value function V (x) admits the following bounds: Taking the supremum over all admissible strategies u shows that the value function V (x) satisfies inequality (a). It remains now to validate inequality (b). The choice of the admissible strategy u 0 which corresponds to buying continuously full reinsurance until the time of ruin leads to a deterministic reserve X u 0 with negative drift. As a consequence, the time of ruin τ u 0 x can be explicitly computed, that is, The following result presents bounds on increments of the value function and also provides its continuity. Proposition 2 For x > y ≥ 0, the value function satisfies: For given x > 0 and given > 0, consider an admissible -optimal strategy u such that Since u is also admissible for initial capital y with x > y ≥ 0 (up to time τ u y ), we have where E x , E y indicate the starting value of the corresponding process. Now we are going to use a pathwise argument, let Notice that on E the paths (for fixed ω) of the reserves started in x and y move parallel with a distance x − y > 0 and get ruined at the same point in time. Therefore, we can rewrite the above inequality in the following way, The first inequality is just a restatement of (6). It incorporates the fact that the two values, the values of the strategy u for surplus processes started in x and y, only differ on E c . This difference is given by the third expectation, in which E x indicates that the surplus within the integral is started at x. The second inequality follows from the observation that The last inequality uses X u τ u y ≤ x − y for the reserve started in x and that consequently the corresponding expectation is smaller than Let us now prove inequality (b). Let y ≥ 0 and > 0 be given, consider an admissible strategyū such that Vū(y) + ≥ V (y). For x > y, we have, Again, let E = {τū x = τū y } and let T 1 be the time of the first claim occurrence. We can write From the arbitrariness of > 0, we get the result. Additionally, we can derive the following. Lemma 1 The value function V is locally Lipschitz continuous. Proof For given x > 0 and > 0, consider an admissible strategy u = (u x t ) t≥0 such that Finally, after explicitly evaluating the last estimate we derive for x > y ≥ 0, This implies that V is locally Lipschitz continuous. Finally, we can summarize the following elementary properties of the value function V (x). Notice that absolute continuity follows from the local Lipschitz continuity mentioned in the previous Lemma. Corollary 1 The value function V is strictly positive, linearly bounded, monotone increasing and absolutely continuous. Remark 3 Suppose we assume in the proof of part (a) of Proposition 2, that for all u ∈ U the random variable r (Y, u) admits a bounded density f u r . Then, we can formally derive we get from (7) that the value function is globally Lipschitz continuous. For example, this case appears when dealing with proportional reinsurance. For further investigations, we need to improve on the lower bound from Proposition 1. When dealing with a contraction operator later on, the refined bound will allow us to describe the growth behaviour of the value function in a more precise way. We start with showing that for is the infinitesimal generator of the uncontrolled process X . For that purpose, we define which can be rewritten as Lemma 2 The value function V is bounded from below by Proof Since g(x) is differentiable we can apply Dynkin's formula and get From above, we already know that Lg(X s ) − δg(X s ) ≥ −X s + c−λμ δ , using this estimate, we arrive at, where T 1 denotes the time of the first claim occurrence. Using linear boundedness of g(X t∧τ ) in t and monotone convergence, we arrive at From its definition, we get Remark 4 Characterization of the value function Based on the elementary properties of the value function which are collected in Corollary 1, we can work out the dynamic programming approach for solving the optimization problem. We start with observing that V fulfills the dynamic programming principle, that is, for every F t -adapted stopping time S ≥ 0 the following relation is valid: The proof of this fact is mainly based on the continuity of V and follows standard arguments from the corresponding literature, see for instance the proof of Azcue and Muler (2014, Prop.2.3). The following Lemma shows that at least in some weak sense V fulfills the associated Hamilton-Jacobi-Bellman equation. (5) is a.e. a solution to: Lemma 3 The value function V defined in Proof In a first step we show that (10) is smaller equal to zero. Fix x > 0, h > 0 and let u ∈ U. Defineũ = (u t ) t≥0 such that u t = u for t ∈ [0, h] and u t =ũ t−h for t > 0 for someũ ∈ U. If necessary, we choose h small enough such that x + c(u)h > 0. Let T 1 denote the time of the first claim occurrence and set S = min{T 1 , h}. Then, (9) yields Since u is a constant control which applies on the time horizon [0, S] we can apply Rolski et al. (1999, Th.11.2.2) and get that V ∈ D(A u ), i.e., V lies in the domain of the generator. In the present situation the generator A u of the constantly controlled process X u is given by The particular result from Rolski et al. (1999, Th.11.2.2) applies, because the map t → V (x + c(u)t) is absolutely continuous, the so-called active boundary is empty and the bounds from Proposition 1 and Proposition 2 guarantee the asked for integrability condition. Therefore we can apply Dynkin's formula, identifying V with the measurable density of V , and can rewrite (11) to After regrouping and division by h we have The integral in the first expectation can be interpreted in the Riemann sense, V is continuous, such that sending h → 0 leads to The second limitation procedure needs a bit more care since the integrands as functions in t are only measurable and the respective integral is interpreted in the Lebesgue sense. For this purpose consider where in the second equality we used Lebesgue's Differentiation Theorem from Wheeden and Zygmund (1977, Th.7.16) which applies since the measurable density V certainly is locally integrable in the Lebesgue sense because of the bounds on the function V and its increments. One may notice that since the ds integrand equals zero for s = 0. The choice of the control parameter u ∈ U was arbitrary, such that we have We can turn to the second step, showing that (10) is also larger or equal to zero. Set again S = min{T 1 , h} for some h > 0 and let the strategy u 1 = (u 1 t ) t≥0 be h 2 −optimal for the right hand side of (9), that is where we added the term h with some arbitrary ε > 0 for achieving strict positivity. In the above equation we can use T 1 ∼ Exp(λ) and regroup a little bit to arrive at We kept E x since u 1 is still stochastic on the time interval under consideration. In the following we divide A, B, C, D by h and study the limits as h tends to zero -for interchanging limitation and expectation we will repeatedly make use of the dominated convergence Theorem. We start with discussing B: which follows from continuity of V . Next we deal with C: which is derived by an application of Wheeden and Zygmund (1977, Th.7.16). For part D we exploit a similar procedure together with the absolute continuity of V , Part A is resolved in the same way and delivers Finally we arrive at which concludes the proof since ε was arbitrary. At this point, we know that the value function is in some sense a solution to the associated HJB-equation. What remains to be done for a complete analytical characterization is a complement on uniqueness. For accomplishing such a result we are going to rewrite (10) in a way similar Schmidli as (2008, p. 47) did, when transforming equation (2.14) into (2.15). Suppose x is meaningful in the sense that V (x) exists. Since the set U is compact and all corresponding terms are continuous in u, a maximizer u(x) exists such that the supremum equal to zero is attained. Replacing the sup u by u(x) in (10) we have from which we can observe, using the lower bound (8) on V (x), that c(u(x))V (x) > 0 ⇒ c(u(x)) > 0. Hence, in the supremum we can replace the set U by the set U = {u ∈ U \| c(u) > 0}. Since V (x) is monotone, we can rewrite (10) into the equivalent form: Formally, we know that a.e. V (x) is a solution to (13). In addition, for x such that V (x) exists, we have the following, where , which can be used in (14), leading to Reinspecting (12) gives a positive lower bound on c(u(x)), where the last inequality is due to Lemma 2. Together with (15) we have As a consequence, we can redefine the crucial set for taking the supremum (resp. inf) U = {u ∈ U \| c(u) ≥ L}. One may notice that in (13) the infimum is taken again over a compact set and that the denominator is uniformly bounded away from zero. The first step towards a unique characterization of the value function is given in the following theorem the proof of which relies on the fixed point property of a certain operator (inspired by a similar approach used in Muler (2005, 2014)). Theorem 1 Let f (0) > 0 be some given initial value, then there exists a unique a.e. differentiable solution to Proof Let x 0 ≥ 0 and a continuous function f : [0, x 0 ] → R be given. Fix h > 0 and set is defined on C and x ∈ [x 0 , x 0 + h] and clearly T g ∈ C. Since for all s ∈ [x 0 , x 0 + h] all terms involving u are continuous in it and the infimum is taken over a compact set, we know that a minimizer u(s) exists. Now let g 1 , g 2 ∈ C and u 1 (s), u 2 (s) be the corresponding minimizers, we get Interchanging the roles of g 1 and g 2 and choosing h = L 2(δ+2λ) we get, such that T is a contraction on C and that consequently an unique fixed point of it exists. Since h and the contraction factor do not depend on x 0 , we can iterate this procedure on the intervals [0, h], [h, 2h], . . .. Finally, we observe that these fixed points, on the end points of the intervals [k h, (k + 1) h] continuously pasted, induce an unique solution to (13) with given initial value f (0). By construction, this solution is absolutely continuous on R + , since one may alter the grid for the construction procedure. We are now able to finalize the analytical characterization of V . Theorem 2 Suppose g : R → R with g(x) = 0 for x < 0 is linearly bounded by x δ + c δ 2 and an absolutely continuous solution to (13), then g(x) = V (x). The optimal strategy u * = (u * t ) t≥0 is induced by the pointwise minimizer u(x) of (13) such that u * t = u(X u * t− ). Remark 5 One can use verbatim the proof from Schmidli (2008, Lem.2.12) to show that the function u defining the optimal strategy is measurable. Consequently the process (u * t ) t≥0 is previsible and constitutes an admissible strategy. Proof Let t > 0 and u = (u t ) t≥0 ∈ U, since the paths of (X u t ) t≥0 are of bounded variation, we can use the Stieltjes integral to obtain The process M = (M t ) t≥0 defined by is a zero-mean martingale, due to compensation. Therefore, taking expectations in (16) leads to Remember that for g (X u s ) we have (at least a.e.) which yields for the particular control parameter u s , From Schmidli (2008, Lem.2.9), we know that either ruin occurs or the controlled surplus tends (linearly bounded) to infinity. Therefore, using bounded convergence in (17) results in . One observes that in (17) we have equality for the strategy u * , defined in the statement of the theorem, such that finally V (x) = g(x). The combination of the statement of the last theorem with the uniqueness result and the properties of the value function enables us to state a complete characterization. Corollary 2 The value function V is the unique solution to (10) in the set of absolutely continuous function g : R → R with g(x) = 0 for x < 0 which are bounded by x δ + c δ 2 . In particular just the initial value V (0) for equation (13) allows for a solution g(x) with the property lim x→∞ Numerical examples In this section, we will illustrate the theoretical results and sketch a numerical solution method by means of two examples. Furthermore, for the particular case of proportional reinsurance and a reinsurer using the expected value premium principle, we can refine the analytical results and state the asymptotic behaviour of the optimal strategy as the initial capital tends to infinity. Since an explicit solution to (10) is unfortunately out of reach, for deriving a solution one needs to rely on a numerical method. Luckily, the theoretical characterization stated in Theorem 2 and Corollary 2 constitutes an implementable procedure. These results tell that an iterated application of the operator T , defined in the proof of Theorem 1, on some linear function g(x) = x δ + g 0 leads to an approximation of the value function if and only if g 0 = V (0) is correctly chosen, cf. Corollary 2. Consequently, the first step in the procedure asks for a good guess of g 0 , which can (and needs) to be improved in later steps. For determining a meaningful approximation of g 0 , we exploit the idea of policy improvement, see for instance Bäuerle and Rieder (2011). The starting point is the value V sr (x) corresponding to the situation of no reinsurance, which in our parameter setting can be explicitly determined. Based on this value V sr , we compute a strategy u 1 = {u 1 t } with u 1 t = u 1 (X u t ) from the HJB-equation (10) via In a next step we determine a good approximation for V u 1 (0), which can be done by using the Monte-Carlo method with direct simulations of the controlled surplus process from (4). Now we know that V u 1 (0) corresponds to an admissible strategy but does not necessarily equal V (0). But with V u 1 (0) at hand we can determine V u 1 (x) for x ≥ 0 either by an iteration of an operator, similar to T but without the infimum in its definition, or by a finite-difference method. We use this value V u 1 as the starting point of iterations of T . After a number of iterations, one can improve the initial value again by using the same method as illustrated above, but with the function obtained from the iterations as basis for the policy improvement step. This newly obtained value V u 2 then serves as the basis for new iterations of T . Remark 6 Alternatively, one can execute a policy iteration procedure on the basis of the original HJB-equation (10). Our experience showed that the obtained strategies are very close to the ones determined via the first method. Unfortunately, the quality of the simultaneously generated return functions is not always trustworthy, a fact which originates from the presence of the control parameter in front of the sensitive derivative term and inside the integral. Nevertheless, the use of these strategies allows for a considerable acceleration of the whole procedure. In this way we create, by the use of policy iterations at intermediate steps, an increasing sequence of initial values and also determine candidates for a fixed point of T . To decide whether an initial value is significantly too small one can check the behaviour of the function obtained from the corresponding iterations of T . If an initial value is far away from V (0) we observe a violation of the lower bound from Lemma 2 for relatively small values of x. We can accept an initial value V * (0) as a good guess for V (0) if the function V * obtained from iterations stays within the theoretically given bounds. If additionally V * matches the value of the implicitly given strategy, we can accept it as an valid approximation of the value function. Remark 7 Instead of starting the iteration procedure always at predetermined values V u , we can also start with g(x) = x δ + V u (0) and all previously stated arguments still apply. Our experience showed that this procedure leads to trustworthy results and representative illustrations of our theoretical findings. Certainly, a theoretical numerical analysis would be necessary and highly interesting but this is out of the scope of this publication. Example: proportional reinsurance In the following, we are going to use the model parameters given by: Gamma(2, γ ) distributed claim amounts. The insurer's premium rate is determined via the expected value principle and reads as c = (1+η)λμ with μ = 2 γ and η > 0. For the reinsurer, we assume the same premium principle but with a safety loading θ > η. The concrete numbers are given in Table 1. The considered reinsurance schema is r (y, u) = u y for a control parameter u ∈ (u, 1] with u = inf{u ∈ [0, 1] \| c(u) > 0}, as discussed before the statement of Theorem 1. For deriving numerical approximations to the value function and to the optimal strategy, we implemented the program we have illustrated in the introduction to this section. In contrast to the case of excess of loss reinsurance, the proportional situation turned out to be numerically demanding, requiring lots of computational efforts for arriving at passably satisfying results. The strategy obtained from 20 policy iterations steps, starting from V sr , is depicted in Fig. 1. In the remark following below, the shape of this strategy is discussed in some detail. Figure 2 contains the graphs of V sr (dotted line), V 1 (dashed line) and V 20 (full line). V 1 is computed from 30 iterations of T starting with g and an initial value g 0 = 212 corresponding to the strategy obtained from 1 policy improvement step based on V sr . The function V 20 is derived from 30 operator iterations, but using the initial value g 0 = 226.436 associated to the strategy from Fig. 1. In Table 2, we present some exemplary function values from the iterations of T towards the computation of V 20 . Remark 8 We would like to discuss lim x→∞ u * (x), which by the numerical computations is suggested to be one. Here, we exclusively deal with the case of proportional reinsurance and the expected value premium principle for both insurer and reinsurer, c(u) = λμ(u(1 + θ) − (θ − η)) for safety loadings θ > η. From the definition of the value function, we have Above, we introduced the martingale M = (M t ) t≥0 which is the compensated compound Poisson process: Now, we can regard τ 0 e −δt M t dt pathwise as a Stieltjes integral and apply integration by parts, Wheeden and Zygmund (1977, Th.2.21), to arrive at In (18), the integral with respect to the martingale is itself a martingale, leading to the second equality. At the same time, using an ε−optimal strategy u * for initial capital x > 0, we have If we suppose that u * is a Markov control, then we certainly have that M * t = t 0 λμu * s ds − N t k=1 u * T k Y k is a zero mean F X t martingale and the same integration by parts procedure as before applies. Consequently, we have for x large such that τ and τ u * (M t is linearly bounded) are tending almost surely to infinity that: Now, we proceed with determining lim x→∞ u * (x). Here, u * (x) denotes the pointwise maximizer in u of the HJB-equation (10), which due to continuity exists. Plugging in c(u) = λμ(u(1 + θ) − (θ − η)) and regrouping, we see that yd F Y (y)+(δ + λ) ηλμ δ 2 − λ 2 μη δ 2 F(x/u * (x)) λμ δ (1 + θ) . If we now assume that lim x→∞ u * (x) = u * exists, it should fulfill u * ≈ θ + u * 1 + θ , which can be fulfilled only if u * = 1. The two plots in Figs. 3 and 4 illustrate the sharp linear upper bound together with V (x) and f (x) = E x τ 0 e −δt X t dt for exponentially ν distributed claims and the following set of parameters given in Table 3. Example: XL-reinsurance As a second example, we consider the case of dynamic XL-reinsurance with E x p(ν) distributed claim amounts. The particular numbers chosen are close to the ones chosen by Hipp and Vogt (2003) and can be found in Table 4. The numerically determined approximative optimal strategy is displayed in Fig. 5. The corresponding value function's numerical approximation (full line) is shown in Remark 9 (Comparison with ruin probability minimization) When numerically determining the approximative optimal strategies, one observes some similarities but also differences to the situation of optimal dynamic reinsurance strategies for minimizing ruin probabilities, see Schmidli (2008, Ch. 2.3.1) and Hipp and Vogt (2003). In both situations, proportional and XL, the behaviour for small initial capital is similar, one finds that for some x 0 > 0 on [0, x 0 ], it is optimal to take no reinsurance. From that point on, a certain amount of reinsurance is bought. For larger x, the reinsurance choice is either returning to the no reinsurance case (proportional) or converging towards a constant level (XL). Here, the proportional case is in contrast to the situation when minimizing the ruin probability. There, for small claims the optimal reinsurance choice converges to a finite value as x tends to infinity. This different behaviour may be explained by the underlying performance measure which in the present framework is profit orientated. Because of discounting, a ruin event late in time does not bother the insurer which implies that above a certain surplus level (large enough for having early ruin just with a low probability) one is focusing on the maximal drift and not buying reinsurance. The question: " why does the numerically optimal XL strategy behave differently?" is interesting as a future research project on its own. The answer to this question may be based on the comparison of solutions to integro-differential equations. Conclusion In this paper, we studied a dynamic optimal reinsurance problem which is derived from an economical valuation criterion in risk theory. An interplay between analytical and probabilistic arguments allowed us to characterize the associated value function and finally the theoretical results were complemented by numerical examples. Based on the alternative interpretation of the studied value function, which is given in (2), we can state, that our results suggest that reinsurance can accelerate the process of building up a free reserve and that the use of reinsurance is beneficial in the economical context.	2022-11-21T14:57:38.958Z	2016-09-24T00:00:00.000	{ "year": 2016, "sha1": "3a3bb065d203a3ad0b9a6755853d33c6dd7b7630", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/s00186-016-0559-8.pdf", "oa_status": "HYBRID", "pdf_src": "SpringerNature", "pdf_hash": "3a3bb065d203a3ad0b9a6755853d33c6dd7b7630", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [] }
119394678	pes2o/s2orc	v3-fos-license	Enabling Narrow(est) IWA Coronagraphy with STIS BAR5 and BAR10 Occulters The Space Telescope Imaging Spectrograph's (STIS) BAR5 coronagraphic occulter was designed to provide high-contrast, visible-light, imaging in close (>= 0.15") angular proximity to bright point-sources. This is the smallest inner working angle (IWA) possible with HST's suite of coronagraphically augmented instruments through its mission lifetime. The STIS BAR5 image plane occulter, however, was damaged (bent and deformed) pre-launch and had not been enabled for GO science use following the installation of the instrument in 1997, during HST servicing mission SM2. With the success of the HST GO 12923 program, discussed herein, we explored and verified the functionality and utility of the BAR5 occulter. Thus, despite its physical damage, with updates to the knowledge of the aperture mask metrology and target pointing requirements, a robust determination of achievable raw and PSF-subtracted stellocentric image contrasts and fidelity was conducted. We also investigated, and herein report on, the use of the BAR10 rounded corners as narrow-angle occulters and compare IWA vs. contrast performance for the BAR5, BAR10, and Wedge occulters. With that, we provide recommendations for the most efficacious BAR5 and BAR10 use on-orbit in support of GO science. With color-matched PSF-template subtracted coronagraphy, inclusive of a small (+/- 1/4 pixel) 3-point cross-bar dithering strategy we recommend, we find BAR5 can deliver effective ~ 0.2"IWA image contrast of ~ 4 x 10^-5 pixel^-1 to ~ 1 x 10^-8 pixel^-1 at 2". With the pointing updates (to the PDB SIAF.dat file and/or implemented through APT) that we identified, and with observing strategies we explored, we recommend the use of STIS BAR5 coronagraphy as a fully"supported"capability for unique GO science. Introduction Throughout the course of its more than 27-year long mission, the Hubble Space Telescope (HST) has hosted five instruments with coronagraphic optics for high contrast imaging. Spherical aberration in the telescope primary mirror, unrecognized pre-launch, however, rendered unusable the first two coronagraphs flown, in the first-generation instrument suite with the Faint Object Camera (FOC; f/288 channel) and the Wide-Field/Planetary Camera-1 (WF/PC; PC 8 channel). While Hubble's aberrated vision was corrected individually for each instrument in Servicing Mission 1, this was not the case in the FOC f/288 channel that implemented its coronagraph. The WFPC-2 replacement instrument eliminated its predecessor's coronagraphic "Baum spot", thereby delaying the onset of HST coronagraphic science until the installation of the secondgeneration instruments in 1997 with the Near-Infrared Camera and Multi-Object Spectrometer (NICMOS) and the Space Telescope Imaging Spectrograph (STIS). NICMOS provided near-IR coronagraphic imaging from 1.1 to 2.4 µm with a symmetrical inner working angle (IWA) of r = 0.3" until its shutdown in 2009. STIS included a deployable coronagraphic focal plane mask with four hard-edge occulters (see Fig. 1). STIS provides broadband visible-light coronagraphy with: (a) two long, orthogonally oriented wedge-shaped occulters (WEDGE-A and WEDGE-B), each tapering to a minimum half-width of ~ 0.3". (b) a 3"-wide rectangular bar-shaped occulter (BAR10) extending 10" from one edge of its 52" x 52" FOV. (c) a, planned, narrow IWA bar (or "finger"-shaped) occulter (BAR5), 5" in length (orthogonal to, but non-intersecting with BAR10), with a half-width of only 0.15". The STIS BAR5 occulter was going to provide, at visiblewavelengths, an IWA (in one dimension) half the size of NICMOS's coronagraph. Unfortunately, BAR5 was damaged pre-launch (bent and deformed) without possibility of repair or replacement. Thus, BAR5 was not commissioned with the other three STIS coronagraphic occulters on-orbit, and was consequentially not used for subsequent science operations. The Advanced Camera for Surveys (ACS) was installed in 2004 with a high-performance visible-light coronagraph (well apodizing the HST diffraction spikes, unlike STIS), but with two larger angle (r = 0.9" and r = 1.8") occulters only. The unavailability of the STIS BAR5 occulter to GO programs, however, precluded the conduction of coronagraphic science investigations demanding the smallest IWAs. In GO 12923, despite its physical damage, we explored and confirmed the utility of STIS BAR5 coronagraphy leading to its much-delayed enablement and commissioning for GO science as a supported capability. Herein we discuss the GO 12923 program and its findings leading to the restoration and contrast calibration of STIS BAR5, and BAR10-corner ( § 14), coronagraphy. GO 12923 -Observation Plan The HST/GO 12923 program was designed, with an economy of orbits (6), to explore and validate the utility of the STIS BAR 10 rounded-corner and BAR5 ("bent finger") occulters for narrow-angle coronagraphy, despite BAR5 having sustained pre-launch damage and physical deformation. Primary concerns (which were put to rest) included the possibilities of afocality of the "finger" after bending leading to contrast loss, and/or introduction of edge-defects that could induce instrumental scattering or other optical artifacts in the unocculted field beyond the finger edges. Additional concerns for robust narrow-angle coronagraphy resulting from target acquisition (placement) imprecision with a (pre-execution) highly-uncertain occulter-edge metrology were also evaluated (and retired). This included in the observing (and analysis) plan a necessary high-precision re-determination of the occulter aperture effective location in the focal plane. This necessitated multiple fine repointings (image scans) with sufficient, but only relatively very shallow depth (low SNR), imaging as would be otherwise be implemented differently with other (scientific) priority. The HST/GO 12923 program was conducted in three parts: 1) Pre-execution redetermination and estimation of the locations of the deformed BAR5 and supporting BAR10 aperture locations and pointing fiducials that may have changed compared to "best available" data from pre-launch design due to zero-gravity release, two decades of desorption in the OTA and instrument optical bench, and astronauts stomping around the HST aft shroud during various servicing missions including a major on-orbit repair to STIS itself. 2) A coarse coronagraphic step-and-dwell image scan test of the (step 1) best-assumed aperture metrology to update the BAR10 (reference aperture) to BAR5 inter-aperture metrology, develop fine corrections for then "routine" use of BAR5, and obtain a first assessment of coronagraphic viability for the use of BAR5 using the coarse image scan data. 3) Pointing corrections derived from step 2 were applied to obtain follow-on observations of a second test target, using finer imaging scans to verify the (step 2) updated BAR5 pointing and coronagraphic performance results. Most importantly, this includes the quantifying both raw and PSF-template subtracted contrast curves for BAR5 coronagraphy, and use of the BAR10 corners. In steps 2 and 3, rather than image only point-source position and contrast calibrators, in GO 12923 we imaged two targets with well-studied edge-on circumstellar disks of well-established morphology and photometry in spatial regions commonly accessible to BAR5 and BAR10 corners beyond their unique IWA domains. This enabled an ability to test and validate with high confidence the BAR5 and BAR10 image fidelity and photometric efficacy with a priori knowledge in spatial regions of representative high-priority science targets previously observed. Targets Potential performance-degrading effects for coronagraphy can have wavelength dependencies in the unfiltered STIS 50CCD broadband response. Hence, GO 12923 was carried out with one "red" and one "blue" test target: AU Mic and β Pic, respectively (see Table 1). Both of these stars host bright, edge-on, debris disks with nearly equal visible-light scattering fractions, a priori characterized with STIS coronagraphy using its Wedge A and/or B occulters at their narrowest working angles (2x larger than BAR5). This allowed direct comparison of performance in regions of spatial overlap and, for BAR5, extending interior to the Wedge A/B IWA. Each test target was contemporaneously observed with a Δ\|B-V\| ≤ 0.04 color-matched PSF template star (see Table 1), for BAR5 and BAR10 PSF template subtraction. Orbits and Orientations Each target and its color-matched PSF template star were observed in three contiguous orbits, with the PSF star interleaved between two target orbits each at different spacecraft roll angles: Target (roll 1)--PSF--Target (roll 2). For BAR5, the absolute orientation angles for the target orbits placed the edge-on disk straddling a line roughly perpendicular to the long axis of the bar ± 15° from roll 1 (ORIENTAT for AU Mic 295.06°, for β Pic 208.06°) to roll 2; see Fig. 2. This also cleanly placed the disk between the "upper" and "lower" sets of diffraction spikes. For BAR10 the disk celestial orientations were the same as for BAR5. BAR10-Edge and Cross Mid-BAR5 Step-and-Dwell Imaging Scans A BAR5 aperture fiducial was not established or known to APT or other elements of the ground system when the observation plan 1 was designed (but later resulted from this test). Hence all target pointings in the observing plan were specified relative to (offset by POS TARGs from) the BAR10 aperture fiducial. By design, the latter was intended to be on the long axis of the BAR10 occulter, interior to the mask --though with detailed examination of on-orbit flatfield images we found that not to be the case with high accuracy. Thus, for this program to ascertain the relative metrology of the BAR10 and BAR5 occulters with highest precision, we designed imaging scans to observationally determine any unintended offsets w.r.t. BAR10 rounded corners, as well as across the BAR5 occulter. For details of planning and process, see Appendix A. The AU Mic visits (with PSF star, Visit ids 04 -06) executed first on 31 July 2013. Image-structure and pointing analysis revealed small, correctable, offsets in the commanded vs. executed positions. Details of the metrical analysis are presented in Appendix B. The residuals from these initial results were used to update/adjust the second epoch (23 Sep 2013) scans for β Pic and its PSF star δ Dor (Visit ids 01 -03) and lead, with post-execution confirmation, to recommend Project Data Base (PDB) and derived updates for relevant STIS apertures parameters. The planned imaging scan positions for the first-executed, AU Mic observations, as illustrated in Fig. 3, are tabulated in Science Instrument Aperture Frame (SIAF) and POSition TARGget coordinates (all relative to the SIAF and APT supported BAR10 aperture fiducial); see Table A-1. Coronagraphic Image Acquisition Each visit began with a two-FGS guide star acquisition, followed by a standard mode-2 target acquisition exposure to place the target at the first pointing position in the visibility period. Short, individual, coronagraphic exposures were designed to approach, but not exceed, safely conservative 50% -80% full well depth at the edge of the coronagraphic aperture, 0.15" from the core of an occulted star. Exposure times for the first epoch observations were estimated from prior coronagraphic imaging so as not to saturate. Exposure times were adjusted with both the STIS imaging ETC and TinyTim template PSFs and allowing for as much as 1/2-pixel of target acquisition decentering: for AU Mic 2.3s, for HD 191849 0.2s, for β Pic and δ Dor 0.1s (the minimum possible exposure time). At each step and dwell scan position, multiple short exposures were taken to build SNR and to mitigate effects of cosmic ray hits: 5 repeats for all targets except HD 191849 with 6 repeats. These exposure times were sufficient to enable both astrometric pointing analysis (and corrections) and contrast performance measurements while also revealing the host-star disks (though at low SNR) for comparison to existing observations obtained with other (larger) STIS coronagraphic occulting masks. Image Calibration/Reduction We used the calstis pipeline S/W with flat, bias, dark, and other calibration references files provided by STScI's calibration database system, to instrumentally calibrate the raw STIS images then transformed into count-rate images with CCD-GAIN=4 used in all observations. STIS "herringbone noise" was removed using autofillet (Janson et al. 2003, Proc. HST Cal. Workshop,193). The multiple individual instrumentally calibrated images at each scan position then were manually inspected for any anomalies (none found) and median combined into a single countrate image at each scan position. Astrometric Determination of the Stellar (SIAF) Position The SIAF location of the occulted (or partially occulted) star in each image was determined using the "X marks the spot" method that has previously been validated and used extensively in GO program 12228 (see Schneider et al. 2014, AJ 148 59; henceforth Sch14). Simply, the photometric mid-line intensity peak along the orthogonal linear OTA diffraction spikes were found, fit, and the intersection computed. Over the few arcseconds of the region considered the (very small) effects of differential geometrical distortion are negligible. For BAR5 this is very straight forward, as the diffraction spikes are symmetrically seen uninterrupted in the four diagonal directions originating at the target location. For the BAR10 corners (lower left illustrated in Fig. 4), one of the two spikes is nearly fully occulted on one side of the star by the mask itself -though it reappears in the sub-array used for data readout on the side opposite the star and is used to better constrain the fit. Note that the eye is highly biased in assessing "where is the star?" in images (such as in Fig. 4) where part of the PSF core, with FWHM appx 1.4 pixels, is partially obscured by the BAR10 edge. The eye tends to place the star at the bright but asymmetric and partially occulted photocenter of the PSF --which is NOT where the star is. The diffraction spikes (and origin at the true stellocenter) are unaffected by the introduction of the first focal plane BAR10 mask. For example, Fig. 4 is a BAR10 image (left; OBZE06030) that was planned to have the stellocenter 0.2" interior to the BAR10 mask along a 45° diagonal ([+2.76, -2.76] pixels in SIAF X/Y) tangent to the mask lower-left rounded corner (green dot, right panel). The "X marks the spot" solution reveals the star was not placed there, but was offset from that planned pointing by +0.068 -1.052 pixels (red dot). Details relating to that offset, and recommendations for remediation, are discussed below. Where is the Star? At the intersection of the two blue lines (red dot). Planned position indicated by green dot. Also note, the occulter edge itself is not geometrically sharp (in part because the mask plane is not perfectly conjugated to the reimaged focal plane, i. e., it is very slightly afocal when re-imaged on the detector, but also due to diffraction and scattering effects. In planning the point-position scans in this program we ascertained and used the edge-metrology from back-lit bright-sky and flat-field images at the half-power point of the brightness profile. To optimize for coronagraphic starlight suppression of the PSF core at a fixed baseline position, this 50% criterion may have been somewhat under-conservative, but was robust with the as-executed scan positions. In Fig. B-1, as can be seen visually, the best centering on the mid-line of the BAR5 occurred between planned bar-perpendicular offsets of +0.04" and +0.08" in all three target visits: 04 = AU Mic (roll 1), 05 = HD 191849 PSF, 06 = AU Mic (roll2). The small red circles indicate where the star was actually placed as determined from "X-marks the spot" centroiding. Table B-1 quantitatively gives the corresponding as-planned vs. as-executed target positions and offsets derived from these images. Note that the intra-scan repointings are well correlated between all three visits. All are also consistent within an expected dispersion (of about 1/4 pixel = 0.0125") in target-to-target non-repeatability due to Mode Selector Mechanism (MSM) non-repeatability in mask repositioning with deployment. N. B.: The VERY large tabulated differences in SIAF-Y is an artifact due to the fact that only small sub-array strips (100 rows wide, with offset Y-index about the regions of interest) were read out to reduce on-orbit dead time: centered at read-out row 720 for BAR5, and 840 for the BAR10 corners (Figures and Tables B-2 and B-3). The subarray start index in Y is 670 and for the BAR5, and 790 for the BAR10, images. This sub-array start index is later subtracted to get the actual difference w.r.t. the SIAF locations on the full (but not fully read-out) detector. The intra-visit differences from planned to achieved positions were highly repeatable with very small dispersion within expectations of rms jitter for two-star fine lock (~ 4 mas = 0.08 pixels). The inter-visit mean differences have a larger dispersion, posited as due to the repeatability limits of initial target placement with independent target acquisitions (at different rolls and targets), separate from MSM deployment imprecision. • For BAR10 LL (Appendix B, Fig. & Table B-2): In a similar method of analysis at each of the scan points, the target "missed" the planned position an average of (0.0388, 0.928) ± (0.069, 0.191) pixels. The as-planned BAR10 POS TARGS (for LL corner) should have been altered by (+0.00197, +0.04711) arc seconds. • For the BAR10 LR (Appendix B, Fig. & Table B-3): At each of the scan points, the target "missed" the as-planned position by a very similar average of (0.0716, 0.9624) ± (0.087,0.267) pixels. The as planned BAR10 POS TARGS (for LR corner) should have been altered by (+0.00364, +0.04886) arc seconds. These pointing updates implicitly correct for geometrical distortions between the BAR10 aperture fiducial and the determined "best" BAR5, and BAR 10 LL & LR locations, and were applied to the β Pic and δ Dor PSF visits 01 -03 before execution on 23 Sept 2013. Additionally, for both the BAR10 LL and LR corners, the stellar PSF core disappears between scan step planned positions of +0.20" and +0.25" from the as defined rounded edge of the corners. We therefore adopt a "best" position mid-way between these scan points (+0.225" w.r.t round edge) as the corrected mid-point of the BAR10 scans. BAR5 IWA Performance (AU Mic, Visits 04-06) AU Mic, as imaged in both Visits 04 and 06, was very close to BAR5-centered at the step-anddwell image scan last position (3rd exposure set in time in both visits, planned for a +0.08" offset). The PSF-template observations from Visit 05 do not precisely replicate the AU Mic visit scan offsets. However, as executed, they nearly equally flanked the near BAR5-centred AU Mic positions with the PSF pointings at positions 4 and 5 (+0.04" and +0.08" planned offsets). For these reasons, only these occulter position-matched data (AU Mic Visit 04/06 scan position 5, and PSF Visit 05 scan positions 4 and 5) were used to evaluate the BAR5 coronagraphic performance. The two BAR5 best-centered AU Mic images (one each from Visits 04 and 06) and the two most-closely corresponding PSF images from Visit 05, flanking the AU Mic positions, are shown in Fig 5. In all cases, the unocculted PSF halo at the closest proximity to the BAR5 edges is well exposed, but not saturated. So (in principle) the exo-BAR5 data are valid for PSF-subtraction in all regions (if not photon-limited), but are degraded in directions of the HST diffraction spikes. An empirically determined PSF flux-scaling (intensity renormalization) factor of x0.58 relative to AU Mic was used. For each subtraction, the PSF template image was aligned to the location of the star in its corresponding AU Mic image initially by "X marks the spot" centroids, then with very fine empirically determined offsets minimizing the residuals in subtraction in the OTA diffraction spikes (see Sch14). The final set of the four individual PSF-subtracted images are shown in Fig. 6 (left and center panels). As previously noted for the purpose of this commissioning program, the exposure depth is sufficient to verify coronagraphic performance, but is very shallow (photon limited) in revealing the AU Mic disk. With a different (science) priority, deeper integrations (e. g., following GO 12228; Sch14) would be warranted. As can be seen in the Fig. 6 images, the further suppression of much of the remaining starlight after BAR5 coronagraphy, by PSF-template subtraction, is both effective and highly repeatable. This is evidenced, after target/template star alignment and scaled-subtraction, by the very deep near-nulling of the OTA diffraction spikes. Aligning the imperfectly BAR5-centered template star to the position of the imperfectly BAR5-centered target star, however, causes a mis-registration of the BAR5 edges. Indeed, the stellocenter and BAR5 edges cannot be simultaneously aligned unless both the target and template star were exactly pointed to the same place (ideally BAR5-centered). By taking out the unintended target/template small pointing offsets to null underlying light in the stellar PSF halo, the star-illuminated BAR5 edges become mis-aligned and in difference images result in opposite edge ± brightness patterns at the opposing BAR5 edges. This is easily seen in the Fig. 6 images where the disk appears at the two different celestial orientations used in Visits 4 and 6. An inversion in parity in the ± brightness pattern at the BAR5 edges (on opposite sides of the star) is seen comparing the PSF subtracted images using the PSF templates from offset positions 4 & 5. This indicates that a better nulling of the ± pattern would have been achieved with a template image taken between these two positions. The brightness of these artifacts at the + andedges is (to first order) "close to" equal in amplitude (but opposite in sign) in the position 4 and position 5 PSF subtracted images (though slightly different for the respective Visit 04 and 06 images). This implies that an intermediately placed PSF template, acquired approximately mid-way between scan positions 4 and 5 should have been able to simultaneously null the PSF halo, and light scattered by the BAR5 edges. None exists for these observations. However, by digitally masking the + and -artifacts and then combining the images, we are able to produce a rudimentary science quality image (Fig. 7). The achieved IWA of this reduced image at 0.20" closely approaches the hard-edge limit of 0.15". The image scans identified the ideal POS TARG offsets from the BAR10 aperture fiducial required, and with ± 1/4-pixel cross-bar dithers to mitigate pointing uncertainties should reliably allow high quality and small IWA data acquisition (see § 15, item 5). Note: These AU Mic data preceded the Visit 01 -03 β Pic data by ~ 2 months. For those later β Pic observations, along with the initial pointing correction offsets previously discussed applied, finer scan spacings were implemented and compared to the coarser "out of the box" AU Mic observations. The pointing corrections applied were based upon the above nearly nulled positions. To first order in this data set, the phase reversal of this ± BAR5-edge artifact brightness pattern from Visit 04 to 05 suggested that, as a proxy to having an "intermediate" PSF template, applying linear combination of the two available PSFs could reduce the magnitude of this pollutant. This was done, for simplicity here just assuming equal weighting for the two images, with averaged results shown in the third column in the above mosaic of images. The two averaged images of the AU Mic disk at different celestial orientation angles ( Fig. 6. right panels), were then "roll combined". I. e., co-aligned at the location of the star, rotated to celestial north "up", and then averaged with digital masking (rejection) of regions unsampled or degraded by the presence of the BAR5 occulter and the HST diffraction spikes. This two-roll combined image is shown in Fig. 7, clearly detecting and resolving the AU Mic disk along its edge-on major axis to a smallest stellocentric angle of r = 0.20", closely approaching (less than an innermost resel 2 from) the 0.15" half-width of physical edge of the BAR5 aperture. Fig. 7. Two-roll PSF-subtracted BAR5 imaging of the AU Mic inner disk Visit 04 & 06 best finger-centered images (3rd exposure set in each visit, planned for +0.08" offset). PSF template using star co-registered average of PSF template images from 2nd & 3rd exposure sets (planned for +0.04" and +0.08" offset). Log 10 image display dynamic range [-0.3] to [+2.0] {dex} counts s -1 pixel -1 . The green circle is r = 0.20" (2 AU at AU Mic's distance of 10 pc). Total integration time combining all images used in both rolls is 23 s. Comparison to Prior Deeply-Exposed Wedge-A Results In HST/GO 12228 we obtained deep imaging of the AU Mic disk using 6 HST orbits at differing spacecraft rolls, coronagraphy at two different WedgeA occulting positions: WedgeA-1.0 and Wedge-A0.6, and utilizing a total of ~ 12 ksec of integration time. This produced a very high (photometric and astrometric) quality image for analysis, but with an achieved inner working angle limit of r = 0.5" (r ≈ 10 pixels). In HST GO/12923 (Fig. 7), using BAR5, we newly imaged the inner part of the larger FOV explored in GO/12228, 2.5x closer than had previously been possible. In Fig. 8 we reproduce the GO 12228 (Sch14) 6-roll combined image of the AU Mic "inner disk" region, in detail comprised of a 3.3 ksec integration using WedgeA-0.6 for the innermost part of the disk at r < 0.7"-1.0", and an additional 9.8 ksec using WedgeA-1.0 beyond. We show this along with our 23 s total integration time GO/12923 two-roll combined BAR5 image (same data as in Fig. 7) at the same image scale, orientation, and display stretch. The image is derived from a single scan position with five 2.3 s exposures from Visits 04 and 06 of AU Mic. PSF subtraction was done using PSF templates from scan positions 4 & 5 combined from Visit 05. The GO 12932 image noise floor is, obviously, significantly higherdue to the differences in the observations, notably only 23 s of total integration time compared to up to 12 ksec, but also only two-rolls on the target compared to six. Despite the differences in sensitivity for these reasons, the IWA achieved in the bright, inner, region of the disk in GO/12923 improves over the GO/12228 imaging by a factor of 2.5. I. e., reducing the IWA from r = 0.5" (GO/12228) to r = 0.2" (GO/12932). The higher fidelity GO/12228 image serves to validate the disk photometry in the regions commonly sampled in the GO 12932 image (which agree very well, despite the "noise", as can be seen in these side-by-side images); thus indicative of no significant amount of stray starlight outside the BAR5 centered subarray in the commonly measured region r > 0.5" from the star. Follow-On Imaging Resultsβ Pictoris (Visits 01 -03) The first epoch AU Mic image scan observations provided the necessary occulter metrology information to compute post priori pointing corrections for subsequent imaging. The computed offsets 3 were applied in redefining the POS TARGs of the follow-on β Pic and δ Dor (PSF) visits. The a priori known intrinsic repositioning uncertainty in the deployment of the coronagraphic optics by the STIS MSM (resulting in deviations from perfect target centering) is only ~ ±1/4 pixel. To test/confirm the efficacy of derived pointing corrections necessary to optimize coronagraphic centering, performance, and effective IWA, we thus conservatively planned image position offsets 3x larger at the extrema then the MSM redeployment uncertainty and with inter-point scan position steps finer than in the initial AU Mic orbits. Specifically, for β Pic (Visits 01 and 03) and its PSF star (Visit 02) we executed 7-point stepand-dwell coronagraphic imaging scans with incremental offsets of +0.015" (0.3 pixels). For the BAR10 LL and LR corners the target was initially placed +0.225" along the respective 45° lines across the rounded edges. For BAR5, the cross-bar scans went from -0.045" to +0.045" across the mid-line of the occulter. "X marks the spot" stellar position determinations for the BAR10 scans confirmed the target positioning from the updated POS TARGS within an uncertainty of ~ 1/4 pixel. For BAR5 we then performed PSF-template-subtracted coronagraphy for all permutations of the disk and template star (multi-exposure median combined) images from all scan positions. With the pointing corrections derived from the prior AU Mic observations used to command the scan steps, the predicted "best" position on the mid-line BAR5 was at scan step 04, (a "half step" from the middle of the scan), and ± ~1 step for both β Pic and its PSF star δ Dor. In Fig. 9 we show all possible combinations of PSF subtractions from Visit 03, stretched to simultaneously show: (a) the β Pic disk, (b) the BAR5 occulter and its ± edge gradients, and (c) the HST diffraction spike residuals. The best, and smallest IWA, images of the disk result when both the HST diffraction spikes and the BAR5 opposing light and dark edges are simultaneously nulled, anticipated in PSF subtracted images when both the target and template stars are positioned on the mid-line of BAR5. By visual inspection of Fig. 9, this occurs for β Pic position 04 (central position, as planned) and PSF template position 03 --the dark green cell in Fig. 9. I. e., the template star at its planned position 04 was mis-positioned by (only) -1 scan step in SIAF Y w.r.t the mid-line of the bar. This also indicates that the system metrology is stable to ≤ ± 0.015" on greater than monthly timescales, as we determined the requisite POS TARGs from the AU Mic observations acquired 2 months prior to the β Pic observations. In Fig. 9, below each image, are the post-facto determined PSF template image position shifts in SIAF X and Y pixels that were required to minimize the diffraction spike residuals by co-aligning the target and template stars independent of their possibly de-centered locations from the mid-line of BAR5. As can be seen, the co-registration correction needed to produce the best image was a template shift of (-0.03, +0.11) pixels, i.e., ~ half of the 1/4 pixel absolute positioning uncertainty. I. e., with perfect placement, the best image would have been in the central cell of the 7x7 mosaic of scan position images. Given the ~ 1/4 pixel relative pointing precision and MSM redeployment uncertainties, we recommend for highest fidelity PSF-subtracted imaging that all BAR5 observations be executed with a linear 3-point position "dither" of ~ (-1/4, 0, +1/4) pixel orthogonal to the bar with post-facto stellar co-registration prior to image combination for science analysis. Very similar results were found from the Visit 01 step-and-dwell image scans (identically executed, but at different celestial orientation). Fig. 10 illustrates, side-by-side at the display same stretch as Fig. 9, the best three PSF-subtracted images from the Visit 01 and Visit 03 in the SIAF frame (so the disk rotates about the occulted star from Visit 01 to 03), while the diffraction spike residuals and BAR5 occulter imprint remains rotationally invariant. ] with both targets commanded to the mid-line position of BAR5, but was offset by -1 step in the template position due to ~ 1/4 pixel mis-centering of the image scan. Movement along the green diagonal preserves the 1 step in X and 1 step in Y relative positioning of both targets w.r.t. the BAR5 location, so bar edge brightness gradients remain shallow. In the orthogonal direction (lower-left to upper right) the co-aligned stars become increasingly decentered w.r.t. BAR5 with the opposing edge brightness gradients of opposite parity increasing toward the respective coroners. Fig 10. The as-determined best, and flanking, PSF-subtracted BAR5 images of the AU Mic disk at two celestial field orientations exploring as closely as possible the IWA space adjacent the BAR5 edges with small (1/4 pixel) cross-bar dithers, prior to later dither and roll combination into a single image. This is a demonstration observation using only two rolls. Enabled science observations using six field orientations would completely sample around the BAR5 occulter at all azimuth angles ≥ 0.15", and decorrelate quasi-static PSF-subtraction residuals in the celestial FOV (see Sch14 for analogous discussion and proof of concept with Wedge-A 0.6 and 1.0 occulting masks). Prior to this enabling experiment for smallest IWA coronagraphy with HST, the β Pic disk had been most aggressively probed in GO program 12551 (Apai et al. 2015, ApJ 800 136) using a combination of: (a) STIS occulting wedges A and B at, (b) their 0.6" and 1.0" full-width taper positions, (c) with two spacecraft roll angles, (d) total integration time 1178 s, (e) contemporaneously interleaved PSF template observations. The reduced image of the inner part of the disk from the GO 12551 data set, combining all images in a north-up orientation while masking the "here be dragons" regions in the input images obscured by the STIS wedges and edge artifacts, is shown in Fig. 11 (left panel). The effective smallest IWA achieved, optimized in the GO 12551 observing plan 4 to image the edge-on disk in only two-rolls, was r = 0.35" -0.40" (Apai et al. ibid) and larger at other stellocentric azimuth angles. We compare the GO 12551 reduced image in Fig. 11 (left panel) to a similarly-combined image using the exploratory BAR5 data obtained in GO 12923 (Fig. 11, right panel) also at two rolls (Visits 01 and 03) but with total integration time only 3 s. The GO 12923 reduced image is derived from only the three cross-bar dithered images (in each visit) centered on and immediately flanking the BAR5 mid-line as shown in Fig. 10 in comparative accordance with our recommended 3-point dither imaging strategy. The BAR5 reduced image is derived from these six images. Though it is noisier than the GO 12551 image (see Fig. 11 caption), it successfully reveals the inner disk to an IWA ≈ 0.15" in the two-roll optimized direction along the edge-on disk plane after digitally masking the BAR5 occulter in the original images. In the regions commonly sampled, the morphology and surface brightness (SB) distribution of the β Pic starlight-scattering disk is well reproduced with the BAR5 imaging, with an ~ 2.5x improvement in IWA over the GO 12551 image. The high degree of morphological correlation in these images, gives confidence to the GO 12923 newly revealed image structure in the region interior to the GO 12551 IWA. Fig. 11 compares the morphology, and verifies the imaging reproducability, of the inner β Pic edge-on disk at r ≲ 3" in the regions commonly sampled in the GO 12551 (left) and GO 12923 (right) data sets. The latter, designed to push the IWA, is limited in outer stellocentric distance by its short integration time and the subarray readout. These apples-to-apples image displays are optimized to allow a best comparison with the smaller FOV and more limited dynamic display range appropriate for the GO 12923 BAR5 data. The much more deeply exposed GO 12551 image, by virtue of its ~ 400x greater total integration time and use of four occulting wedge positions, has higher SNR and sensitivity to lower SB dust much further out (to r ~ 10") though is not illustrated in this figure; see Apai et al. 2015. In Fig. 12, we overplot the radial SB profile of the edge-on β Pic disk measured identically from the independent GO 12551 (Wedge A/B, 0.6+1.0) and the GO 12923 (BAR5) reduced image data sets (i.e., from Fig. 11). The high degree of reproducability of the GO 12551 profile from the GO 12923 data in the regions commonly sampled validates the efficacy of the BAR5 disk photometry, and utility of the BAR5 occulter for narrow-angle PSF-subtracted coronagraphic photometry. The region interior to r ~ 7 pixels is measured only from the GO 12923 data, though the smoothly contiguous profile is indicative of the absence of image artifacts of significance beyond the masked edges of the BAR5 occulter. BAR5 Starlight Suppression & Image Contrast The ability to image faint point-, and low-SB spatially-resolved, sources in the close angular proximity to their much brighter host stars relies on observing techniques to suppress the background light due to the stellar PSF halo. With HST/STIS this is accomplished with a two-tier approach. (1) With coronagraphy using a simple Lyot stop and image plane occulters of different angular sizes. (2) Post-facto, through PSF-template subtraction further suppressing the stellar light in the PSF halo not rejected with coronagraphy alone. The image contrast at any unocculted field point provides a useful metric to assess the depth of the starlight rejection that then sets a detection floor for circumstellar objects. By contrast we mean: the ratio of the background light due to the coronagraphically suppressed PSF halo in a pixel (or resel) at any field point compared to the brightness of the central pixel (or resel) at the peak of the PSF if measured directly without coronagraphy or other methods of starlight suppression/rejection. Image contrast is both radially and azimuthally dependent, generally with improved (deeper) contrasts further from the star, but with azimuthally dependent structure at any radius. A contrast curve expresses the contrast as a function of stellocentric angle averaged, or medianed, in circumstellar annuli of increasing radii from the coronagraphically occulted star, i. e. an azimuthally averaged or medianed radial surface brightness profile in dimensionless contrast units. To quantify the intrinsic starlight suppression with BAR5, we directly measured (in instrumental units of counts s -1 pixel -1 ) the SB of the coronagraphically well-centered δ Dor (Fig. 13, left panel) and HD 191849 template PSFs (Fig. 5, right panel) from the GO 12923 observations. In doing so we digitally masked the imprint of the occulting bar itself and the regions degraded by the unapodized diffraction spikes (e. g., Fig. 13, center). In the GO 12923 BAR5 enabling program we did not, however, image the unocculted stellar PSF template stars directly. Indeed, the BAR5 occulter was required to prevent image saturation at, and near, the PSF core. For background discussion on direct vs. coronagraphic imaging with STIS see Grady et al. 2003; PASP, 115, 1036. Thus, for direct to coronagraphic comparison (but not for subsequent PSF subtraction) we used TinyTim 5 model PSFs (Krist et al. 2011; Proc. SPIE, 81270J) in concert with the STIS imaging ETC (built on synphot) to obtain very close estimates for the flux (instrumental count rate) in the central pixel of the model PSFs: 2.63 x10 7 counts s -1 pixel -1 for δ Dor, and 1.26 x10 6 counts s -1 pixel -1 for HD 191849. These scale factors were used to transform the SB profiles in instrumental units of counts s -1 pixel -1 of the BAR5 images to a dimensionless contrast curves. Fig. 14 shows the raw BAR5 coronagraphic contrast curves as measured from the masked PSF images, calibrated through a direct image based on the central-pixel scaled TinyTim model PSF. Despite their significant differences in brightness and [B-V] color, thanks to the instrumental stability, the azimuthally medianed contrast curves for δ Dor and HD 191849 are essentially identical. The greater dispersion about the medians for HD 191849, increasing with stellocentric angle, are due to the relatively shallow depth of the only 2x longer integration time (constrained by the experimental design) compared to δ Dor though HD 191847 is 3.6 magnitudes (28x in flux) fainter. Fig. 14. BAR5 raw coronagraphic contrast (i.e., without PSF subtraction) -radial dependence. 360° azimuthal median with the BAR5 occulter and OTA diffraction spikes excluded (as shown in Fig. 13, middle panel). Error bars represent the 1-sigma dispersion in contrast around the median values. The angular distance from the star to the physical edge of the BAR5 occulter, and one full resolution element (~ 72 mas) beyond is indicated. BAR5 Contrast Augmentation with PSF-Template Subtraction In GO 12923, the δ Dor PSF template was applied to the β Pic images to unveil the disk, but no calibrator for the template star itself was obtained -but is needed to show quantitatively the benefits of PSF subtraction clearly (i.e., in the absence of a disk signal). This was beyond the scope of the GO 12923 orbit allocation. To that end, however, here we make use of PSF calibration observations from the GO science program 13786 (Schneider et. al. 2016 AJ 152 64) that followed the commissioning of BAR5 and used the newly enabled occulter in the manner suggested in this ISR. In GO 13786, HR 4735 (V=5.56, B9V, central pixel count rate 9.86 x10 6 counts s -1 pixel -1 with CCDGAIN=4) was used, as a contemporaneously imaged PSF template (applied to observations of the HR 4796A debris disk at two epochs six months apart). The two-epoch set of reduced BAR5 HR 4735 template images is shown in Fig. 15, left and middle panels. Importantly, they were obtained widely separated in time with very different Sun, beta, and roll angles, and hence with the OTA in different thermal states driving wavefront instabilities; exactly the systematics one would normally take effort to mitigate or minimize through observational design. Here, as a demonstration observation, we advantageously use the secular non-reproducibilities in these two different images of the same star in quantifying the raw repeatability, and PSF-subtracted image contrast limits. Subtracting a second identically exposed and reduced image of the same star obtained six months later (from Visit 47 of the same program, middle panel), with the telescope in a different thermal pointing profile, reveals contrastlimiting differences in PSF structure (right) quantified in Fig. 16. Both the raw BAR5 coronagraphic, and PSFtemplate subtracted images, are shown with the same linear display stretch chosen to best show the PSF-subtraction residuals close to the BAR5 edges. With a prior, fully independent determination of the BAR5 raw contrast curve as measured from the GO 12923 imaging of δ Dor (Fig. 14), in Fig. 16 we present raw contrast curves derived from the first and second epoch observations of HD 4735. Both of these curves, as in the case of δ Dor that is overplotted in this figure, are 360° azimuthal medians with the regions degraded by the OTA diffraction spikes, and obscured by the BAR5 occulter itself, excluded. Despite the low-amplitude differences in the fine (high spatial frequency) image structure revealed in Fig. 15 (right panel), the azimuthal median contrast curves for both HD 4735 template observations are virtually indistinguishable. Both, also, are nearly identical to the δ Dor curve that does deviate by a small amount only very close to the BAR5 edge, and (inconsequentially) beyond ~ 2". For clarity of presentation we did not overplot error bars in Fig. 16, though those found for both epochs of the HD 4735 contrast curves very closely replicate those found for δ Dor presented in Fig 14. I. e., all three raw contrast curves are very highly repeatable. Raw (top; black, red, blue) and PSF-template subtracted (bottom, green) 360° (excluding diffraction spikes) median contrast curves for BAR5 coronagraphy. The raw contrast curve determined from observations of δ Dor (black), used as a PSF subtraction template for β Pic (with results presented earlier in this ISR), is compared to two highlyrepeatable, independent contrast curves derived from similarly executed observations of HR 4735. The latter were then used to produce the azimuthal median BAR5 PSF template subtracted coronagraphy (PSFTSC) contrast curve (green). To the eye, the two HD 4735 PSF template images in Fig. 15 (left and middle panels) look identical. They do, however, have differences in image structure, beyond photon noise, due to secular instabilities and non-repeatabilities in the telescope plus instrument system that are revealed with "self-subtraction" (Fig. 15, right panel). Such differences (PSF subtraction residuals) also appear imprinted (as "noise" or "image artifacts") on PSF-subtracted images of circumstellar disks and close-angular proximity faint companions. These residuals, however, are more are readily and cleanly seen using the star as its own (imperfect) PSF-subtraction template. In the non-photon noise dominated regime, these PSF-subtraction residuals ultimately set the contrast-limited detection sensitivity. It is visually apparent, from inspection of Fig. 15, that PSF-template subtraction has done a laudable job in reducing the residual light in the circumstellar PSF halo. To quantify the level of starlight suppression achieved, we use the PSF-template "self-subtracted" image of HD 4735 as shown in Fig. 15 (right panel) to assess the with BAR5 coronagraphic contrast limits with PSF subtraction. This is shown in the green curve in Fig. 16. With PSF subtraction, the residuals can be (and are) biased both positively and negatively with respect to the true background. I. e., imperfect nulling will result in both positive or negative signals (e. g., light and dark compared to "true gray" in Fig. 15, right panel), of non-astronomical origin, setting the contrast floor. Thus, the azimuthal median residual amplitudes that set that floor are plotted in absolute value, i.e., the median absolute deviation 6 (MAD). We overplot the best power-law fit to the measured 1-MAD contrast limit (dashed line) that may be smoothly approximated as: PSF-template subtracted 1-MAD contrast ≈ 4.45 x10 -7 x r -3.80 ; with r in arcseconds A note on "color" and companion & disk imaging: Using the same star as its own PSF subtraction template has the benefit of fully eliminating chromatic differences in PSF structure that are otherwise apparent with stars differing in optical color indices. Fully eliminating this one variable (difference in template "color") provides a result equivalent to ideal "color-matching" with a different template star. This (by elimination) decouples image residuals in PSF-subtractions due to mis-matches in template star colors from those that are more difficult to control except (for optimal science) through contemporaneous, and pointing-constrained, PSF calibration observations. As an observing technique, this two-roll method is particularly useful for imaging faint point-source companions where the roll angle difference is sufficient for the companion stellocentric distance to result in well separated positive and negative imprints of the companion in the difference image. This two-roll "selfsubtraction" method, avoiding chromatic PSF-subtraction artifacts, also has utility for disk detection for edge-on and high inclination disks (e. g., Heap et al. 2000, ApJ 539 435; Krist et al. 2012, AJ 144 45). Augmentation with additional rolls and dither positions is discussed on: http://www.stsci.edu/ hst/stis/strategies/pushing/coronagraphy_bar5. However, intermediate to low inclination (face-on) disks will suffer from partial to total flux cancellation with "self-subtraction". The mitigation of this effect requires, instead, PSF-template subtraction with a well color-matched PSF template star. Using the BAR10 Rounded Corners The STIS image plane coronagraphic mask includes a 10" long by 3" wide rectangular occulter with rounded edges extending into the FOV from the "top" (SIAF +Y) of the frame known as BAR10; see Fig. 1. The conceived utility of BAR10 was for wide-angle coronagraphic imaging of the circumstellar fields around the brightest stellar targets such as Fomalhaut. I.e., imaging low surface brightness CS dust or low-mass companions at large stellocentric angles from the brightest disk-or planet-hosting stars. BAR10 is wider than the tapered "tops" of either of the occulting wedges. BAR10 was designed to be used by placing a target in the middle of the bar, or possibly pushed "downward" (with a POS TARG) closer to its bottom edge, but still along the mid-line of the bar. Despite its posited utility for such observations, BAR10 has found very little use since it was commissioned in Cycle 7, with Wedge A/B occulters favored by observers. The GO 12923 program was devised to enabling narrow(est)-angle coronagraphy with the 0.15" half-width BAR5 occulter (10x narrower than BAR10). Success, however, was not a priori assured. Hence, we conceived a possible, non-conventional, use of the BAR10 occulter for offcentered narrow-angle coronagraphy by placing a star near one (or both) of its rounded corners. Each would provide a FOV beyond its IWA covering ~ 270° in stellocentric azimuth angle. BAR10 was not designed for this purpose, and the fabrication quality (precision in manufacturing) for starlight rejection at and beyond its rounded edges at the extreme apices of the mask was unknown. We thus took the opportunity, in GO 12923 to explore this possibility. To explore the potential utility of BAR10 "rounded-corner" coronagraphy, we conducted step-and-dwell image scans using β Pic and its PSF template star δ Dor. This was done in the second set of visits (#'s 1 -3) after executing and evaluating the earlier coarser scans (visit #'s 4 -6) that were used to verify/update aperture positions and pointings with AU Mic as previously described. We found from those data that placing a star closer than ~ 0.225" from the BAR10 rounded corner edges would produce too much edge-illuminated light to be useful -see Figs. B-2 and B-3. We then performed finer-spaced BAR10 imaging edge-scans from 0.225" to 0.300" (the half-width of WedgeA/B-0.6), repositioning the target in incremental steps of +0.015" interior to the occulter LL and LR rounded edge on a 45° diagonals (e.g., see Fig. 17). Fig. 17 illustrates the raw coronagraphic imaging results for the first β Pic visit 7 #1 (BAR10 LL corner). The presence, but incrementally decreasing level, of un-rejected edge-scattered light at and beyond the rounded-edge corner is visually apparent from the first to the last target position. Table 2 gives the actual stellar locations at each LL scan position (in SIAF coordinates), recovered from these data with "X-marks the spot" centroiding, used in the later generation of stellocentric contrast curves for each image. In Fig. 18 we present ~ 180°-azimuthally medianed radial surface brightness profiles for each of the six target positions shown as contrast curves. For contrast calibration, we adopted an estimate of the flux in a modeled unocculted central pixel for a β Pic PSF as 6.49 x10 7 counts s -1 pixel -1 by the same method described prior for δ Dor and HD 191849 in § 11. In stellocentric azimuthal medianing, we digitally excluded nearly all of the BAR10 mask area, except very near and including the rounded corner that is the source of potentially problematic light limiting the image contrast in this region of interest. The digital masks also excluded the "upper-left" to "lower-right" diffraction spikes, but did not exclude the "lower-left" spike that is superimposed upon the rounded-edge scattered starlight, as it would appear in the region of particular interest at small stellocentric angles; see the masked-image inserts provided in Fig 18. for illustrative examples of digital masking for scan positions 1 and 6. Each contrast curve turns over very close to the occulter-edge effective IWA, corresponding the distance from the star to the rounded-edge of the BAR10 corner. In Fig. 18 the locus of the turn-over points is connected, in approximation, by the black dotted line. To the left of that line the "turned-over" data are corrupted or invalid being impacted by the (small) digitally unmasked region of the BAR10 edge. To the right of this line, the data are valid and there the raw IWA vs. contrast performance may be assessed. As can be seen in Fig. 18, on the azimuthal median the IWA contrast is best (smallest) by a factor of ~ 3 with the star positioned +0.300" (POSition 6) compared to a smaller positioned IWA of +0.225" (POSition 1). I. e., at 0.300" the contrast is improved, but IWA comparatively suffers. The corresponding contrasts, however, suggest that with augmenting efficacious PSF subtraction, the BAR10 rounded-corners may provide additional useful capability. BAR10 LL and LR Rounded-Corner Imaging Scans The differences in the azimuthally medianed contrast performance diminishes at larger stellocentric angles, and vanishes beyond r ≈ 0.5"; see Fig. 18. Very similar results are obtained with the BAR10 lower right corner, and also with the second β Pic visit #3 and also for the δ Dor PSF visit #2. The two-dimensional morphology of this stellar background light in this region, however, is complex, as can be seen for the LL corner in Fig. 17. The BAR10 rounded-edge brightness pattern has dependency upon the two-dimensional (im)precision of the target centering w.r.t. the round-edge center of curvature location. This presents additional complexities, and uncertainties, in PSF-template subtraction that are obviated (in one dimension) with BAR5 straight-edge coronagraphy. BAR10 Rounded-Edge PSF Subtraction We next evaluated the efficacy of BAR10 rounded-corner PSF-subtracted coronagraphy. As is not atypical for STIS coronagraphy in general, the Visit 2 PSF images at same commanded scan-point offset positions as the β Pic Visits 1 and 3 images have (a) mis-alignments in the projected location of the BAR10 occulter on the CCD due to MSM repositioning nonrepeatabilies. They also have (b) mis-alignments of the target placement w.r.t. the BAR10 itself (and as seen re-imaged at the CCD focal plane) from target acquisition imprecision that must be "corrected" in post-processing prior to PSF subtraction. (See Sch14 for details as discussed analogously for the Wedge-A occulter). (b) is correctable (except very close to the BAR10 edges) in post-processing by PSF image registration for (a) best-matched scan positions that approximately compensate for MSM position re-deployment non-repeatabilities. E. g., in normal science observations, as recommended separately for BAR5, with ± 1/4 pixel "dither" steps providing a full-range of a half-pixel offset at the extrema for target:template position matching and correction. The β Pic/δ Dor BAR10 exploratory image scans in GO 12923 used an inter-scan step size of 0.015" that is very close to 1/4-pixel. Because of (a), in Visit 1 for β Pic at the LL corner of BAR10, only images at scan steps 3 - Fig. 19. PSF-template subtracted images of the β Pic disk with the star occulted by the BAR10 LL rounded corner. Leftto-right with incremental offsets of 0.015 arcseconds closely compensated in position matching of the BAR10 imprint in PSF image scans with the post-factor co-alignment (co-registration) of target and template stars before subtraction. 100x100 pixels (~ 5.1" square) FOV centered on the star. Within each visit we separately executed identically designed, but "mirror symmetric", imaging scans at both the LL and LR BAR10 corners. Following the PSF-subtraction process described above, Fig. 20 (left and middle panels) illustrates by example the high degree of imaging repeatability in commonly sampled regions (i. e., "below" BAR10), and averaged in the right panel. Here we compare the (third) scan position in visit 1 at the LL (same image as left panel in Fig. 19) and LR corners. For either the LL or LR corners, BAR10 imaging provides an unocculted stellocentric field approaching 270° in azimuth beyond the effective IWA. There is a "gap" however in the overlapped regions set by both the IWA and artifacts along the BAR10 edges; see Fig. 20, right panel. Additionally, the post-facto stellar registration process results in BAR10 artifacts that are manifested as steep positive or negative brightness gradients that externally flank the bar edges. This can invalidate the data up to a few pixels beyond the physical edges. In Figs. 19 and 20, as presented, these artifacts appear as hard black (negative) or hard white (positive) near-linear features at the bar edges, through the latter (positive) is difficult to distinguish from the disk itself in the display stretch used for other illustrative purposes. In Fig. 21 we illustrate the typical extent of these edge artifacts in a stretch where more easily seen. To test the robustness of BAR10 corner PSF-subtracted coronagraphy in concert with a "standard" two-roll observing paradigm, we followed an identical process for the same β Pic scan point images (3-6) in visit 3. In this visit, the disk was re-oriented +30° w.r.t. visit 1, thus requiring independent MSM deployments and target acquisitions. In this case, best position matches for each of these scan points were found from the δ Dor LR corner scan positions 3 -6 (i. e., the same as for the β Pic points, rather than 1 -4 as in visit 01), respectively. Those resulting images are shown in Fig. 22. As can be seen, despite the differences in the scan-position dependent raw contrast floor close to the BAR10 rounded corner edges the images all exhibit a very high degree of repeatability. Fig. 22. PSF-template subtracted images of the β Pic disk with the star occulted by the BAR10 LR rounded corner, from visit 03 image scans (positions 3 -6) with the disk axis rotated +30° clock-wise w.r.t. visit 01 (Fig. 19). BAR10 Rounded-Edge 2-Roll PSF-Subtracted Image Combination The rotation of the field by 30° from visits 1 to 3, in combination, reduces the gap in spatial coverage flanking the BAR10 vertical edges to a small triangular wedge beyond the IWA, as shown in Fig. 23. Fig. 23 combines the four LL and four LR PSF-subtracted images shown in Figs. 19 and 22 by co-registering the eight images at the location of the occulted star, rotating all to a common celestial frame, and individually applying digital masks to the BAR10 occulted, and (most) diffraction-spike affected, regions prior to median combination. The disk is recovered at r ≥ 0.36" along the NE major axis, and at r ≥ 0.46" along the SW major axis (being impacted interior by the remaining (only) two-roll gap in the LL + LR spatial coverage). The red circle indicates a stellocentric angular radius of 0.25", the smallest effective IWA over the range of azimuth angles where the disk had been recovered in the two-roll reduced image. Fig. 23. Two-roll BAR10 LL and LR combined image of the β Pic circumstellar disk to a field limit of 100 x 100 pixels (5.1" x 5.1") commensurate with the total depth of integration of 0.8s for all eight images. Left: linear display from -20 to 100 counts s -1 pixel -1 . Right: Log 10 display from [+1] to [+4] dex counts s -1 pixel -1 . Both "north up" with the a priori known orientation of the disk major axis (PA = +29.1°) indicated by the green line in the left panel. Red circle: r = 0.25". The morphology of the inner (r < 2.6") region of the β Pic disk is well recovered in the 2-roll BAR10 LL + LR combined image. To evaluate the photometric efficacy, we measured and compare in Fig. 24 the disk surface brightness profile along its major axis from the 2-roll combined image, to the a priori established profile previously ascertained using the Wedge A+B occulters in GO 12551, and as reported and compared in § 11 of this ISR with BAR5. Despite the very small number (8) of only short (0.1s) exposure time images contributing to the 2-roll combination used for comparison, the major axis radial surface brightness profile beyond the NE and SW effective IWAs, is well reproduced with a precision of a few percent. The BAR10 profile is systematically lower (by a few percent) on the NE (left) side of the disk. This may be due to the residual offsets in target:template positioning w.r.t. the BAR10 corners not fully remediated with quantized scan position matching. BAR10 Corner Rounded-POS TARGs For the BAR5 finger, a target is always optimally placed when located on the mid-line of the bar exactly between the two, parallel, edges that are 0.3" apart. In § 15 of this ISR we give specific POS TARG offsets (from the BAR10 aperture location fiducial) to achieve such a pointing, specifically mid-way along the unbent (but tilted) long axis of BAR5. For each of the BAR10 corners, in placing the target with respect to only a single rounded edge, the observer has a degree of freedom in placing the "back-off" distance from the edge along a 45° diagonal passing through the corner center of curvature based on a trade between inner working angle and contrast at r < 0.5" (see Fig. 18). Here we similarly give specific POS TARGs for the BAR10 LL and LR corners for effective IWAs predicated on the contrast "turn over" as shown in Fig. 18. Because of target placement imprecision, we recommend obtaining observations of both the target and PSF template star with a three-point dithers of 1/4-pixel steps centered on the desired POS TARG along the 45° passing through the rounded corner. Summary and Recommendations 1) Use of the STIS BAR5 coronagraphic occulter, though damaged pre-launch, has been verified and validated as functional to produce high-contrast images in close (≥ 0.15") angular proximity to bright point sources, and for the BAR10 LL and LR "rounded corners" with targetable IWAs in the range ≥ 0.225" -0.300". 2) Target pointing offsets and/or aperture metrology updates as codified in the PDB SIAF.dat file and/or implemented through APT are required to properly position a target centrally on the mid-line of the BAR5 occulter, or with respect to the BAR10 LL or LR corners, to achieve useful/optimum coronagraphic starlight suppression. 3) Item (2) pointing corrections may be implemented in the STScI/HST ground system either as POS TARGs from the existing BAR10 aperture, or by the equivalent update/definition of a BAR5 aperture position (and/or BAR10LL/BAR10LR mnemonics) in SIAF.dat and flowed down to other elements of the ground system. N.B: We specify (2) throughout this ISR as POS TARG offsets from the BAR10 aperture pointing fiducial as propagated into the STScI ground system observation planning S/W systems per the PDB SIAF.dat file circa mid-2013. STScI may elect to update the BAR5 fiducial and/or develop new aperture definitions for the BAR10 LL and LR corners based upon these offsets. If so, observers should be aware not to then additionally apply those offsets if built into updated aperture definitions. 4) Optimum mid-BAR5 pointing (with uncertainty of ± ~ 1/4 pixel) is achieved with: POSTARG (17.48579, -7.42153) from the BAR10 aperture 5) Because of the relative narrowness (0.15" half-width) of the BAR5 occulter, and given the a priori known non-repeatabilities of the STIS MSM that deploys its coronagraphic optics, all BAR5 observations (target and PSF template) would greatly benefit from being executed as a 3-point linear cross-BAR5 "dither" with the central point at a planned mid-BAR position, and flanking positions at +/-1/4 pixel orthogonal to the bar. Such observations are implementable, simply, with the following Phase 2 special pointing requirements: POSTARG 17.48331, -7.43398 from BAR10 aperture; 1/4 pixel below nominal BAR5 mid-line POSTARG 17.48579, -7.42153 from BAR10 aperture; on nominal mid-line of BAR5 POSTARG 17.48827, -7.40908 from BAR10 aperture; 1/4 pixel above nominal BAR5 mid-line	2019-04-13T18:52:24.940Z	2017-07-26T00:00:00.000	{ "year": 2017, "sha1": "f17b83a46a2fcdf3b9f6bb10c144c80c877f3f77", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "a36588bfa163227359913df774d9ae57e1756ea5", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics", "Computer Science" ] }
119481046	pes2o/s2orc	v3-fos-license	Double Band Inversion in $ \alpha $-Sn: Appearance of Topological Surface States and the Role of Orbital Composition The electronic structure of \graySn(001) thin films strained compressively in-plane was studied both experimentally and theoretically. A new topological surface state (TSS) located entirely within the gapless projected bulk bands is revealed by \textit{ab initio}-based tight-binding calculations as well as directly accessed by soft X-ray angle-resolved photoemission. The topological character of this state, which is a surface resonance, is confirmed by unravelling the band inversion and by calculating the topological invariants. In agreement with experiment, electronic structure calculations show the maximum density of states in the subsurface region, while the already established TSS near the Fermi level is strongly localized at the surface. Such varied behavior is explained by the differences in orbital composition between the specific TSS and its associated bulk states, respectively. This provides an orbital protection mechanism for topological states against mixing with the background of bulk bands. Since the theoretical prediction of topological insulators (TIs) [1,2], many materials were found to belong to this new class of solids [3,4]. The interest in these materials is based on the idea to exploit and to manipulate spin-polarized metallic surface states with forbidden backscattering due to time-reversal symmetry. These socalled topological surface states (TSSs) are induced at an interface between systems that belong to different topological phases. The topologically non-trivial low temperature phase of Sn (α-Sn) is a particularly interesting example for such a system. In contrast to other TIs, like the Bi 2 X 3 compounds, in which the inverted bands are separated by a global band gap [5], unstrained α-Sn is a semimetal [6,7] with diamond structure and the Fermi energy (E F ) pinned to the Γ + 8 level (Fig. 1). Spin-orbit coupling (SOC) shifts the p-derived split-off Γ + 7 band below the s-derived Γ − 7 band which in turn is situated below the four-fold degenerate p-derived Γ + 8 band [8,9]. The latter degeneracy can be lifted by applying strain [ Fig. 1(b)]. The key feature of topological materialsthe band inversion -occurs in α-Sn between the Γ + 8 and the Γ − 7 levels at the Γ point of the Brillouin zone (BZ). Importantly, yet another band inversion exists at the Γ point of the BZ within the occupied electronic structure of α-Sn: between Γ − 7 and Γ + 7 bands, which has gone unnoticed in previous studies. Although α-Sn has been predicted to be topologically nontrivial [10], neither the electronic structure nor the existence of surface states was discussed in detail until the experimental observation of a TSS which emerges between the Γ − 7 and Γ + 8 levels and exhibits a Dirac point close to E F [8,11,12]; we refer to this state as TSS1 from now on. Interestingly, although TSS1 is largely degenerate with the surface-projected bulk Γ + 8 band, in angle-resolved photoelectron spectroscopy (ARPES) it is detected as a sharply defined dispersive peak [ Fig. 1(c)]. Moreover, a very recent experimental report on spin-tocharge conversion supposedly involving the TSS1 of α-Sn [12] demonstrates a practical utility of this material for spintronics. The above results raise two questions: what is the mechanism that protects TSS1 from hybridization with the bulk states, and can we expect another TSS (TSS2) induced by the other band inversion? Here we report on a new surface state in the occupied electronic structure of strained α-Sn(001) films, explored in a combined theoretical and experimental approach. Electronic-structure calculations reveal that this state bridges the Γ − 7 and Γ + 7 bands and is located in the subsurface region, thus penetrating deeper into the bulk than TSS1 [8,11]. Bulk-sensitive Soft X-ray ARPES (SX-ARPES) provides a direct experimental evidence for this new subsurface state, which has been overlooked in previous low-photon-energy ARPES experiments with smaller probing depths. Analyzing the orbital characters of bulk and surface states, we show that the different decay character of two surface states is explained by hybridization with the corresponding bulk states. In particular, TSS1's orbital composition suppresses hybridiza-tion with surface-projected bulk states, while for the new surface state such hybridization is favored. Moreover, we review the topological properties of the bulk-band structure by calculating the Z 2 invariant as well as the mirror Chern numbers. We find that the new 'buried' surface state is topologically nontrivial despite being located further away from the topological phase discontinuity (interface TI/vacuum). Thin films of α-Sn were grown in situ on InSb(001) substrates. Prior to deposition, substrates have been cleaned in a series of sputter-anneal cycles, while the surface quality was monitored by low-energy electron diffraction (see Ref. 13). Sn was deposited on the substrates at room temperature, simultaneously Te was deposited to n-dope the film and to provide a smooth growth [8]. The thickness of the α-Sn film was at least ≈ 6 nm (9 unit cells) as estimated from photoemission data [13]. SX-ARPES measurements have been performed at the ADRESS beamline of the Swiss Light Source. Details of the experimental setup are given in Ref. 14. While bulk α-Sn is stable only below ≈ 13.2 • C [15], thin films of α-Sn epitaxially deposited on InSb(001) substrates are stable up to ≈ 200 • C [16], which guarantees successful growth at room temperature. The InSb substrate induces compressive strain in α-Sn due to the lattice mismatch of 0.13 %, the latter causing a local band gap of ≈ 30 meV at the Γ point of the bulk Brillouin zone. Similar to HgTe [17,18], a global band gap in the entire reciprocal space is opened only by tensile in-plane (001) strain, whereas for compressive in-plane (001) strain the local band gap around the bulk Γ point closes along the Γ-Z direction [13,19]. The resulting band crossing gives rise to a Dirac point in the bulk band structure. These properties suggest that compressively strained α-Sn is a Dirac semimetal. We note, that the closing of band gap does not alter the topological character [8,11] because the former occurs be-tween bands whose associated states have even parity [ Fig. 1 To explore the bulk and surface electronic structure of α-Sn we first performed ab initio-based tight binding (TB) calculations for bulk and semi-infinite systems. The Slater-Koster parameters [13,20] for firstand second-nearest neighbors as well as the spin-orbit coupling strength were obtained by optimizing the TB band structure with respect to ab initio data [6,9]. Special attention has been paid to reproducing the correct band ordering near the Γ point. Layer-resolved spectral density near Γ for TSS1 (red and black squares) and TSS2 (black circles). While TSS1 is surfacelocalized, TSS2 is located in the subsurface region, with an LDOS maximum in the sixth DL. Red and black squares show identical but differently scaled data for TSS1. The TB results are presented in Fig. 2 (see also [13]), where we show a depth distribution of the electronic states' wavefunctions. When projected onto a Sn double layer (DL; half of the cubic unit cell) within the bulk [ Fig. 2(a)], the Γ + 8 , Γ − 7 and Γ + 7 bands remain unconnected, in agreement with the bulk electronic structure [6]. On the contrary, when projected onto the DL near the surface [Figs. 2(b) and 2(c)], TSS1 shows up as a consequence of the Γ − 7 / Γ + 8 band inversion, exhibiting the maximum spectral density in the topmost DL. Importantly, a new 'connecting' spectral density located between the Γ − 7 and Γ + 7 bulk bands appears at ≈ −0.8 eV [ Fig. 2(c)]; we refer to this feature as TSS2 in the following. The spectral density of TSS2 is enhanced in the sub-surface region, having an extended maximum at and around the sixth DL, as shown in Fig. 2(d). Although the dispersive behavior may at first glance look like a Rashba split Γ − 7 band, such an explanation can be ruled out, as its overall band width does not change from surface to bulk, whereas a Rashba effect would occur only at the surface and inevitably increase the width of the Γ − 7 band. Based on our slab-calculations [13] a quantum well origin of this state can be excluded as well. As pointed out by Barfuss et al. [8], spin-orbit coupling lifts the degeneracy of the Γ + 7 and Γ + 8 bands at E F and pushes the Γ + 7 band below the Γ − 7 band, thereby introducing another band inversion. Therefore, the so far undiscovered TSS2 exists in an 'inverted' SOC-induced band gap, which suggests a topological origin [ Fig. 1(b)]. This conjecture and, thus, the topological nontrivial band ordering are affirmed by calculations of topological invariants. Originally defined for insulating systems [21], the Z 2 invariant has also been calculated for systems without a global band gap [22,23]. A necessary condition for such calculations is the presence of a finite energy difference at each wavevector k (i. e., a k-dependent band gap). This is the case for α-Sn in which the Γ + 7 band is separated from the Γ − 7 band in the entire bulk BZ, irrespectively whether moderate compressive or tensile strain is applied (strain less than 3 % is assumed). The Z 2 invariant of the bands below the Γ − 7 band was calculated to (ν 0 ; ν 1 ν 2 ν 3 ) = (1; 000) by using the Fu-Kane formula [24] discretized by Fukui and Hatsugai [22] and by tracing maximally localized Wannier functions [25]. Moreover, the lattice is invariant upon reflection at the (110) and the (110) mirror planes; the respective mirror Chern numbers are n M = −1 for both planes (further details in the Supplemental Material [13]). Both topological invariants confirm the non-trivial topology: moderately strained α-Sn is both a strong TI and a topological crystalline insulator. This necessitates that according to the bulk-boundary correspondence [3] a TSS has to bridge ('interconnect') the Γ + 7 and Γ − 7 bands: this state is unambiguously identified as TSS2. The absence of both TSS1 and TSS2 in the bulk electronic structure [ Fig. 2(a)] is clear evidence for their surface character. It is also evident from comparing Fig. 2(b) with Fig. 2(c) that TSS1 has its LDOS maximum at the surface while TSS2 is located well below the surface. The different surface localization is addressed in Fig. 2(d). Due to overlap with projected bulk bands, both TSS1 and TSS2 are strictly speaking so-called 'surface resonances', which are characterized by an enhanced LDOS in the surface or subsurface region being degenerate with a finite LDOS in the bulk. Yet TSS1 decays in general exponentially toward the bulk, thereby being more reminiscent of a surface state; in contrast, the LDOS of TSS2 decays and oscillates, which is more typical of a surface resonance. To probe the subsurface-localized TSS2 experimentally we utilized SX-ARPES which has a higher probing depth compared to low-photon-energy ARPES [26]. Owing to an excellent k ⊥ resolution, the experimental SX-ARPES data acquired with polarized 352 eV and 684 eV photons detect mostly states close to the bulk Γ points [ Fig. 3(a)]; they show clearly all three occupied bulk valence bands, i. e., Γ + 8 , Γ − 7 , and Γ + 7 [ Figs. 3(b) and 3(c)]. While at both photon energies the intensity from TSS1 near E F is very weak, TSS2 located between the surface-projected Γ − 7 and Γ + 7 bulk bands at ≈ −0.8 eV is unequivocally resolved [Fig. 3(e) and 3(f)]; it is strikingly similar to TSS2 in the calculations [ Fig. 2(c)]. The fact that along the surface perpendicular [001] direction TSS2 appears only near bulk Γ points and persists in spectra acquired with high probing depth (hν = 684 eV) clearly indicates its surface resonance character. The apparent broadening of the Γ + 7 band at hν = 352 eV is most likely a consequence of a small yet finite k ⊥ resolution [27]. While TSS1 appears very pronounced between the Γ − 7 and the Γ + 8 bands at photon energies around 20 eV, the intensity of TSS1 is small at soft X-ray photon energies (see Ref. 13). Interestingly, in a study by Liu et al. on HgTe(110) [28], which closely resembles the electronic structure of α-Sn, a counterpart of TSS1 was observed over a wide SX-photon energy range. The small intensity of TSS1 in α-Sn might be better understood by future SX-ARPES photoemission simulations. As depicted in Fig. 3(d), the Γ + 8 band crosses E F , indicating that the film is p-doped despite the Te doping; this may be explained by Sn vacancies or diffusion of In atoms from the substrate into the film [8]. We note that the small negative partial band gap between both Γ + 8 bands described above and in Ref. 13 is likely to occur above E F and is, for the small strain induced through the InSb substrate (0.13 %), below the resolution limit of our experiment. In order to better understand the different spatial distribution of the TSS1 and TSS2, we explore their orbital composition using two distinct polarizations of the incident light (Fig. 4). In the experimental geometry, the measurement plane coincides with a (110) mirror plane (XΓZ). Therefore, for light polarized parallel (even) to the measurement plane (p-polarized light), photoemission from initial states whose wavefunction is antisymmetric (odd) with respect to the (110) mirror plane is suppressed, since the final state of the photoelectron is even [29]. Thus, only states with dominant s, p z , and symmetric combinations of p x and p y orbitals are visible in Fig. 4(a). In turn, with perpendicular polarization (odd or s-polarized light) one detects states with antisymmetric wavefunctions [i. e., antisymmetric combinations of p x and p y orbitals; Fig. 4 very well to the experimental SX-ARPES data and provide a strong hint towards a similar orbital composition of TSS2 and the projected bulk states. Further investigations of the orbital compositions show that the projected bulk states between the Γ + 7 and Γ − 7 levels are mainly composed of p x and p y orbitals but lack p z contributions ( Fig. 4 in Ref. 13). The same orbital composition is found for TSS2, which explains the high degree of its hybridization with the bulk states and its pronounced surface resonance character. On the contrary, TSS1 is mainly composed of p z orbitals; it therefore hybridizes much weaker with the p x and p y bulk states below Γ + 8 , which results in its exponential decay toward the bulk. Recalling that despite the absence of a global band gap, TSS1 has sharp dispersive peak in ARPES (i.e., long photo-hole lifetime), we conclude that the decay channel of the TSS1 photo-hole through the bulk states is reduced due to their different orbital composition and strong surface localization of TSS1. These results verify an "orbital protection mechanism" for such surface resonant TSSs, which can occur in addition to a topological protection from backscattering. Bearing resemblance to Weyl or Dirac semimetals in which both topological bulk and surface states cross the Fermi level [17][18][19][30][31][32], TSSs can appear degenerate with the surface-projected bulk band structure as long as they reside within local energy-and wavevectordependent band gaps [22,23,33,34]. Moreover, such states were observed to appear also in spin-orbit induced partial gaps away from the E F in well-known layered topological insulators, e.g. Sb 2 Te 3 [35,36] and Bi 2 Se 3 [37,38]. However, a pair of surface-resonance TSSs which share the same bulk band and have very different spatial decay character is so far a unique property of α-Sn. In conclusion, the surface and subsurface electronic structure of α-Sn thin films strained compressively in the (001) plane is studied theoretically and experimentally. In agreement with ab initio-based electronic structure calculations, the bulk-sensitive SX-ARPES data allow to reveal an additional topological surface state located between the projected Γ − 7 and Γ + 7 bands, to the best of our knowledge unreported so far [8,11,12]. The nontrivial topology of this surface state is proven by the topological invariants of the associated bulk states. This new TSS of α-Sn in the projected valence bands has a pronounced surface resonance character, in clear contrast to the surface-localized TSS1. Based on the analysis of orbital compositions, the different decay characters of TSS1 and TSS2 are attributed to their hybridization with surface-projected bulk states. Our findings thus reveal a new type of TSS in α-Sn that is coupled by orbital symmetries to the bulk states in the whole energy range. In addition, we find that an "orbital protection mechanism" can effectively protect the surface-resonant TSS against hybridization, as in TSS1 in the present case. Recently, a pair of surface states reminiscent of TSS1 and TSS2 discussed in the present work was also found in half-Heusler compounds whose bulk-band structure is quite similar to those of α-Sn and HgTe [39,40]. The authors of those studies attribute a nontrivial topology to the surface state near E F ; considering our findings one may speculate that the other surface state is of topological origin, too.	2017-02-16T17:03:14.000Z	2017-01-12T00:00:00.000	{ "year": 2017, "sha1": "e941d760105fccb0e0e6cb2e3d6d1f358a0dc183", "oa_license": null, "oa_url": "https://www.dora.lib4ri.ch/psi/islandora/object/psi:4469/datastream/PDF/Rogalev-2017-Double_band_inversion_in_%CE%B1-Sn-(published_version).pdf", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "e941d760105fccb0e0e6cb2e3d6d1f358a0dc183", "s2fieldsofstudy": [ "Physics", "Materials Science" ], "extfieldsofstudy": [ "Physics" ] }
211019294	pes2o/s2orc	v3-fos-license	The neural basis of belief-attribution across the lifespan: False-belief reasoning and the N400 effect The current study examined how social cognition - specifically, belief-state processing - changes across the lifespan, using a large sample (N = 309) of participants aged 10-86 years. Participants completed an event-related brain potential study in which they listened to stories involving a character who held either a true- or false-belief about the location of an object, and then acted in a manner consistent or inconsistent to this belief-state. Analysis of the N400 revealed that when the character held a true-belief, inconsistent outcomes led to a more negative-going N400 waveform than consistent outcomes. In contrast, when the character held a false-belief, consistent outcomes led to a more negative-going N400 waveform than inconsistent outcomes, indicating that participants interpreted the character's actions according to their own more complete knowledge of reality. Importantly, this egocentric bias was not modulated by age in an early time window (200-400 msec post-stimulus onset), meaning that initial processing is grounded in reality, irrespective of age. However, this egocentric effect was correlated with age in a later time window (400-600 msec post-stimulus onset), as older adults continued to consider the story events according to their own knowledge of reality, but younger participants had now switched to accommodate the character's perspective. In a final 600-1000 msec time window, this age modulation was no longer present. Interestingly, results suggested that this extended egocentric processing in older adults was not the result of domain-general cognitive declines, as no significant relationship was found with executive functioning (inhibitory control and working memory). Introduction The ability to understand other people's mental states e in terms of their motivations, beliefs, desires, and knowledge states e plays a critical role in daily life, allowing successful interactions to occur. These 'Theory of Mind' (ToM) abilities are key predictors of successful social outcomes; individuals who report difficulties in their ToM abilities often report impairments in social interactions, both in clinical populations (e.g., Baron-Cohen, Leslie, & Frith, 1985;Couture, Penn, & Robert, 2006;Frith, 2001), and also in typical human aging (e.g., Bailey, Henry, & von Hippel, 2008;Moran, 2013;Sullivan & Ruffman, 2004). Whilst there has been much empirical work on social cognition abilities, it remains unclear how these social cognition capacities are utilized in our daily lives, and, importantly, the extent to which these abilities change or remain stable across the lifespan. The current study examined the electrophysiological basis of tracking other peoples' (false) beliefs and integrating events according to those beliefs (i.e., understanding events from another person's perspective). Specifically, we tested whether and how belief-attribution capacities change across the lifespan in a large sample (N ¼ 309) of participants aged 10e86 years old, and examined whether these changes reflect domain-general declines in cognitive skills (e.g., executive functions), versus domainspecific changes in social cognition. Recent research has shown that the social brain continues to develop throughout adolescence (Blakemore & Mills, 2014;Dumontheil, 2016), and that these structural changes underlie major developmental progressions in social cognition, which interact with improvements in cognitive control (Humphrey & Dumontheil, 2016;Mills, Dumontheil, Speekenbrink, & Blakemore, 2015). At the other end of the age spectrum, older adults experience greater difficulties in engaging ToM abilities compared to younger adults (Bailey & Henry, 200 9;Castelli et al., 2010;German & Hehman, 2006;McKinnon & Moscovitch, 2007). For instance, older adults have been shown to be impaired compared to younger adults in their ability to recognize emotions (Grainger, Henry, Phillips, Vanman, & Allen, 2015), attribute mental states to story characters (Maylor, Moulson, Muncer, & Taylor, 2002), and switch from their own to someone else's perspective (Martin et al., 2019). The social consequences of difficulties with ToM can be severe, limiting social-communicative abilities, which in turn can lead to feelings of loneliness and depression, particularly in later life (e.g., Bailey et al., 2008;Sullivan & Ruffman, 2004). Further, it has been shown that ToM abilities predict social functioning outcomes across the neurotypical human lifespan, highlighting the importance of furthering our understanding of how ToM changes and develops across the lifespan (Apperly, Samson, & Humphreys, 2009;Dumontheil, Apperly, & Blakemore, 2010;Schneider, Slaughter, & Dux, 2017). Whilst prior studies have aimed to localise the underlying mechanisms of ToM capacities using neuroimaging or source localisation methods (e.g., Gallager & Frith, 2003;Kang, Schneider, Schweinberger, & Mitchell, 2018;Vogeley et al., 2001), it remains largely unknown how these social cognitive mechanisms may change across the healthy lifespan, and how this may be reflected in temporally sensitive neuroimaging measures, such as event-related potentials. Results ascertained from healthy, neurotypical adults can significantly further our understanding of ToM processes, including advancing our understanding of the neural changes associated with ToM processes across the lifespan, and how social changes during different stages of life may impact e or be impacted by e ToM engagement. Apperly and Butterfill (2009) proposed a distinction between explicit and implicit ToM, where implicit ToM refers to automatic tracking of an agent's perspective, and explicit ToM refers to deliberate consideration of an agent's mental states. In other words, explicit ToM involves an individual purposefully and deliberately reasoning about others' beliefs, whereas implicit ToM, which is argued to exist alongside the explicit mental state tracking system, involves processing of another person's mind without a conscious, explicit intention to do so (e.g., Schneider et al., 2017;Back & Apperly, 2010). Neuroimaging research has revealed substantial overlap in brain areas involved in implicit and explicit ToM (e.g., Naughtin et al., 2017), though differences in performance have been observed in these two domains. When examining explicit ToM abilities, tasks tend to require responses that demonstrate an overall understanding of different ToM scenarios (e.g., yes/no answers; explicit verbal/pointing responses). Prior studies have shown that older adults perform consistently worse than younger adults on these explicit measures, indicating a decline in explicit ToM abilities with advancing age (see Henry, Phillips, Ruffman, & Bailey, 2013 for a meta-analysis). More recently however, research has begun to use more sensitive measures, such as eye-tracking, response times, and electrophysiological methods (e.g., event-related potentials, ERPs), to examine the implicit cognitive processes underlying ToM engagement in adults (e.g., Ferguson, Scheepers, & Sanford, 2010;Schneider, Bayliss, Becker, & Dux, 2012). This research has demonstrated that, despite declines in older age seen in explicit ToM abilities, some aspects of implicit ToM processing may remain intact in older age. For instance, Grainger, Henry, Naughtin, Comino, and Dux (2018) conducted a study in which younger and older adults watched a series of videos where true-or false-belief states of characters were manipulated, whilst eye movements were recorded. Analysis of anticipatory looking behaviour revealed similar success levels for implicit falsebelief processing in the two groups, despite older adults showing a deficit compared to the younger adults in explicit ToM processing abilities (assessed using an emotionrecognition task). This distinction between implicit and explicit ToM task performance is further demonstrated by developmental research that has shown advanced ToM skills are present in explicit measures from around 5e7 years old (Baron-Cohen, Joliffe, Mortimore, & Robertson, 1997;Brent, Rios, Happ e, & Charman, 2004;Wellman, Cross, & Watson, 2001), but that even healthy adults make errors when judging the perspective of another person, suggesting a resilient egocentric bias/'pull of reality' in processing of mental states during ToM use (Birch & Bloom, 2007;Bradford, Gomez, & Jentzsch, 2018;Ferguson, Apperly, Ahmad, Bindemann, & Cane, 2015;Ferguson & Breheny, 2011Keysar, Barr, Balin, & Brauner, 2000). However, other research has suggested that, at least at an implicit level, another person's perspective is automatically processed (e.g., Kov acs, T egl as, & Endress, 2010;Samson, Apperly, Braithwaite, Andrews, & Bodley Scott, 2010;Schneider et al., 2017). For example, Surtees and Apperly (2012) found that, in a visual perspective taking task in which an avatar could see either the same or a different number of dots as the participant in a virtual room, both children and adults were slower to respond when the avatar's view differed from the participant's, even when this was irrelevant to the task (i.e., when asked to respond from the 'self' perspective, so that the avatar's perspective could be ignored). These results were consistent with findings from Samson et al. (2010) and Qureshi, Apperly, and Samson (2010) who report that participants' judgments about their own perspectives were both slower and more error prone when an avatar in the task had a different perspective. This research indicates a differentiation between implicit ToM e where the mere presence of a social agent can slow down an individual's responses, indicating automatic processing of other people's perspectives e and explicit ToM, where adults are shown to make egocentric errors, suggesting a failure to engage in ToM abilities. Alongside eye-tracking measures, recent research has also utilized ERPs to examine the neural substrates of ToM processing, either using tasks that require participants to make an explicit judgment about a belief-state of a story character (e.g., Liu, Sabbagh, Gehring, & Wellman, 2004;Sabbagh & Taylor, 2000;Zhang, Sha, Zheng, Ouyang, & Li, 2009) or tasks that require participants to observe pictorial depictions of different belief-states, providing a more implicit measure of ToM processing (e.g., Kü hn-Popp, Sodian, Sommer, D€ ohnel, & Meinhardt, 2013;Meinhardt, Kü hn-Popp, Sommer, & Sodian, 2012). When participants were explicitly prompted to make belief inferences this elicited a frontally distributed late slow wave (LSW) (Liu et al., 2004;Sabbagh & Taylor, 2000), which was similar to the more widely distributed LSW that was elicited when belief inferences were implicit. These overlapping brain responses indicate that people continue to consider other's beliefs, even when they were not directly prompted to track the other person's mental state. Further ERP evidence for implicit ToM processing has been found among young adults by examining modulations of the N400 effect. The N400 is a centro-parietal component that is sensitive to stimulus predictability and semantic integration processes during language comprehension, showing more negative-going amplitudes when an anomaly is detected (Kutas & Hillyard, 1980;Lau, Holcomb, & Kuperberg, 2013;Nieuwland et al., 2018;Westley, Kohú t, & Rueschemeyer, 2017). In Ferguson, Cane, Douchkov, and Wright (2015)'s study, participants read a series of short narratives in which a character held a true or false belief about the location of an object, and subsequently acted in a manner that was consistent or inconsistent with this belief state. ERPs were timelocked to the visual onset of the sentence-final critical word (i.e., the object's location). For instance, if a character witnessed an object being moved (true-belief) from Location A to Location B, the character should look for the object in Location B (consistent outcome, where both the character and they know it to be located), rather than Location A (inconsistent outcome). In contrast, if the character did not witness the object being moved (false-belief), the character should look for the object in the original Location A (consistent outcome, even though the object is no longer there) and not Location B (inconsistent outcome, even though this is the true object location). Results showed a classic N400 effect in true-belief scenarios, whereby inconsistent actions elicited a more negative-going N400 than consistent actions. However, this pattern was reversed for false-belief scenarios, as consistent trials elicited a more negative-going N400 than inconsistent trials. This indicated that, when reading the stories, participants were not automatically integrating events according to the character's beliefs, but rather were processing the stories from their own egocentric perspective (i.e., their own knowledge of reality). Interestingly, contrasting effects have been found among adolescent participants (aged 10e15 years) who completed a similar reading paradigm, while another person was either present or absent in the room (Westley et al., 2017). In Westley et al.'s study, adolescent participants read a story presented on a computer screen in the presence of a confederate. The start of each story was presented to just the adolescent, with the confederate in the room but unable to see the computer screen, creating different knowledge states for the participant and the confederate. The final sentence of the story presented either a plausible or implausible outcome, and was seen by both the participant and the confederate. Results of this joint comprehension task showed a significant N400 effect (i.e., more negative-going N400 effect following implausible outcomes vs plausible sentence outcomes) when the final sentence was implausible for the confederate who had not read the full story, even though it was plausible for the fully informed participant. These results suggest that the adolescent participants were automatically simulating the perspective of the other person during the task, and considered this more limited perspective when integrating events in the stories, even though events in the story made sense from their own perspective. Though there are clear methodological differences between Ferguson et al.'s (2015) and Westley et al.'s studies (in terms of language content, and social environment), it is puzzling that adolescents should have better awareness of others' perspectives compared to adults. These mixed results therefore highlight the importance of examining implicit mental state attribution capacities across the lifespan in a single task, which will allow novel insights into how these abilities and underlying neural mechanisms may change at different ages. Understanding of the processes underlying implicit mental state attribution is important, allowing further insight into potential underlying causes of age-related difficulties in ToM, which in turn can have important functional outcomes (Grainger et al., 2018;Henry, Von Hippel, Molenberghs, Lee, & Sachdev, 2016). Critically, it is important to further our understanding of how these processes may change across normal aging, including during adolescence, and younger, middle, and older age, as well as building on our knowledge of what may be driving these changes. For instance, there is significant evidence of a link between ToM and executive functioning abilities, particularly inhibitory control (Bradford, Jentzsch, & Gomez, 2015;Carlson & Moses, 2001;Perner & Lang, 2000) and working memory ( 1996;Lin, Keysar, & Epley, 2010). These links make sense given that successful social cognition requires one to hold in mind multiple perspectives (i.e., working memory), and suppress irrelevant perspectives (i.e., inhibitory control). Moreover, there is now robust evidence that these cognitive abilities decline with advancing age (Elderkin-Thompson, Ballmaier, Hellemann, Pham, & Kumar, 2008;Salthouse, Atkinson, & Berish, 2003), with emerging evidence suggesting that some aspects of this age-related cognitive decline are manifest from 20e30 years old, and decreases at a faster rate with increasing age (Klindt, Devaine, & Daunizeau, 2017;Salthouse, 2009;Singh-Manoux et al., 2012). Taken together, these findings raise the question of whether changes in ToM abilities (such as increases in egocentric bias during ToM engagement) are driven by domain-general cognitive factors, such as executive function abilities, or whether they are specifically related to changes in social cognition abilities? The current study adapted the false-belief ERP task in Ferguson, Apperly, et al. (2015) and to the auditory domain, examining the N400 as a marker of belief integration, in an age sample of participants spanning adolescence (10 years) to older age (86 years). We examined effects over early and late phases of the N400 (200e400 msec and 400e600 msec), based on previous observations that auditory N400s tend to begin earlier and last longer than N400s elicited in the visual modality (Kutas & Federmeier, 2011;Kutas & Van Petten, 1994), as well as in a later 'wrap-up' period (600e100 msec). Moreover, this early/ late analysis of the N400 allowed us to examine whether an age-related temporal delay exists in narrative-based ToM processing, following research showing that typical N400 semantic congruity effects at the scalp gets smaller, slower and more variable with age (Kutas & Iragui, 1998). Participants also completed measures of inhibitory control and working memory, to provide measures of general cognitive skills, allowing analysis of whether any age-related changes in ToM abilities are due to declines in social skills per se, or whether they are due to general changes in cognitive domains. By including both adolescents e a period of rapid, significant cognitive and social development e as well as middle-aged and older individuals, the current study includes groups of participants that have, to-date, not been at the forefront of research into ToM processes, allowing us to further our understanding of how and when implicit mental state attribution processes change across the lifespan. In line with the study aims, there were three key predictions: First, we expected to replicate Ferguson et al.'s (2015) findings in a large sample of participants, and in the auditory modality. That is, a typical N400 effect was expected in truebelief conditions, but a reversed N400 effect was expected in false-belief conditions. Second, if social cognitive skills undergo general declines with advancing age, then we predicted that this reversed N400 effect in false-belief conditions would be larger with increasing age. Alternatively, if implicit social skills remain intact in older age, we would not expect age to correlate with N400 effects in false-belief trials, indicating that all participants experience the same degree of egocentric interference on these trials. Finally, we expected to see cognitive skills (i.e., working memory and inhibitory control) declining with advancing age, and it was predicted that these capacities would correlate with the consistency effects in false-belief story conditions. Method We report how we determined our sample size, all data exclusions (if any), all inclusion/exclusion criteria, whether inclusion/exclusion criteria were established prior to data analysis, all manipulations, and all measures in the study. Participants A total of 339 participants (112 males) took part in this study as part of a larger task battery. Of this full sample, 20 participants did not have EEG data (due to computer difficulties or participant declining to complete the task), and a further 10 participants were excluded from analysis due to noisy EEG recordings (retaining less than 60% segments in one or more of the critical test conditions after pre-processing). This resulted in a final sample of 309 participants (101 males , and household income and occupational class were coded on a scale of 1e7. These three scores were summed to derive an SES index, with lower scores indicating lower SES. All participants were native English speakers, had normal or corrected-to-normal vision, had no known neurological disorders, and no mental health or autism spectrum disorder diagnoses. Participants were recruited from a community sample in the local area of Kent, U.K., using a variety of recruitment strategies (e.g., newspaper adverts, local groups, word-of-mouth, Kent Child Development Unit). Sample size was determined a-priori during grant application, allowing a large data sample across the lifespan to be collected. The Ethical Committee of the School of Psychology, University of Kent, U.K., approved the study. Materials and design The tasks described below were programmed using E-Prime software. False-belief ERP task The false-belief ERP task was modified from Ferguson, Apperly, et al. (2015) and . Here, verbal stimuli were presented to participants auditorily through headphones, rather than visually, which allowed us to control for different reading times among the different aged participants. The task involved 120 experimental items (see Table 2 for examples; full experimental stimuli are available at: https://osf.io/pw7h6/ https://osf.io/pw7h6/). Each item was constructed of three sentences, depicting classic change of location true/false belief scenarios. The first sentence introduced a character and described the character placing a target object in a specific location. The second sentence then described a second character moving the target object to a new location. Importantly, this action was either observed (resulting in a true-belief about the target object's location) or not observed (i.e., not seen, resulting in a false-belief about the target object's location) by the first character. The third, final, sentence described the first character looking for the target object, referring to them looking in a location that was either consistent or inconsistent with their true/false belief state. This resulted in a 2 (belief: true vs. false) Â 2 (consistency: consistent vs. inconsistent) within-subjects design. There were four presentation lists of experimental stimuli, with each list containing 120 experimental items, 30 in each of the four conditions, using a Latin square design. In addition, there were 65 unrelated filler stories (approximately the same length as experimental items, but did not depict beliefs) that were randomly interspersed among the experimental items. Participants were randomly assigned to read one of the four test lists, meaning that each participant heard each target sentence in one of the four conditions (i.e., they did not hear the same story more than once). In the task, each trial began with the presentation of a single centrally located red fixation cross shown for 500 msec to signal the start of a new trial. After this time, the fixation cross turned black (remaining on screen for the duration of the trial), and the audio stimuli began following a 500 msec delay. Target words were always the final word of the third sentence. Event related potentials for analysis were timelocked to the auditory onset of the target word. After each sentence, there was a blank screen for 1000 msec before the next trial began. One third of the experimental stories (i.e., 10 in each condition) and 27 of the filler stories were followed by a question about how appropriate the target character's actions were, and participants were prompted to respond using a Likert-Type scale from 1 ¼ very appropriate to 5 ¼ very inappropriate. These questions were included as a check that participants were paying attention to the stories, and to ensure they interpreted events appropriately according to the character's perspective (i.e., an inconsistent action from the character should be rated as 'very inappropriate'). There were eight practice trials to familiarize participants with the procedure, three of which had an example 'appropriateness' question after them. The main experimental items were presented in a random order in 10 blocks of 12 stories. Each block was separated by a break, the duration of which was determined by the participant. Stroop task A classic Stroop word-based task (Stroop, 1935) was used to measure inhibitory control. Participants were told they would see a word on the computer screen, and were required to respond to the ink colour of the presented word as quickly and as accurately as possible. Responses were made using a button box with four colour buttons e red, green, blue, and yellow. Participants first completed 20 practice trials, including 10 neutral and 10 congruent trials in pseudo-randomised order. For the experimental trials, participants were presented with words printed in red, green, blue, or yellow ink, shown on a grey background. There were 50 congruent trials (e.g., "RED" printed in red ink), 50 incongruent trials (e.g., "RED" printed in blue ink), and 50 neutral trials (i.e., a neutral word such as "MEET"), presented in pseudo-randomised c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 order, in which the same colour word, printed colour, or the same colour word/printed colour pair did not appear on two consecutive trials. A blank screen appeared for 1000 ms at the start of the experimental trials, and the next trial was started immediately after the participant made a response. For analysis, response times for congruent and incongruent trials were log-transformed to adjust for age-related slowing; the dependent variable used for analysis was the Stroop congruency effect (i.e., log-transformed incongruent trial mean response time minus log-transformed congruent trial mean response time). In the task, participants were asked to solve maths equations whilst concurrently remembering letter sequences. During the task, a maths equation appeared on screen (e.g., (2 Â 2)þ 1 ¼ 3); participants needed to indicate whether the answer presented was correct or incorrect, and there was no time constraint for this response. After the maths equation, a single letter was presented; this was repeated two to seven times (i.e., a minimum of two letters to remember, a maximum of seven letters to remember, in a given trial). Letters were shown for 800 msec. Trials were presented in randomized order, with three trials for each level. This resulted in 18 trials total, including 81 maths problems and 81 letters. At the end of each trial, a 4 Â 3 matrix was displayed on the screen, from which participants were required to indicate the letters they had seen, in the correct order, by clicking on a box next to the appropriate letter. There was a 'blank' button available to select if participants could not remember the letter in the sequence. There was no time limit for completing this matrix. Participants were given feedback after each trial detailing their word recall accuracy and percentage correct for the maths equations. Participants were encouraged to keep their maths accuracy score at or above 85% throughout the task. The dependent variable used for analysis was a partial OSpan score, calculated as the total number of letters recalled in the correct position. 2.2.3.1. PROCEDURE. The tasks were completed as part of a larger task battery involved in the 'CogSoCoAGE' study which lasted approximately 5 h in total. No part of the study procedures or analyses were pre-registered in a time-stamped, institutional registry prior to the research being conducted. Participants completed the tasks over one or two days, and tasks were administered in a counterbalanced order. Electrophysiological measures A Brain Vision Quickamp amplifier system was used with an ActiCap cap for continuous recording of EEG activity from 30 electrodes, referenced to FCz. Vertical electro-oculogram (VEOG) activity was recorded from one extra electrode (below right eye), and horizontal electro-oculogram (HEOG) activity was recorded from one extra electrode (to the left of the left eye). EEG and EOG recordings were sampled at 1000 Hz, and electrode impedance was kept below 10 kU. Prior to segmentation offline, a vertical ocular calculation was applied (1Fp2þ(À1VEOG)). All data were re-referenced to a left/right mastoid average reference. EEG and EOG activity were bandpass filtered (.1e30 Hz, notch filter at 50 Hz). The raw data were visually inspected for noisy sections or channels, and for other general artifacts. EEG activity containing blinks or horizontal eye movements was corrected using a semi-automatic ocular ICA correction approach (Brain Vision Analyzer, 2.1). An average of four ICA components were removed per participant. Semi-automatic artifact detection software (Brain Vision Analyzer 2) was run, to identify and discard trials with non-ocular artifacts (drifts, channel blockings, EEG activity exceeding ± 75 mV). This procedure resulted in an average trial-loss of 10% per condition. Ten participants were removed from analysis due to less than 60% segments remaining in one or more of the critical test conditions after pre-processing, resulting in a final sample size for analysis of 309 participants (101 males). ERP data analysis For analysis of scalp-based ERP effects, the signal at each electrode site was averaged separately for each participant and experimental condition, time-locked to the onset of the target word. Waveforms were aligned to a 200 msec baseline prior to target word onset. Mean ERP amplitudes were determined across three time windows (200e400 msec, 400e600 msec, and 600e1000 msec), which allowed us to examine how the detection of belief-inconsistencies evolves over different phases of the N400, and a later 'wrap-up' period (reflecting processing at sentence conclusion), and whether effects in these time periods are differentially modulated by age. To assess ERP amplitudes over lateral electrodes four regions of interests (ROIs) were used, dividing electrodes along a left/right dimension and an anterior/posterior dimension, as in Ferguson, Apperly, et al. (2015) and . The two ROIs over the left hemisphere were: leftanterior (F7, F3, FC5, FC1), and left-posterior (CP1, CP5, P7, P3, O1); two homologue ROIs were defined for the right hemisphere. ERP amplitudes over midline electrodes (Fz, Cz, Pz, Oz), where the N400 is maximal, were analysed separately from lateral electrode sites. To analyse the N400 in each experimental condition, a repeated-measures ANOVA was performed over lateral electrodes in a 2 (Belief: true vs false) Â 2 (Consistency: consistent vs. inconsistent) Â 2 (Hemisphere: left vs right) Â 2 (Ant/Pos: anterior vs posterior) design. For midline electrodes, a 2 (Belief: true vs. false) Â 2 (Consistency: consistency vs. inconsistent) Â 4 (Electrode: Fz, Cz, Pz, Oz) repeated-measures ANOVA was used. Appropriateness ratings First, participants' judgements of appropriateness were analyzed to examine whether participants explicitly detected belief inappropriate behaviors in the stories. A 2 (Belief: true vs false) Â 2 (Consistency: consistent vs inconsistent) repeatedmeasures ANOVA was run on appropriateness ratings (1 ¼ very appropriate, 5 ¼ very inappropriate). N400 analyses Full statistical effects for the three time windows (200e400 msec, 400e600 msec, and 600e1000 msec), over lateral and midline electrode sites are summarised in Table 3; full data for each measure is available on the Open Science Framework (https://osf.io/pw7h6/). Note that due to space constraints, only significant effects that involve the Belief or Consistency variables are presented in the text. Fig. 1 shows the topographical maps of ERP difference waveforms for the consistency effect in each belief condition, shown separately for participants in the four age groups described in Table 1 for illustration purposes. Fig. 2 shows the grand average ERP waveforms over three midline electrodes (Fz, Cz, and Pz). 200e400 msec time window Analysis of lateral electrodes revealed the expected significant interaction between Belief and Consistency. There were also significant interactions between BeliefHemisphere, ConsistencyHemisphere, and BeliefConsistencyHemisphere, which were further subsumed under a four-way interaction between Belief, Consistency, Hemisphere, and Ant/Post. To explore this four-way interaction, post-hoc analyses was conducted using a 2 (Belief) Â 2 (Consistency) Â 2 ( c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 Where this BeliefConsistency interaction was significant, it reflected the classic N400 for true-belief stories (i.e., a more negative-going wave-form for inconsistent vs consistent; all ts > 2.98, ps < .003), but a reversed N400 effect for false-belief stories (i.e., a more negative-going wave-form for consistent vs inconsistent; all ts > 3.23, ps < .005). Over midline electrodes, there were significant interactions between BeliefConsistency, and ConsistencyElectrodes, both of which were further subsumed under a three-way interaction between Belief, Consistency, and Electrode. Post-hoc analyses, using Bonferroni corrections to control for multiple comparisons, leading to a new significance criterion of .0125, revealed that the BeliefConsistency interaction was significant at Fz In summary, results in the 200e400 msec time window supported our first hypothesis, showing a typical N400 effect in true-belief conditions (i.e., more negative-going waveforms for inconsistent vs consistent story outcomes), and a reversed N400 effect in false-belief conditions (i.e., more negative-going waveforms for consistent vs inconsistent story outcomes), indicating egocentric processing of the stories in false-belief scenarios. 400e600 msec time window Analysis of lateral electrodes revealed a significant main effect of Consistency, which was further modulated by Hemisphere. This ConsistencyHemisphere interaction revealed more negative-going N400 amplitudes for consistent versus inconsistent conditions in the left hemisphere [t(308) ¼ 4.85, p < .001], but no significant difference in the right hemisphere [t(308) ¼ .60, p ¼ .55]. There was also a significant interaction between Belief and Hemisphere, however, none of the post-hoc comparisons reached significance (all ts < 1.7). More importantly, the three-way interaction between Belief, Consistency, and Ant/Post was significant. Post-hoc analyses using Bonferroni corrections to control for multiple comparisons, leading to a new significance criterion of .025, showed that the BeliefConsistency interaction was significant for both anterior Over electrodes Fz and Cz, the interaction showed that the N400 was more negative-going for consistent versus inconsistent in false-belief stories (ts > 4.61, ps < .001), but did not differ in true-belief stories (ts < .72 ps > .47). In summary, results in the 400e600 msec time window showed a reversed N400 effect in false-belief conditions across anterior electrodes (i.e., more negative-going waveforms for Fig. 1 e Topographic maps of the ERP difference waveform for each belief context condition (inconsistent minus consistent), between 200 and 1000 msec from critical word onset, shown separately for participants in the four age groups described in Table 1 for illustration purposes. c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 Fig. 2 e Grand average ERPs over midline electrodes elicited by critical words in the target sentence for each of the four conditions. Note that negativity is plotted upwards. c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 consistent vs inconsistent story outcomes), indicating egocentric processing of the stories in false-belief scenarios. In true-belief contexts, the typical N400 effect was now reversed so that the N400 was more negative-going for consistent versus inconsistent stories over posterior electrodes. 600e1000 msec time window Analysis of lateral electrodes revealed a significant main effect of Consistency, which was further modulated by Hemisphere. This ConsistencyHemisphere interaction revealed more negative-going N400 amplitudes for consistent versus inconsistent conditions in the left hemisphere [t(308) ¼ À4.90, p < .001], but no significant difference in the right hemisphere [t(308) ¼ .359, p ¼ .72]. There was also a significant interaction between Consistency and Ant/Post; this interaction reflected more negative-going amplitudes for consistent versus inconsistent conditions in anterior regions [t(308) ¼ À3.52, p < .001], but no significant difference in the posterior regions [t(308) ¼ À. Midline electrode analysis showed a significant main effect of Consistency (consistent < inconsistent), as well as significant interaction effects between BeliefElectrode and ConsistencyElectrode, which were further subsumed in a threeway interaction between Belief, Consistency, and Electrode. Post-hoc analyses, using Bonferroni corrections to control for multiple comparisons, leading to a new significance criterion of .0125, revealed that the BeliefConsistency interaction was significant at electrodes Fz [F(1,308) ¼ 9. In summary, results in the 600e1000 msec time window showed a reversed N400 effect in false-belief conditions across anterior electrodes (i.e., more negative-going waveforms for consistent vs inconsistent story outcomes), indicating continued egocentric processing of the false-belief scenarios. In true-belief stories, the N400 was again more negative-going for consistent versus inconsistent stories across posterior electrodes. Correlations Next, we investigated whether and how age influenced the detection of story character's inconsistent behaviors, within true-belief and false-belief contexts. 2 In particular, we aimed to examine whether the initial egocentric interpretation of story events (i.e., the reversed N400 effect within false-belief stories) was driven by, or enhanced in, the older participants in our sample, reflecting an age-related decline in ToM ability. In addition, given the robust evidence for age-related decline in cognitive skills that have been associated with false belief reasoning (i.e., working memory and inhibitory control), we examined whether individual differences in these skills predicted the detection of inconsistencies in the story comprehension task. To facilitate these analyses, a 'consistency effect' variable was calculated for each individual, separately for true-belief and false-belief conditions. N400 amplitudes were pooled over posterior electrode sites (CP5, CP1, CP2, CP6, P7, P3, P4, P8, O1, O2), and difference scores were calculated by subtracting the mean N400 amplitude in the consistent condition from the inconsistent condition, separately for the three time-windows (200e400 msec, 400e600 msec, 600e1000 msec). Here, a negative score indicates a larger N400 effect for the inconsistent compared to consistent condition (i.e., appropriate anomaly detection), and a more positive score indicates a larger effect for the consistent compared to inconsistent condition (i.e., interpreting events egocentrically in false belief contexts). Individual differences in working memory capacity and inhibitory control were assessed using the partial OSpan score and the Stroop congruency effect, respectively. Note that due to computer error or participant refusal to complete task, data was missing for eight participants on the OSpan task, and three from the Stroop task. Full statistical effects for these correlation analyses are presented in Table 4. Fig. 3 shows correlations with age and the consistency effect in the three N400 time windows. Analysis of the posterior electrodes showed that in the 200e400 msec N400 time window, age did not significantly correlate with the inconsistency effect in either true-belief or false-belief contexts. Importantly, this suggests that across all ages, participants initially anchored comprehension of falsebelief stories to their own egocentric perspective rather than to the character's more limited knowledge, and that this egocentric tendency did not increase with advancing age. In the 400e600 msec time window, age did not significantly correlate with the consistency effect in true-belief trials, however results revealed a significant positive correlation between age and the consistency effect in false-belief trials 3 2 Please see Supplementary Materials on OSF (https://osf.io/ pw7h6/?view_only¼9bb79f149f02419285ccaf8c8b89caf3) for details of the main ANOVA analyses run separately for the four age groups described in Table 1. 3 Note that follow-up stepwise multiple regression analysis showed that the linear model best described the data [F(1,307) ¼ 4.75, p ¼ .030], and adding the quadratic coefficient (that might best fit a pattern where belief reasoning skills continue to develop through adolescence, reaching a peak at some point in adulthood, before declining in older age) did not significantly change the R 2 from the linear model [F(1,306) c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 Here, the size of the consistency effect increased with increasing age, which suggests that false-belief events were more likely to be interpreted from an egocentric perspective in older age. For the 600e1000 msec time window, there were no significant correlations with age for the consistency effect in either true-or false-belief trials. Table 4 e Correlations (N ¼ 309) between age and the N400 consistency effect (posterior region) in three time-windows (200e400 msec, 400e600 msec, and 600e1000 msec), across true/false belief contents, and executive function abilities (working memory [Ospan] and inhibition [Stroop] Fig. 3 e Correlations between age (years) and the N400 consistency effect (inconsistent minus consistent), plotted separately for ERP effects in the three time windows (200e400 msec, 400e600 msec, and 600e1000 msec) across posterior regions. Data for true-belief stories are shown with the letter 'T' in the upper panels, and data for false-belief stories are shown with the letter 'F' in the lower panels. Discussion The ability to understand another person's perspective is an important part of daily life, requiring rapid and efficient computation of how another person's perspective may differ from one's own current perspective. In this paper, ERP measures were used to assess implicit ToM abilities in a beliefstate manipulation task. We examined effects of expectancy during narrative comprehension on the established N400 effect, comparing neural responses to belief-consistent versus belief-inconsistent outcomes in true and false belief contexts. The key aim of the study was to examine whether and how implicit ToM changes across the lifespan (i.e., whether the extent of 'egocentric bias' changes across different ages, or whether it remains stable from adolescence through to older age). A unique feature of our study is the large sample (N ¼ 309) of individuals, aged between 10 and 86 years old, which allowed us to track changes across an extended period of development. In addition, we explored whether any agerelated changes in ToM capacities can be explained by domain-general changes in cognitive abilities (specifically working memory and inhibitory control) as opposed to domain-specific changes in social-cognitive abilities. Our results first established that people can successfully track a story character's beliefs in true-and false-belief contexts (i.e., correctly rating story outcomes as appropriate vs inappropriate), indicating explicit understanding of falsebelief scenarios across the lifespan. Following this, we examined implicit belief processing by analyzing N400 responses to different task conditions (belief-type and consistency) across the full sample to establish whether participants were sensitive to violations of expectation from the perspective of a story character, or were biased to their own situational knowledge (i.e., how egocentric was the processing of the stories). This analysis was conducted across three time windows e 200e400 msec, 400e600 msec, and 600e1000 msec e to allow examination of the timings of these effects and whether effects in these time periods are modulated by age. Finally, an N400 consistency effect was calculated for trueand false-belief contexts separately, to assess whether individual differences among participants e specifically, inhibitory control and working memory abilities e modulates their ability to successfully consider another person's perspective, and how this ability changes with age. Across all participants, results showed that when a story character was in possession of a true-belief about an object's location, the N400 was more negative-going for belief-inconsistent than for belief-consistent outcomes, as is typical for semantic and discourse anomalies (Camblin, Gordon, & Swaab, 2007;Nieuwland et al., 2019;Van Berkum, Hagoort, & Brown, 1999). In the true-belief context, this pattern was attenuated or reversed in later time windows, likely reflecting a tendency for participants to reprocess the critical word given the uncertain context of the experiment (i.e.,~33% of items included a violation). Critically, and in line with our first hypothesis, overall analysis of false-belief trials showed the opposite pattern of results. That is, in false-belief contexts, belief-consistent outcomes led to more negative-going N400 effects than belief-inconsistent outcomes, suggesting that participants were processing the stories from an egocentric perspective. These results replicate the pattern seen in Ferguson, Apperly, et al. (2015) and and indicate that when listening to the stories, participants do not immediately integrate the character's beliefs, but instead initially anchor understanding to their own more informed knowledge of the world. These findings fit with previous research that has indicated that when engaging ToM capacities, events are often processed using an initial egocentric bias (e.g., Birch & Bloom, 2007;Bradford et al., 2018;Ferguson & Breheny, 2012;Ferguson, Apperly, et al., 2015;Keysar et al., 2000), despite explicit measures of ToM (such as the appropriateness ratings in the current task) highlighting that participants are able to ultimately overcome this initial bias, and successfully consider the wider context of the story to identify whether a character's actions made sense or not. This study makes further important contributions to the literature by examining how implicit ToM abilities mature and develop across the lifespan, from adolescence to healthy old age. Correlation analyses revealed no significant relationship between age and the consistency effect in the early time window (200e400 msec after stimulus onset) for either true-or false-belief conditions, indicating that during this period, all participants e regardless of age e showed an egocentric bias in processing of the stories presented. This is important, because it shows that egocentric processing is the default perspective for information integration, and that this spontaneous 'pull of reality' effect does not increase with advancing age (i.e., goes against the suggestion that belief processing gets worse in older age). However, in the 400e600 msec time window, age was found to influence the magnitude of the consistency effect in false-belief (but not true-belief) contexts; as age increased the N400 consistency effect also increased. Although this effect was relatively weak (r ¼ .123) and thus requires further research to more firmly establish the magnitude of this finding, the result provides preliminary evidence that the tendency to interpret story events egocentrically, rather than according to the story character's more limited beliefs, increases with advancing age. In other words, older adults' neural activity continues to reflect an interpretation of the story events according to their own knowledge of reality rather than the character's false belief (i.e., a reversed N400, reflecting more negative-going N400 for consistent vs inconsistent), while younger participants had now switched to accommodate the character's actions according to that character's own (false) belief state (i.e., a classic N400, reflecting more negativegoing N400 for inconsistent vs consistent). By the 600e1000 msec 'wrap-up' window, this age modulation was no longer present. We interpret these results cautiously, given the magnitude of the correlation in the 400e600 msec time window, and the challenge of linking cognitive outcomes to that of c o r t e x 1 2 6 ( 2 0 2 0 ) 2 6 5 e2 8 0 neural data. Taken together, however, the results tentatively suggest that whilst all participants initially processed events in the stories egocentrically, anchoring comprehension to the 'self' perspective, younger participants more quickly switched to integrate events according to the story character's perspective compared to older adults, who continued to experience interference from their own egocentric perspective until later in the trials. It is noted that, in using a cross-sectional design, such as the current study, it is difficult to ascertain to what extent these changes are due to age-related differences e for this, longitudinal analysis would be required, assessing participants across different time points. However, within the scope of this study design, our data provides novel evidence that social cognitive skills do undergo changes with advancing age, although even younger participants experience some initial difficulty adopting others' mental states, originally processing the stories from an egocentric perspective. Future research using a longitudinal sample would be valuable for teasing apart the role of ageing, within-person changes, as opposed to cross-sectional differences (e.g., Lindenberger, von Oertzen, Ghisletta, & Hertzog, 2011). It is important to note that we have interpreted our ERP effects within the constraints of the N400 component, and the established cognitive processes that modulate it, since our apriori predictions were based on this. However, we note that the topography of some of our effects is uncharacteristic of typical N400 effects (i.e., the consistency effect in false-belief contexts was maximal over fronto-central rather than centro-parietal scalp sites). It is possible that this unusual pattern was driven by the auditory modality in which the true/ false belief stories were presented (it has been shown that auditory language elicits a slightly more frontal N400 than language presented in the visual modality; Kutas & Federmeier, 2011;Kutas & Van Petten, 1994). Alternatively, it could reflect a different component all together. Specifically, prior research has observed a later occurring 'semantic P600' component that is elicited when participants are detecting 'semantic reversal anomalies', or pragmatic violations and ambiguities (e.g., Kim & Osterhout, 2005;Osterhout & Holcomb, 1992;Schacht, Sommer, Shmuilovich, Martienz, & Martin-Loeches, 2014). The P600 has been suggested to play a role in indicating a need for reanalysis following a language/ lexical violation in a sentence (e.g., Kuperberg, Sitnikova, Caplan, & Holcomb, 2003;Mü nte, Heinze, Matzke, Wieringa, & Johannes, 1998). Within this definition of the semantic P600, it is possible that the 'reversed N400' seen in later time windows for the true-belief context and all time windows for the false-belief context (i.e., more positive-going waveform for inconsistent vs consistent critical words) reflects greater reanalysis of the input following a belief-inconsistent word compared to a belief-inconsistent word (since listeners reintegrate information after detecting the violation), and thus reflects appropriate language processing and reinterpretation in the false-belief context (i.e., reanalysis to fit with the character's false belief). This interpretation fits with the explicit ratings of appropriateness, which showed that participants successfully tracked the character's true and false beliefs by the end of the story. Moreover, if the correlation effects in the later 'N400' time period were interpreted as a semantic P600, this would indicate that older adults showed a greater reanalysis for false-belief inconsistencies than younger participants, indicating a processing advantage for older adults who are more likely to re-analyse and 'repair' sentences in order to make sense of them. Distinguishing these two components using source analysis on the current data is beyond the scope of this paper, but future research may focus on examining the role of age in predicting N400 and semantic P600 components separately in language processing tasks. Our third question asked whether any age-related changes in ToM can be attributed to domain-general changes, or whether they are specific to social abilities. Correlation analysis highlighted a negative correlation between age and both inhibitory control and working memory, indicating that these abilities decline with age, as expected (Braver & West, 2008;De Luca et al., 2003;Salthouse, 2009;Schroeder & Salthouse, 2004). Both of these executive skills are argued to play an important role in ToM engagement e inhibitory control allowing suppression of one's own perspective in favor of another person's, and working memory allowing multiple mental states to be represented simultaneously. Importantly, the current results highlighted that, contrary to our third hypothesis and despite these age-related declines in executive functions, age effects seen in the N400 consistency effect for false-belief trials in the 400e600 msec time window were not reflective of a domain-general decline (due to the absence of correlations with either executive function ability), but rather a more domain-specific decline in implicit social-cognitive processing abilities. In fact, this lack of domain-general input from cognitive abilities concords with recent data from Martin et al. (2019) showing that an age-related decline in visual perspective-taking (specifically switching from the self to other perspective) is only partially mediated by declines in inhibitory control (and not at all by working memory). In our study, this is further supported by the correlation with age being restricted to false-belief conditions, rather than truebelief conditions, indicating that declines in abilities as a result of increasing age are specifically related to an individual's ability to infer and use ToM online. In true-belief conditions, there is no conflict between the participant's and story character's belief states, which are both aligned with reality, thus removing any potential interference effects between the two different perspectives. These data suggest that agerelated declines in ToM are specific to the social domain, although it remains to be ascertained to what extent the changes in ERP amplitudes established in the current study relate to explicit social cognition performance, and thus the extent to which ageing effects influence engagement in social cognition capacities in everyday life. Interestingly, the results from the current study were established despite the older adult group having, on average, an IQ over 10 points higher than other groups, reflecting the high-functioning cognition of the older participants in this sample. To further examine the role of age in predicting differences in the N400 effect, it may be of interest in further research to examine how individuals with higher versus lower IQ within one age category differentially perform on behavioural and neural measures of ToM, establishing whether IQ influences age-related changes in ToM performance. Additionally, the adolescent group in the current study showed considerable variation in performance, as highlighted in Fig. 3. The presence of this variation supports prior research that has suggested large individual differences in neurodevelopmental trajectories across individuals during adolescence (e.g., Dumontheil, 2016;Foulkes & Blakemore, 2018;Tamnes et al., 2018), and highlights the importance of including adolescents in research examining the development of ToM into adulthood, with significant changes present throughout adolescence. This critical period of social, psychological, and biological development (Foulkes & Blakemore, 2018) would be interesting to examine in more refined detail than permitted by the scope of the current paper, exploring how ToM processing may change across the adolescent period, and how individual differences may influence these changes. In sum, the results of the current study make novel contributions to the field by demonstrating that, although older adults were able to explicitly evaluate story events according to a character's (false) beliefs, they experience age-related changes at an implicit level. Older participants maintained an egocentric stance for longer than younger participants, who were able to switch to take the character's more limited perspective at a later stage of online comprehension. Prior research has suggested a differentiation between explicit and implicit ToM abilities, with older adults experiencing greater difficulties than younger adults when engaging in explicit ToM tasks (e.g., Bailey & Henry, 200 9;German & Hehman, 2006;Grainger et al., 2015;Maylor et al., 2002), but showing less age-related changes in tasks assessing implicit ToM, where older and younger adults perform comparably (e.g., Grainger et al., 2018). In the current study, the need to draw ToM inferences was not strictly implicit in the task (i.e., participants were asked to judge the appropriateness of the character's actions after some sentences, thus were cued to attend to the character's perspective), but our ERP data highlights agerelated changes in implicit measures, indicating a possible explanation for experiences of declining social skills at older ages. The results highlight that with increasing age, interference from the self perspective is more pervasive, and inhibition of this egocentric perspective takes longer to achieve. In turn, this could lead to repercussions during social interactions, with delayed processing of the perspectives of other people when engaging in social situations. Further research is needed to establish the impact that this elongated egocentric bias in online ToM may have on everyday social functioning. Conclusion In conclusion, the results of the current study provide novel evidence that people experience an egocentric bias in implicit ToM processing that persists throughout the lifespan. When listening to stories in which a character is described as having a true-belief about an objects location, a standard N400 effect is elicited when the character acts in an inconsistent manner (vs consistent). In contrast, in false-belief conditions an enhanced N400 effect is seen when a character acts in a beliefconsistent manner (vs belief-inconsistent), indicating that the stories have been processed from the participant's own perspective, failing to spontaneously consider the perspective of the character. ERP results showed that in an early time window (200e400 msec post stimulus onset) there was no significant effect of age, suggesting that all individuals demonstrated this egocentric effect to a comparable degree. In contrast, a significant effect of age emerged in the later time window (400e600 msec), suggesting that increasing age led to a longer time taken to disengage from the self-perspective to consider the character's perspective. In the final 600e1000 msec time window, this age modulation was no longer present. Importantly, these age-related changes were not as a result of domain-general cognitive changes, with no significant relationship to executive function abilities, suggesting a more domain-specific social-cognitive decline in online ToM. This study demonstrates the importance of examining implicit ToM processes in healthy aging populations e including both adolescence, which is a period of rapid developmental change, and older aged adults e allowing further understanding of how ToM functions and develops across the lifespan to be established. Open practices The study in this article earned Open Materials and Open Data badges for transparent practices. Materials and data for the study are available at https://osf.io/pw7h6/.	2020-02-04T14:25:12.854Z	2020-02-04T00:00:00.000	{ "year": 2020, "sha1": "eaa89ed8feb994b40daa05d58482e52e3d582d5d", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1016/j.cortex.2020.01.016", "oa_status": "HYBRID", "pdf_src": "Elsevier", "pdf_hash": "eaa89ed8feb994b40daa05d58482e52e3d582d5d", "s2fieldsofstudy": [ "Psychology" ], "extfieldsofstudy": [ "Psychology", "Medicine" ] }
123778074	pes2o/s2orc	v3-fos-license	The possible conection between q-defomed hamornic oscillator formation and anharmonicity In our previous article, the connection between q-deformed harmonic oscillation and Morse-like asymmetric potential is investigated. In present work, a possibility of the connection between q-deformed harmonic oscillator and anharmonic symmetric potential is in detail considered. For simplicity, we take the inverse square cosine-hyperbolic form of potential, i.e Pöschl-Teller potential. The relation between the deformation parameter q and the set of parameters of anharmonic symmetric potential was found. The correspondence of two types of connections between q-deformed harmonic oscillator with asymmetric and symmetric potentials are discussed. Introduction In the last twenty years, quantum group and deformed Heisenberg algebras with q-deformed harmonic oscillator have been a subject of intensive investigation. This approach has found some useful applications in various branches of physics and chemistry [1,2,3,4,5,6,7]. The method of q-deformed quantum mechanics was developed on the base of Heisenberg commutation relation (the Heisenberg algebra) within some typical generalizations. The main parameter of this method is the deformation parameter q which is usually considered to variety in the range 0 < q < 1, and the models have been constructed that the behaviors of studying objects reduce to theirs conventional counterparts as q → 1. The Morse potential finds an important role in describing the interaction among atoms in diatomic and even in polyatomic molecules [8,9,10,11,12,14] of atomic and molecular physics. Despite its quite simple form, the Morse potential describes very well the vibrations of diatomic molecules. This is because that four-particle complex system (two heavy atomic nuclei with positive charge and two light electrons with negative charge) can be reduced to relative motion of two atomic nuclei in an effective potential which is average Coulomb interaction of nuclei and electron clouds. The Morse-like potential models just work with a simple one-dimensional threeparameter effective potential, and find many applications in condensed matter, bio-physics, nano-science and quantum optics. The Morse potential in algebraic approach can be written in terms of the generators of SU (2). The quantum relation between q-deformed harmonic oscillator and the Morse potential was considered in [10], where then the anharmonic vibrations in the Morse potential have been described as the levels of q-deformed harmonic oscillator. The extended SU (2) model (q-Morse potential) has been also developed to compare with phenomenological Dunham expansion and experimental data for numbers of diatomic molecules [10]. In this work, by considering deformed algebra as mathematical object and atomic effective potential as physical model, we use this relation in inverse way to investigate properties of q-deformed harmonic oscillator on the base of the Morse potential. In one hand, the potential of harmonic oscillation is parabolic with infinity equal-step levels. In other hand, we show that the potential of q-deformed harmonic oscillator can be described as Morse-like anharmonic potential with finite unequal-step levels. The relation between the deformation parameter q and the set of parameters of Morse-like anharmonic potential was found. We have also investigated the partition function and some thermodynamic properties of q-deformed harmonic oscillator. In the our previous work [14] we have shown that mathematical deformation properties can be represented and understood in the language of physical object, which can be described by an anharmonic potential. The asymmetric representation of deformed harmonic oscillators was investigated in [14] with the Morse potential. And as a further step, in this work we study the symmetric representation of deformed harmonic oscillators with corresponding potential. Harmonic oscillator and q-deformed harmonic oscillator In q-deformed harmonic oscillator, creation a † and annihilation a operators of q-deformed harmonic oscillator satisfy the commutation relation where q is deformation parameter taking values in [0, 1]. In the second quantization representation, the Hamiltonian operator of q-deformed harmonic oscillator is written as As the results of simple algebraic manipulations, energy spectrum of q-deformed harmonic oscillator is obtained as follows where [n] q = 1−q n 1−q is the q-integer which differs from natural numbers. For very small derivation from unity ε = 1 − q, the energy spectrum becomes quadratic if the higher order contribution In result (4), the energy levels are represented by a system of parallel lines are not equidistant. The extent depends on the deformation parameter q. Physical model for q-deformed harmonic oscillator In the case of q = 1(ε = 0), the energy levels return to non-deformed expression, i.e. we obtained the energy levels of the harmonic oscillator, the gaps between the energy levels are constant ( Figure 1). Figure 1. Energy spectrum has infinity equal-step levels form. In the range of parameter values 0 < q < 1, the energy levels are represented by a system of parallel lines are not equidistant. In all previous work, we have studied the deformation parameter q through Morse potential, which is the asymmetric anharmonic potential. In this ongoing one, the deformation parameter q with symmetric anharmonic potential form ( Figure 2) will be under consideration. To easily compare potential energy shapes, we observe their appearance on the same coordinate system (Figure 3). The symmetric potential is anharmonic and energy spectrum has finite unequal-step levels, for comparison also plotted harmonic spectrum of corresponding parabolic potential. Symmetric representations The symmetric potential has an important role in describing the interaction of atoms in diatomic, and even polyatomic molecules. In the case, the symmetric potential is anharmonic, energy spectrum has finite unequal-step levels as presented in the Figure 4, where energy spectrum of symmetric anharmonic and corresponding parabolic potentials are compared. The symmetric anharmonic potential chosen in study is Pöschl -Teller potential Figure 3. The harmonic potential, anharmonic potential and symmetric potential. which is an effective potential with a set of two parameters U 0 , α taken from experimental data. The Schrödinger equation of symmetric anharmonic potential (6) has been exactly and analytically solved [13]. The eigenvalues of this Hamiltonian read By introducing new quantity the energy level of system under consideration can be expressed as where ω s is the frequency of symmetric oscillator. Following the tasks presented in our last work [14], the deformation parameter q s of the given potential is found as Figure 5. q-deformed symmetric representation depending parameters U 0 , α. It is obvious that this system has a finite number of energy levels with maximum number n S.max n S.max = where the notation [f ] is the integer part of the number f . Using the above relationship, we propose the deformation model (SPD model), based on the symmetric potential, for investigating the properties of q-deformed harmonic oscillator. The main idea is that the role of parabolic potential for harmonic oscillator would be replaced by the symmetric representation potential for q-deformed harmonic oscillation. In above proposed model, every given value of deformation parameter q S in the interval from zero to unity, q S ∈ [0, 1], can be described by a symmetric potential with the largest number n S.max determined by the expression (11). Here, we note the well-defined one-to-one correspondence between q S and n S.max . The values of largest number depending on deformation parameter q S and on parameters of material system are plotted in figure 6 and in figure 7. It is shown that the total number of energy levels rapidly grows up when q tends to unity. In the limit of weak deformation q → 1, n max → ∞ . In contrast, when q tends to 0.5 only one level can be found, i.e. in the strong deformation limit q → 1/2, n max → 1. The actual working range of deformation parameter q is not to be in q ∈ [0, 1], but in more narrow half range 0.5 < q < 1, where the physical system has more than one energy level. The last but not least, the main parameters of harmonic, q-deformed harmonic, symmetric and asymmetric potentials are compared in Table 1. Table 1. Comparison of asymmetric and symmetric representations. Discussion Comparing the energy spectrum of q-deformed harmonic oscillator (4) and energy spectrum of symmetric anharmonic potential (7) and neglecting the higher order contribution O ε 2 = O (1 − q) 2 , we realize the relations This is the relation between q-deformed harmonic oscillator and symmetric anharmonic vibrations, providing the analogy of q-deformation and symmetric anharmonicity. The change of energy spectrum from linear to quadratic form is reasoned by deformations of commutation relations in the mathematical algebraic approach, while in case of the symmetric potential by appearance of periodic distribution of atoms or molecules. The main results of this work are the consequences of the expression of energy levels (7), on which we proposed a new symmetric potential deformation SPD model for investigating properties of q-deformed harmonic oscillator via the symmetric potential (6). Instead of parabolic potential for harmonic oscillation, the symmetric anharmonic potential is used to study the q-deformed harmonic oscillation, and as it is expected, the mathematical deformation properties now would be described and understood in the language of physical anharmonic behaviors. In our proposed SPD model, q-deformed harmonic oscillator can be described by the corresponding symmetric Pöschl-Teller potential. With every given value of deformation parameter in the interval q S ∈ [0, 1], we would determine the largest number n S.max and the deformed energy spectrum E q (n) of q-deformed harmonic oscillation. In one hand, the asymmetric representation of deformed harmonic oscillators by using a symmetric Morse potential [14] can be applied to describe interaction between two atoms molecules, where the space among them is clearly non-symmetric. In other hand, the possible applications of symmetric representation for deformed harmonic oscillators are atomic chain, symmetric atoms, where the space is translationally symmetric. This models could be expanded and generalized to investigate many other physical problems such as composite bosons with deformation which is the topic of our ongoing research.	2019-04-21T13:12:52.348Z	2016-06-01T00:00:00.000	{ "year": 2016, "sha1": "dee43a00b540f1856b21eed25417fae8d566b036", "oa_license": null, "oa_url": "https://iopscience.iop.org/article/10.1088/1742-6596/726/1/012018/pdf", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "7fe1bdbe683d18761636c5036e4c3edae1166816", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Mathematics", "Physics" ] }
85457422	pes2o/s2orc	v3-fos-license	Quasiparticle Interference and Symmetry of Superconducting Order Parameter in Strongly Electron-Doped Iron-based Superconductors Motivated by recent experimental reports of significant spin-orbit coupling (SOC) and a sign-changing order-parameter in the Li$_{1-x}$Fe$_x$(OHFe)$_{1-y}$Zn$_y$Se superconductor with only electron pockets present, we study the possible Cooper-pairing symmetries and their quasiparticle interference (QPI) signatures. We find that each of the resulting states - $s$-wave, $d$-wave and helical $p$-wave - can have a fully gapped density of states (DOS) consistent with angle-resolved photoemission spectroscopy (ARPES) experiments and, due to spin-orbit coupling, are a mixture of spin singlet and triplet components leading to intra- and inter-band features in the QPI signal. Analyzing predicted QPI patterns we find that only the spin-triplet dominated even parity $A_{1g}$ (s-wave) and $B_{2g}$ (d-wave) pairing states are consistent with the experimental data. Additionally, we show that these states can indeed be realized in a microscopic model with atomic-like interactions and study their possible signatures in spin-resolved STM experiments. I. INTRODUCTION In iron-based superconductors, it has been widely believed that superconductivity is driven by repulsive interactions, enhanced by the presence of the spin fluctuations associated with the parent antiferromagnetic state. In this scenario, these fluctuations drive a sign reversal (s ± state) 1-5 between order parameters (OP) on the electron and hole Fermi surface pockets at the M -and Γ-point, respectively. The discovery of superconductivity in intercalated or monolayer FeSe at a critical temperature of the order above 40K revived interest in Fe-based superconductivity, but raised further questions on the origin of superconductivity in these compounds [6][7][8][9][10] , because, unlike bulk FeSe, ARPES experiments show that many of these FeSe-derived systems appear to be missing the hole pockets at the Γ-point required in the the conventional scenario. Initial model calculations based on the multiorbital spin-fluctuation framework for systems manifesting only electron pockets at the M -point predicted d-wave symmetry state in this case 11,12 , driven by the spin fluctuations connecting the electron pockets that remain when the hole pockets are removed. In the proper 2-Fe unit cell, such a state must have gap nodes on the Fermi surface 13 . This is because the electron pockets located near (π, 0) or (0, π) points of the Brillouin Zone (BZ) in the 1-Fe unit cell fold onto (π, π) point of the folded BZ as the crystallographic symmetry lowers due to the Se positions. This may lead to hybridization between the electron pockets 14 , which then forces the d x 2 −y 2 -state to acquire gap nodes, although in principle the nodal area may be very small, proportional to the hybridization ("quasinodes"). On the other hand, ARPES experiments in most of the electron-intercalated materials indicated a nodeless superconducting (sc) state 15,16 . Several proposals for the gap structure have been put forward, including a conventional s ++ -wave scenario based on the electron-phonon interaction and orbital fluctuations 17 , as well as the "bonding-antibonding" scenario 13,18 in which the order parameter on the inner electron pocket (mostly d xz /d yz character) has one sign, and on the outer electron pocket (mostly d xy character) the other 14 . Furthermore, it has been argued that the hybridization of the electron pockets is mainly due to SOC [19][20][21] , which within a 3D spin fluctuation framework may stabilize the bonding-antibonding s ± state against d-wave 19 and mixes a spin-triplet component into the even parity s +− -wave state 20,21 . Overall, one can see that the sign structure of the superconducting order parameter is intimately related to the pairing mechanism. Therefore, experiments allowing to determine it could be of great potential importance. One rapidly developing technique to determine the phase structure of the order parameter makes use of QPI as measured by Fourier transform scanning tunneling microscopy (FT-STM). This probe measures the wavelengths of Friedel oscillations caused by impurities present in a metallic or superconducting system, which in turn contains information on the electronic structure of the pure system. A subset of scattering wave vectors q corresponding to peaks in the FT-STM can be enhanced or not according to the type of disorder and the phase structure of the superconducting gap 22,23 . Recently it was proposed by Hirschfeld, Altenfeld, Eremin and Mazin (HAEM) 24 that the sign structure of the order parameter in a multiband system can be extracted from the Fourier transform of the local density of states QPI pattern near an impurity in the superconducting state. The antisymmetrized QPI intensity integrated over the wavevectors corresponding to scattering between two bands was shown to have a dependence on frequency very different for sign-changing and sign preserving scenarios leading to a strong, single-sign enhancement of the integrated response in the former case. This qualitative result was also confirmed by extensive numerical simulations with finite disorder 25 . Recently, a complementary phase sensitive technique to detect sign-changing gaps in the presence of strong impurity bound states was proposed 26 . Using QPI analysis, the authors of Ref. 27 were able to identify a sign changing order parameter in FeSe. Most importantly for our purposes, similar conclusions were recently drawn for the strongly electron doped iron-based superconductor lithium hydroxide intercalated FeSe 28 . In other words, the order parameter in Li 1−x Fe x (OHFe) 1−y Zn y Se, alternates sign, either between the Fermi surface sheets, or within individual sheets. However, distinguishing between these alternatives was beyond the resolution of the experiment. In any case, the situation is somewhat more complicated than anticipated in Ref. 28, since the effect of spin-orbit interaction on pairing needs to be taken into account as well. Moreover, recent observation of Majorana zero modes in (Li 0.84 Fe 0.16 )OHFeSe 29 suggests possible broader implications of the spin-orbit coupling for the Cooper-pairing in electron doped intercalated iron-based superconductors. Note that in contrast to Ref. 28, no Zn substitution was used in Ref. 29. The amount of Zn, however, is relatively small (less than 2 percent). This amount does not affect the superconducting transition temperature or the electronic structure in a significant way and is done only for the purpose of enhancing the QPI signal in the scanning tunneling microscopy. In this manuscript we study the possible Cooper-pairing symmetries and their QPI signatures for strongly electrondoped Fe-based superconductors using the effective three-orbital model of Refs. 20 and 21 with spin-orbit coupling and proper consideration of all lattice symmetries of the FeSe space group. We find that each of the resulting states -A 1g -wave, B 2g -wave and helical E u -wave -can have a fully gapped DOS consistent with ARPES experiments and, due to spin-orbit coupling, are a mixture of spin singlet and triplet components leading to intra-and inter-band features in the QPI signal. Analyzing predicted QPI patterns we find that A 1g -wave pairing state, with the two dominant peaks in the DOS roughly corresponding to the gap energies on each pocket, and B 2g -wave pairing state both with a significant even parity spin triplet component are consistent with the experimental data. Moreover, we show that pairing states with dominant spin triplet component can be identified using spin-resolved STM. 1. (a) Single layer of the iron based superconductors lattice structure. Red and green dots are iron and pnictogen or Se atoms, respectively. One pnictogen sublattice is puckered above the iron layer (filled green dots) one is puckered bellow (empty green dots) which divides the iron atoms into sublattices A and B. One-iron unit cells for sublattices A and B are denoted by dashed squares. The two-iron unit cell, taking the puckering into account, is shown by the solid square. The vector τ0 connects sublattices A and B. We wish to describe the low energy states near the M,-points of the Brillouin zone using the orbitally projected band model of Ref. 20 and 21 for the two-iron unit cell. Near the Fermi level, only the xz, yz and xy orbitals contribute significantly; hence, the full 10-orbital tight binding model is projected onto the subspace of these three orbitals. The effective low energy Hamiltonian near the M-point that takes into account all the lattice symmetries of the FeSe space group as well as time reversal symmetry is defined as II. MODEL where the four component spinor Ψ † M,σ (k) = Ψ † X,σ (k), Ψ † Y,σ (k) describes the states at the M-point for each spin projection σ. The doublets Ψ X,σ (k) and Ψ Y,σ (k) are defined as Moreover, we have and where the Pauli matrices {σ x , σ y , σ z } and {τ 1 , τ 2 , τ 3 } act on spin and orbital space, respectively. The λ z , p z1 and p z2 terms in eq.(3) describe the k-dependent intra-band SOC which does not couple the two Fermi pockets but lifts the out-of plane spin degeneracy. The inter-band SOC term which hybridizes X-and Y-pocket is given by In order to describe intercalated FeSe we use the Luttinger invariants, Tab.(I), which were evaluated in Ref. 21 based on the available ARPES data. The value of the intraband spin-orbit coupling, λ z is in agreement with those found in ab-initio calculations and ARPES experiments 30,31 . Furthermore, the value of the interband spin-orbit coupling, λ SOC between the electron pockets separated by the large momentum yielding their hybridization and splitting on the Fermi surface is found to be smaller and is taken to be ∼ 5 meV 31 . Diagonalizing Eq.(1) yields four bands: two regular ones that form the inner and outer electron pocket, see Fig.1(b), and two incipient bands that do not cross the Fermi level. The effect of inter-band SOC on the band structure is visualized in Fig.1(c) and Fig.1(d). III. MEAN FIELD PHASE DIAGRAM Although phenomenologically the classification of superconducting orders for two electron pockets was considered previously 20,21 we analyze here its microscopic formulation via mean-field treatment of the atomic on-site interactions given by the Hubbard and Hund's couplings U , U ′ , J and J ′ which enter the Hubbard-Kanamori Hamiltonian as Here α ∈ {A, B}, {σ, σ ′ } ∈ {↑, ↓} and {µ, ν} ∈ {yz, xz, xy} label lattice sites, spins and orbitals, respectively. d α † µ,σ (r) and d α µ,σ (r) are the second quantized operators creating and annihilating particles on sub-lattice A and B, see Fig.1(a). Using the results presented in Ref. 20 and assuming sharply localized Wannier functions of the xz, yz and xy orbitals, we can relate, up to a constant, the d α µ,σ (k) operators acting in the one iron unit cell to the components of the doublets Ψ X,σ (k) and Ψ Y,σ (k) in the two iron unit cell via where we absorb the constant prefactors into the Hubbard and Hund terms. As the xy orbital contributes to both Xand Y-point eq.(10) leads to "Umklapp" terms at the M-point. We assume that even parity solutions are still the leading pairing instabilities (for odd parity solutions see Appendix B) and use Eqs. (8)(9)(10) to project eq.(7) onto the low energy model decoupled into the spin singlet A 1g s-wave and B 2g d-wave symmetry states which at the M-point in presence of inter-band SOC couple to the E g even parity spin triplet state. When defining the two doublets Ψ T 1σ (k) = (c yzσ (k), c xzσ (k)) and Ψ T 3σ (k) = (c xy X σ (k), c xy Y σ (k)) the pairing terms read The 2x2 block in eq.(11) corresponds to A 1g s-wave pairing. We write J ′ 13 = αJ ′ 11 and find two eigenvalues E A1g = 1 4 J ′ 11 2 + 3U ± (J ′ 11 ) 2 + α16J ′ 11 − 2J ′ 11 U + U 2 which correspond to ordinary "plus-plus" (s ++ ) s-wave and sign-changing "plus-minus" (s ± ) s-wave pairing, respectively. While the former channel is purely repulsive without spin-orbit coupling the latter becomes attractive once J ′ 11 > (U + U √ 1 + 8α)/4α. The paring term that leads to B 2g d-wave is E B2g = 1 2 (U − J ′ 11 ) and can be directly read off. It is attractive once J ′ 11 > U and competes with sign-changing s-wave. Since we assume sharply located Wannier functions which yields eqs.(8-10) and on-site interactions only we find that within our simple mean field approximation the xy orbitals do not contribute to d-wave pairing as no "pair-hopping" term J ′ 33 mediates between xy X and xy Y . This changes once the higher-order diagrams (spin fluctuations) are taken into account. The E g even parity spin triplet corresponds to pairing between the first (second) component of Ψ Xσ (k) (Ψ Y σ (k)) and the second (first) component of Ψ Y σ (k) (Ψ Xσ (k)) and is thus inter-band and attractive once E Eg = U ′ 13 −J 13 < 0. We perform a mean-field decoupling of Eq.(11) into A 1g and B 2g spin singlet channels with the pairing terms given by In terms of Ψ M,σ (k) a triplet term can be written as Ψ T M,σ (−k)Miσ y σΨ M,σ (k), whereM and iσ y σ represent orbital and spin part, respectively. Since iσ y σ is symmetric an even (odd) parity triplet requiresM to be antisymmetric (symmetric). iσ y σ can be divided into an in plane iσ y (σ x , σ y ) and out of plane iσ y σ z component which transform as as the two dimensional E g and one dimensional A 2g irreducible representation, respectively. In presence of SOC, orbital and spin degrees of freedom transform together under operations of the space group. We focus on the E g even parity spin triplet that together with the E g in-plane spin component decomposes into a direct sum of one dimensional representations as Using the two anti symmetric components of E g and τ ± = (τ 1 ± iτ 2 )/2 one finds two even parity spin triplets that transform according to A 1g and B 2g and hence, couple to the singlet channel. We refer the reader to Refs. 21,32 for further details. In terms of the spinor Ψ T with and the pairing terms∆ where the gaps in orbital space are given by the equations (A1) and (A2). We self-consistently compute∆ A1g and ∆ B2g as a function of temperature and inter-band SOC for the two cases E A1g < E B2g and E A1g > E B2g and present In the case when solely E A1g (E B2g ) is attractive, ∆ A t (∆ B t ) is induced by SOC and scales with λ SOC . In this case \|∆ \| and we call the state a spin singlet dominated A 1g (B 2g ). If E Eg is attractive, however, ∆ t can develop independently of ∆ 1,3 and allows for states where \|∆ . Equally whether the system is in the singlet or triplet dominated regime, the form of Eqs. (21) and (22) stays the same. Depending on the ratio α = J ′ 11 /J ′ 13 the leading pairing symmetry for low values of λ SOC is either B 2g d-wave or A 1g s-wave with accidental nodes and a sign change between inner and outer Fermi pocket. In Fig.2(a) we show a SOC mediated transition from B 2g d-wave to A 1g s-wave. For small values of λ SOC we find a nodeless d-wave state where the nodes are lifted and the gap is opened by a combination of spin singlet and triplet components with the singlet gap being dominant \|∆ B 1 \| > \|∆ B t \|. With increasing λ SOC the paring symmetry changes from nodeless to nodal d-wave and finally to nodeless s-wave with a dominant inter-band triplet component Fig.2(b) we choose parameters such that the initial state at small λ SOC is A 1g s-wave and has a dominant singlet gap Increasing λ SOC lifts the nodes at the Fermi level and drives the system in an A 1g s-wave symmetry with a dominant triplet gap \|∆ A t \| > \|∆ A 1 \|, \|∆ A 3 \|. We find that for large inter-band SOC the triplet dominated s-wave state wins over triplet dominated d-wave. The presence of an intra-band SOC term λ z does not change the phase diagram qualitatively. However, in the region where λ SOC is strong superconductivity in the d-wave channel is suppressed if λ z > 0 which further stabilizes the A 1g -wave solution. Moreover, a large triplet gap, in addition to SOC, lifts accidental nodes leading to a nodeless s ± pairing symmetry. Figure2(a) is consistent with the phase diagram of phenomenological model in Ref. 14 with several important differences. In particular, in our model the singlet dominated d-wave state competes with singlet dominated bonding anti-bonding s ± which arises from pair hopping between xz(yz) and xy orbitals. Furthermore, with increasing spinorbit coupling strength the A 1g state contains dominant interband spin triplet component which was absent in the simplified analysis of Ref. 14 as its pairing channel turned out to be strongly repulsive in the two band model. Since SOC couples even parity inter-band triplet to intra-band even parity spin singlet pairing, an attraction in the former induces a gap in the latter channel and vice versa. Consequently a gap at the FS opens even though the E g triplet is inter-band. A similar paring state with attraction in the triplet channel was recently proposed for highly doped systems with only hole pockets 33 . A. Local density of states in presence of impurities In order to investigate the QPI signatures of the possible pairing states we need to calculate the local density of states (LDOS) for a multi-orbital system in the superconducting state. We further investigate corrections to the LDOS which arise from scattering at single charge impurities to make a statement whether there is a sign-change between inner and outer electron pocket using HAEM's method. Moreover, we consider scattering from a single magnetic impurity to reveal information about the spin structure in the system. In order to describe superconductivity and spin-resolved STM we introduce the 16 component Balian-Werthammer spinor Within this basis the Hamiltonian in the superconducting state is given by where the additional factor of 1 2 accounts for double counting and From Eq.(24) we find the superconducting Green's functionĜ 0 ] −1 and the local density of states (LDOS) given by where we assumed sharply localized Wannier functions of the xz, yz and xy orbitals. We now introduce a single on-site non-magnetic potential scatterer to the system which is located on either sublattice A or B. In the one-iron unit cell it can be described as We assume that the major contribution to scattering is of intra-orbital nature and project eq.(27) onto the states in the two iron unit cell. Due to the fact that d A/B xy in the projected model is a linear combination of c xy X and c xy Y the intra-orbital term induces scattering between xy X and xy Y thus also contributes to inter-band scattering. In the two iron unit cell the impurity potential is given by Following the HAEM 24 approach, we compute the antisymmetrized correction to the LDOS due to impurity scattering with δĜ q (ω) = kĜ 0 k (ω)ÛĜ 0 k+q (ω) being the convolution of the bare Green's functions dressed by a Nambu scattering matrixÛ = τ 3 ⊗ σ 0 ⊗Ṽ in Born approximation, i.e. with V α,β ≪ E F , where E F is the Fermi energy. The HAEM method states that the momentum integrated and antisymmetrized LDOS, δρ − (ω), qualitatively differs in case of a sign-changing OP from that of a sign-preserving one yielding a strong enhancement of the integrated response for the sign-changing but not the sign-preserving one. This method has been recently successfully applied to confirm the sign-changing nature of the order parameter in FeSe 27 , where the superconducting gaps are also extremely anisotropic. More recently, the method has also been applied to (Li 1−x Fe x )OHFe 1−y Zn y Se 28 where δρ − (ω) shows a strong signal and no sign-change between 8 meV and 14 meV suggesting a sign-changing s ± pairing symmetry. Since our previous analysis shows that most of the conclusions regarding the phase structure of the superconducting gap obtained within Born limit are robust and remain valid also well beyond this limit, we restrict our analysis to weak potential scatterers. B. Local density of states and phase sensitive correction to QPI We would like to discuss singlet and triplet dominated A 1g -and B 2g -wave with respect to their consistency with QPI experiments in (Li 1−x Fe x )OHFe 1−y Zn y Se. Recently in Ref. (21) possible pairing states for intercalated FeSe have been discussed. Following Ref. [21] the pairing has to obey three criteria in order to be consistent with experiments: i) fully gaped LDOS, ii) the quasiparticle energy extrema are at or close to k F of the normal state ("back bending") and iii) two peak features in the DOS with a peak at about 8 meV and 14 meV, respectively. In the context of monolayer and intercalated FeSe it has been suggested that the second peak in the DOS arises due to pure interband gap 21 as in the former the gap size seen by ARPES 16 (13.7 meV) significantly deviates from the peak energy (20.1 meV) seen by STM 8 . This, however, does not have to be the case with the electron-intercalated materials where a gap of 13 ± 2 meV around the Fermi energy was reported 10 . Three pairing states were proposed to be consistent with experimental data. These are A 1g s-wave and B 2g d-wave and E u ⊗ U (1) helical p-wave. The A 1g and B 2g state were assumed to have a dominant intra-band singlet gap in order to ensure a full gap and back bending. An inter-band SOC then mixes in inter-band triplet E g pairing. For the odd parity intra-band triplet E u ⊗ U (1) p-wave state its the odd parity inter-band singlet A 2u that is coupled via SOC. In the following we investigate spin singlet and spin triplet dominated A 1g -and B 2g -wave for their consistency with the citeria i)-iii) and the QPI data. For that we use Eq.(24) and calculate the LDOS (ρ(ω)) and the superconducting band dispersion to ensure i)-iii) are fulfilled. In addition, for each pairing state, we calculate the antisymmetrized correction to the LDOS (δρ − (ω)) to check whether the order parameter changes sign between electron pockets. Since the gaps in the intercalated FeSe are found to be isotropic 10 we further present the angular dependence of the gap projected on the inner and outer electron pocket, respectively. In Appendix (B) we also briefly discuss the QPI data for the odd parity E u -wave state and show that at least within Born scattering the results are not compatible with experiment. We find that A 1g and B 2g states with a dominant spin triplet gap can show back bending and, in contrast to their singlet dominated states and the odd parity E u -state, give an appropriate description of the QPI data found in experiment. One should also further note that for the HAEM method the crucial role is played by the gaps present on the Fermi surface pockets. In the present case there is also additional interband gap. We show in the Appendix C that its phase structure with respect to the gaps present on the Fermi level cannot be elucidated within phase-sensitive QPI analysis. Furthermore, we show that the sign-changing and sign-preserving gaps on the Fermi surface still determine the characteristic features of δρ − (ω). C. A1g-symmetry state We start by examining spin singlet and triplet dominated s-wave pairing state. We use the singlet gaps ∆ 1 and ∆ 3 and the triplet gap ∆ t to fit band-structure and LDOS. To start with singlet dominated s-wave we utilize the fitting parameters presented in Ref. 21. As shown in Fig.3(a), ρ(ω) is indeed fully gaped and has a peak at 8 meV and 14 meV. The first peak between 6 − 10 meV mostly comes from the intra-band gaps at the electron pockets as can be seen by comparison with the gap projections on the inner, ∆ in and outer, ∆ out , electron pocket, Fig.3(c). At θ F = π/4 both gaps are equal in magnitude leading to a peak at ω ∼ 8 meV in ρ(ω). The width of the peak is limited by 6 meV and 10 meV which correspond to the minimum of ∆ in and the maximum of ∆ out , respectively, so that the occupied states below and above 8 meV are due to anisotropy of the gaps. The second peak at 14 meV is indeed due to inter-band pairing. It vanishes for λ SOC → 0 and ∆ t → 0. We find that both gaps are highly anisotropic and most important positive on both Fermi surface pockets, yielding an s ++ -pairing symmetry. In Fig.3(b) δρ − (ω) exhibits a sign change in the region 6 meV < ω < 10 meV which can be interpreted as a sign preserving order parameter and thus is not compatible with the QPI data 28 . Note that this state does not appear in our phase diagram, shown in Fig.2, where the A 1g state is sign-changing. The feature at ω ∼ 14 meV in δρ − (ω) for this state appears due to the small inter-band contribution and does not carry phase information as we explain for a simple model in Appendix C. In Fig.3(d) the superconducting band dispersion in ΓM -direction (θ F = π/4) is shown where the minima of the lower superconducting band are located above the former Fermi level (back-bending). It turns out that the fitting parameters for spin singlet dominated s-wave, proposed in Ref. 21 would give rise to s ++ superconductivity. A possible s ± pairing symmetry with a dominant spin singlet gap, on the contrary, would require sign∆ A 1 = sign∆ A 3 . This state in our anaylsis exhibits highly anisotropic gaps and possible accidental nodes (see for example Fig.2c(1)). This makes it difficult to fit the U-shaped two peaked LDOS in the LiOH-intercalated FeSe data without invoking a spin triplet component, induced by spin-orbit coupling that lifts the nodes. Spin-triplet dominated A1g-wave state In Fig.4(a) and Fig.4(d) the LDOS and the superconducting band dispersion are shown. The LDOS shows a full gap and a peak at 8 meV and 13.7 meV, respectively, while the lower superconducting band shows back bending. The band projected gaps ∆ in and ∆ out we plot in Fig.4(c). Possible accidental nodes are lifted by a large ∆ t leading to a nodeless s ± pairing symmetry with two almost isotropic gaps. In Fig.4(b) we present δρ − (ω) which in contrast to the singlet dominated case does not change sign between 8 meV and 14 meV. If we compare Fig.4(a) and Fig.4(c) we find that the peak positions of the two peaks roughly agree with the the magnitudes of ∆ in and ∆ out and thus δρ − (ω) and the behavior of δρ − (ω) agrees with the experimental one. The first peak is sharper than the second since ∆ in is very isotropic. The second peak is broader as a consequence of an anisotropic ∆ out and the contribution from inter-band pairing. In contrast to the singlet dominated case δρ − (ω) exhibits well pronounced negative peaks at ω 1 ≈ ∆ out and ω 2 ≈ ∆ in without a sign change between them indicating an OP that changes sign between the electron pockets consistent with the experimental QPI data and ARPES data 10,28 . Also note that for the spin triplet dominated A 1g state the direct gap feature caused by ∆ out is very close to the interband feature in the DOS which agrees with the findings in Ref.10 and 28. The lower orange band has its minima centered above the former Fermi-level "back-bending". D. B2g-symmetry state We now discuss the singlet and triplet dominated B 2g d-wave pairing state. We use the singlet gaps ∆ 1 and ∆ 3 and the triplet gap ∆ t to fit band structure and LDOS. Note that even though our simple mean field approach leads to ∆ 3 = 0 for the d-wave case, as it involves decoupling of the repulsive intraorbital Hubbard interaction, a nonzero ∆ 3 is necessary to achieve qualitative agreement when fitting the experimental data. In theoretical calculations, a nonzero ∆ 3 appears due to inclusion of the spin fluctuation diagrams in the Cooper-pairing channel yielding momentum dependent interaction, see Ref. 11. Spin-singlet dominated B2g-state For the spin singlet dominated B 2g -wave state appropriate fitting parameters were found in Ref. (21). As shown in Fig.5(a) and Fig.5(d) the LDOS is fully gaped and has a peak at 8 meV and 14 meV, moreover, the lower superconducting band shows back bending. ∆ in and ∆ out are plotted in Fig.5(c). Both gaps are almost equal in magnitude and nodeless due to smallness of hybridization between the electron pockets, i.e \|∆ in/out \| > λ SOC . The first peak in the LDOS is due to the intra-band gaps ∆ in and ∆ out . We calculate δρ − (ω) and present the results in Fig.5(b). δρ − (ω) has positive peaks at ∼ 7 mev and ∼ 9 meV with no sign change between them indicating the sign change between ∆ in and ∆ out . Yet the third peak at ∼ 14 meV comes from inter-band pairing and causes two sign changes between second and third peak which is at odds with the experimental data 28 . Spin-triplet dominated B2g-state Even though a triplet dominated d-wave scenario is not realized in the phase diagram we quickly discuss this case with respect to criteria i)-iii) and QPI data. As shown in Fig.6(a) and Fig.6(d) the LDOS is fully gaped and has a peak at 8 meV and 14 meV, moreover, the lower superconducting band shows back bending. ∆ in and ∆ out are plotted in Fig.6(c). Both gaps are equal in magnitude and nodeless due to smallness of hybridization between the electron pockets. In contrast to the singlet dominated case for the triplet dominated case we use \|∆ B t \| > \|∆ B 1/3 \| and sign(∆ B t ) = sign(∆ B 1 ) = sign(∆ B 2 ). In Fig.6(b) we plot δρ − (ω). Two small peaks between 7 meV and 8 meV reflect ∆ in and ∆ out contributions to δρ − (ω). The peak at 15 meV is due to the large inter-band gap. Overall, the behavior of δρ − (ω) is nearly consistent with experimental data as δρ − (ω) does not change sign in the experimentally relevant energy range. However, one has to bear in mind that in contrast to the triplet driven A 1g case the phase structure of the gaps on the Fermi surface affect the behavior of δρ − (ω) only near the lower peak, and the structure near the second (interband) peak is determined by the interband gap and does not necessarily bear information about the signs of the order parameters (see Appendix C.2). As we clearly see, to obtain the agreement with QPI experiments 28 the spin-triplet interband component from E g state is necessarily required in both symmetry states and triplet dominated A 1g state appears most likely one. This poses an important question how to detect it. To stay within QPI we propose to employ spin-resolved STM as an additional tool to further specify the underlying pairing state and also to distinguish between B 2g and A 1g -states. E. Spin resolved STM In the previous section we investigated corrections to the LDOS which arise from scattering on a single non-magnetic charge impurity and found the triplet dominated A 1g and B 2g pairing states in the presence of SOC to be consistent with the available experiments. As discussed in the introduction, however, the more conventional spin singlet A 1g and B 2g states have also been proposed. This raises the question whether SOC effects and the spin triplet order parameter can be further verified in experiment. Here we propose spin-resolved QPI as an additional tool to further specify the underlying pairing state. To illustrate the possible capabilities of this technique, we study the corrections to the LDOS in Born approximation from a single magnetic impurity which, within our basis of eq.(23), is given bŷ U = τ 3 ⊗ σ z ⊗V . Here τ and σ matrices act on Nambu and spin-space, respectively, and the matrixV describes orbital scattering. The Fourier transform of the spin resolved σ i -projected correction to the LDOS is given by [34] with the real and the imaginary part being even and odd in q, respectively. Note that a non-vanishing imaginary part requires the scattering potential to break inversion symmetry, which, as we show below can result from interorbital scattering. The scattering matrixV =V z +V x can be written aŝ where, in Born approximation,V z potentially contributes to the σ z -andV x to the σ x(y) -polarized LDOS when taking the trace in eq.(32). The labels intra and inter denote intra-and interorbital scattering, respectively. Note that in each block of the matricesV z (V x ) it is the off-diagonal (diagonal) elements that break inversion symmetry since under inversion Ψ X (−k) = σ z Ψ X (k) and Ψ Y (−k) = −σ z Ψ Y (k). Those elements contribute to Imδρ σ i (q, ω), while inversion symmetric ones contribute to Reδρ σ i (q, ω). Results are presented in Fig.7 exemplary for ω = 8 meV where we show real and imaginary part of δρ σ i (q, ω) as a function of the scattering vector q for a σ z impurity with i = z (a-d) and i = x (e-h). Note that a similar behavior of these quantities is found for all frequencies within the range of the two maxima in the LDOS at 8meV and 14meV, respectively.' In Fig.7.(a,b) Reδρ σ z shows an even in q C 4 -symmetric QPI pattern qualitatively similar for the A 1g and B 2g pairing state. In contrast to that, see Fig.7.(c,d), Imδρ σ z is odd in q and exhibit a few qualitative differences between A 1g and B 2g especially along the intensity edges. Note, however, that Imδρ σ z is caused by the inter-orbital terms in V z and hence is expected to show a weakened intensity if V inter ≪ V intra . In Fig.7.(e-h) we present real and imaginary part of the σ x -polarized QPI pattern δρ σ x (q, ω). We would like to stress that this quantity is non-zero only in presence of either a triplet gap ∆ A(B) t or SOC, which then itself induces a triplet gap. The same is true for the σ y -polarized state which is related to δρ σ x (q, ω) by 90 • rotation. In Fig.7(e-f) we show the real part δρ σ x which is C 2 -symmetric and even in q while the imaginary part is odd, see Fig.7(g,h). In contrast to δρ σ z the patterns for A 1g and B 2g representation show qualitative differences in both real and imaginary part. Moreover, if the impurity occupies one Fe site (either on sub lattice A or B)V x contains scattering between xy X and xy Y components which breaks inversion symmetry and according to eq.(10) is of intraorbital nature. Hence, we expect the QPI signal of Imδρ σ x to be of the same order of magnitude as Reδρ σ z and hence may provide a quantity that allows to distinguish triplet driven A 1g from B 2g . Thus we have shown that the spin-resolved QPI is a useful tool to study the details of the spin-orbit coupling and the triplet order parameter; our results suggest that a combination of spin-resolved QPI with the more conventional probes can be used to determine the symmetry as well as spin structure of pairing in iron-based superconductors. V. CONCLUSION In conclusion, we remind the reader that the highly electron-doped Fe-chalcogenides, in particular the (Li 1−x Fe x )OHFeSe system, have been discussed intensively in the context of the standard model of spin-fluctuation induced spin singlet pairing in the limit of weak spin-orbit coupling. In this case, the nodeless d-wave (B 2g ) and bonding-antibonding s ± (A 1g ) spin singlet states on the electron pockets have been considered most favorable, with the additional possibility of incipient s ± pairing via coupling to the incipient hole pocket at Γ. One of these states may eventually prove to be the correct pairing state for these materials, if the effective attraction in these channels generated by spin fluctuations driven by interpocket repulsion dominate the pair vertex 11 . Here, we have instead studied the new possibility raised previously 20 , namely that interorbital triplet components generated by intrinsic attractions possible only in the presence of spin orbit coupling are responsible for the pairing in these systems. We explored the various possible Cooper-pairing symmetries using a mean-field decomposition of the Hubbard-Kanamori Hamiltonian including A 1g and B 2g states. For nonzero spin-orbit coupling, the superconducting order parameter is a combination of spin singlet and spin triplet gaps in each state. Treating attraction in singlet and triplet channels on equal footing and solving the selconsitency equations, we found for weak spin-orbit coupling a dominant spin singlet and small spin triplet gap yielding a state essentially equivalent to those identified in the usual spin fluctuation approach. For stronger spin-orbit coupling, however, the superconducting order parameter is a combination of spin singlet and dominant spin triplet gaps in each state. Focusing on the (Li 1−x Fe x )OHFeSe system, we identified the even parity A 1g -and B 2g -pairing states with a dominant spin triplet component to be consistent with available experiments, including current quasiparticle interference data, whereby according to our phase-diagram the A 1g state is slightly favored. The spin-singlet dominated A 1g and B 2g -states in this scenario without strong spin fluctuations are not consistent with at least one of the existing experiments. In summary, to obtain a full moderately anisotropic gap on the Fermi pockets and its sign-changing character in agreement with experimental results on (Li 1−x Fe x )OHFeSe, we require either the traditional intraband A 1g or B 2g states generated by spin fluctuations, or the new triplet interband pair states in the same symmetry channels generated by intrinsic attraction in multiorbital correlated models. It is clearly of interest to identify experimental tests to distinguish between these possibilities. To this end we proposed using spin-polarized QPI to identify the possible triplet components present in the more exotic alternative states, and presented results for each of the triplet dominated states. A clear identification of triplet interband pairing using these results would be an important step forward in understanding the unusual superconductivity in the Fe chalcogenides. Blue curves refer to the same bands for zero superconducting gap. The low-lying energy band has its minima centered above the Fermi-level "back-bending". In Fig.8(a) and Fig.8(d) we present the DOS and the energy band dispersion for the p-wave state. The DOS show two peaks tuned to lie at 8 meV and 14.3 meV, respectively and the lower energy band shows the back bending. Fig. 8(c) shows the absolute values of ∆ in and ∆ out projected on the Fermi surface pockets. In Fig.8 Here, one finds that the main feature of E u ⊗ U (1) in δρ − (ω) is that there are two sign changes between the peak energies of the total DOS (i.e. two negative peaks and positive values in between them), which does not agree with experiment 28 . Therefore this state does not seem consistent with the QPI data at least within Born scattering limit 28 . For consistency the spin-resolved QPI for the odd parity E u ⊗ U (1) p-state is presented in Fig.9. The SOC induced coupling between the spin singlet and triplet pairing channels translates into a coupling between intra-and inter-band order parameters in band space. It has been argued 21 that the second peak, seen in STM 28 can be at least partially due to an interband gap. In order to investigate how the predictions of HAEM's theory are affected by inter-band pairing we consider a simple model of a two band superconductor with superconductivity driven by intra-and inter-band pairing. . The band dispersions are assumed to be simple parabolic ones ξ 1 (k) = k 2 2m − µ 1 and ξ 2 (k) = k 2 2m − µ 2 . Hence, ξ 2 = ξ 1 − 2B with B = 1 2 (µ 2 − µ 1 ) > 0. We only distinguish between intraand inter-band pairing and neglect for a moment their spin symmetry (i.e. consider them to be spin singlet) to get a better understanding on how the QPI data is affected by the inter-band pairing. We linearize both bands, and calculate δρ − (ω) and investigate how it depends on ∆ 12 , the relative phase between ∆ 12 and ∆ 21 and the band offset (direct gap). For ω > 0 one needs to consider the cases: i) B < \|∆ 12 \| and ii) B > \|∆ 12 \| where in the former ω 1 and ω 3 and in the latter ω 2 and ω 4 correspond to the peak energies in δρ − (ω). i) B < \|∆\| In Fig.10, we plot δρ − (ω) in arbitrary energy units as a function of ω and B < \|∆\|. For ∆ 12 = ∆ 21 the behavior of δρ − is what we define here as "odd", meaning that between the two intraband gap energies δρ − changes sign, so the QPI-pattern is s ++ -like (solid blue curve). In case of ∆ 12 = −∆ 21 δρ − is "even", and one obtains as "s +− -like" pattern (solid orange curve). Hence, if the inter-band gap is larger than the band offset HAEM is sensitive to the relative phase between ∆ 12 and ∆ 21 . ii) B > \|∆\| If B > \|∆ 12 \| the situation is a different one as can be seen by the dotted curves in Fig.10. In both cases ∆ 12 = ±∆ 21 and δρ − is "odd", mimicking an s ++ -pattern. Consequently, HAEM's method is not sensitive to the relative phase between ∆ 12 and ∆ 21 . Therefore, in a system with a large dominant inter-band gap the QPI signal depends not only on the relative phase between ∆ 12 and ∆ 21 but also on the ratio between \|∆\| and B, hence on the band structure. Inter+Intra-Band The influence of the inter-band gap on the QPI does affect the HAEM results on sign-changing and sign preserving intra-band gaps. To show this we numerically present the δρ − (ω) for ∆ 11 = ∆ 22 and sign(∆ 11 ) = −sign(∆ 22 ) leading to an initial s ± pattern. Then we increase \|∆ 12 \| and show that at a certain magnitude the phase information is lost. The results are plotted in Fig.11 where in the left and right panel we have ∆ 12 = −∆ 21 and ∆ 12 = ∆ 21 , respectively. Our results for the LDOS are presented in Fig.11(a) and Fig. 11(b). If ∆ 12 = 0, see orange curve, only intra-band If we switch the inter-band paring ∆ 12 = 1 on, see black curves, peak 3 and 4 appear as a consequence of inter-band pairing, whereas peak energies 1 and 2 are only slightly affected. If ∆ 12 is increased further, see blue curves, peak 2 and 3 merge. In the absence of inter-band pairing the QPI pattern in Fig.11(c)-(d) is of s ± type, see orange curve. For moderate values ∆ 12 = 1 one can nicely distinguish between intra-and inter-band contributions Peak 1 and 2 show an s ± pattern due to the sign change between ∆ 1 and ∆ 2 , whereas between peak 3 and 4 the pattern is an s ++ since the inter-band gap is smaller that the band offset B, see Fig.10. For moderate values of ∆ 12 , see blue curves, peak energies 1 and 2 start to deviate from ∆ 1 and ∆ 1 as marked by the vertical dashed lines. As soon as peak 2 and 3 merge the s ± pattern between peak 1 and peak 2 becomes more and more difficult to resolve. The blue curve corresponds to the situation where the inter-band gap ∆ 12 is larger than the intra-band gap but smaller that the band offset. The latter condition causes an s ++ pattern between peak 3 and 4 independent of the relative phase between ∆ 12 and ∆ 21 , sec.C 1. This is accompanied by ∆ 11 and ∆ 22 having a s ± symmetry. The combination of both patterns leads to a sign-change of δρ − (ω) in the region 2 < ω < 4 which might be misinterpreted as a an s ++ pattern between ∆ 11 and ∆ 22 and hence carries no concrete phase information.	2019-03-18T00:01:48.721Z	2019-03-14T00:00:00.000	{ "year": 2019, "sha1": "dfc30d21ff852d16cd0e4c213ce51466ec9cece7", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1088/1367-2630/ab2a82", "oa_status": "GOLD", "pdf_src": "Arxiv", "pdf_hash": "47421e3efe4504b5f61785e97f0d00cbb03dd2a6", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
6780295	pes2o/s2orc	v3-fos-license	Evidence for a Fourteenth mtDNA-Encoded Protein in the Female-Transmitted mtDNA of Marine Mussels (Bivalvia: Mytilidae) Background A novel feature for animal mitochondrial genomes has been recently established: i.e., the presence of additional, lineage-specific, mtDNA-encoded proteins with functional significance. This feature has been observed in freshwater mussels with doubly uniparental inheritance of mtDNA (DUI). The latter unique system of mtDNA transmission, which also exists in some marine mussels and marine clams, is characterized by one mt genome inherited from the female parent (F mtDNA) and one mt genome inherited from the male parent (M mtDNA). In freshwater mussels, the novel mtDNA-encoded proteins have been shown to be mt genome-specific (i.e., one novel protein for F genomes and one novel protein for M genomes). It has been hypothesized that these novel, F- and M-specific, mtDNA-encoded proteins (and/or other F- and/or M-specific mtDNA sequences) could be responsible for the different modes of mtDNA transmission in bivalves but this remains to be demonstrated. Methodology/Principal Findings We investigated all complete (or nearly complete) female- and male-transmitted marine mussel mtDNAs previously sequenced for the presence of ORFs that could have functional importance in these bivalves. Our results confirm the presence of a novel F genome-specific mt ORF, of significant length (>100aa) and located in the control region, that most likely has functional significance in marine mussels. The identification of this ORF in five Mytilus species suggests that it has been maintained in the mytilid lineage (subfamily Mytilinae) for ∼13 million years. Furthermore, this ORF likely has a homologue in the F mt genome of Musculista senhousia, a DUI-containing mytilid species in the subfamily Crenellinae. We present evidence supporting the functionality of this F-specific ORF at the transcriptional, amino acid and nucleotide levels. Conclusions/Significance Our results offer support for the hypothesis that “novel F genome-specific mitochondrial genes” are involved in key biological functions in bivalve species with DUI. Introduction Apart from the nucleus, mitochondria are the only known organelles with their own DNA in animal cells. Given the abundance of animal mitochondrial DNA (mtDNA) in animal tissues, strict maternal inheritance, the different evolutionary rates of its genes and the absence (or very low level) of recombination [1][2][3], this genome has come to be considered a reliable and robust marker for phylogenetic and population genetic studies as well as a model for the study of genome evolution [3,4]. Comparative mitochondrial genomics has revealed that animal mtDNAs are very conserved in terms of gene content [4]. These small circular and typically intron-less molecules encode 2 ribosomal RNAs, 22 transfer RNAs and 13 protein subunits of the mitochondrial respiratory chain complexes and ATP synthase. The other subunits of the electron transport system (i.e., ,65 subunits in mammals) as well as all the proteins and factors involved in other mitochondrial functions, such as mtDNA replication and mtDNA expression, are nuclear-encoded [4,5]. However, with the increasing number of published mitochondrial genome sequences, examples of species that deviate from the gene content norm have been described in different animal groups including multiple species lacking one of the standard mitochondrial protein-coding genes [3]. In contrast, additional mitochondrial protein-coding genes, usually found in mtDNAs of the closest unicellular relatives of animals (e.g., mutS, dnaB, atp9, tatC in protists), have only been identified and annotated in the mitochondrial genomes of nonbilaterian animals (i.e., Cnidaria and Porifera; [3]). One intriguing observation that emerged from sequencing studies of whole animal mtDNAs is the occurrence of numerous open reading frames (ORFs) of unknown function that are present in closely related species but for which homologues cannot be determined among more distantly related species [3,[6][7][8][9][10][11]. In bacterial and eukaryotic nuclear genomes, unique proteins that do not have recognizable homologues in other organisms (or that exist only in very closely related organisms) are commonly called ''lineage-specific'', ''taxonomically restricted'' or ''orphan'' genes [12][13][14]. Interestingly, these lineage-specific genes have been shown to be involved in key biological functions and important adaptive processes [13][14][15]. For example, it has been demonstrated that species-specific differences in tentacle formation in the cnidarian genus Hydra correlate with expression of a taxonomically restricted gene encoding a small secreted protein of ,85-105 amino acids [13]. Thus, the lineage-specific open reading frames that occur in animal mitochondrial genomes could potentially have functional significance. Lineage-specific mtDNA-encoded proteins are already known to play a role in sex determination in angiosperm plants exhibiting cytoplasmic male sterility [16]. Recently, we demonstrated the expression of two novel, sexassociated mtDNA-encoded proteins, the F-and M-ORFs, in freshwater mussels (Bivalvia: Unionoida) [11]. Although bioinformatics tools have not allowed us to characterize these genes, we have found that the female-transmitted ORF protein is not only present in mitochondria but, more surprisingly, it is also present on the nuclear membrane and in the nucleoplasm of eggs [17]. These results established novel features for animal mitochondrial genomes: the presence of additional, lineage-specific, mtDNAencoded proteins with functional significance and the involvement of mtDNA-encoded proteins in extramitochondrial functions. Interestingly, this discovery has been made in the only known animal group that does not transmit its mtDNA exclusively maternally [18,19]. Unlike the system of strict maternal inheritance described in other animal species, bivalves belonging to the orders Mytiloida, Veneroida and Unionoida instead possess a system of doubly uniparental inheritance (DUI) of mtDNA [18,19]. Specifically, DUI is a ''mother-to-daughter'' and ''father-to-son'' mtDNA inheritance system where females transmit their mt genomes (F mtDNA) to both sons and daughters, and males transmit their mt genomes (M mtDNA) to their sons [20][21][22][23]. Female offspring are typically homoplasmic ( = containing one mt genome) and male offspring are heteroplasmic [21,23] but see [24,25]. In marine mussels Mytilus spp., male somatic tissue contains predominantly the mtDNA of the mother (F genome) but male gametes contain exclusively the mtDNA of the father (M genome) [26,27]. In contrast, the M-type genome is predominant in male somatic tissues of the marine clam Venerupis philippinarum, and also exclusive in male gametes [28]. Remarkably, amino acid sequence divergences between sex-associated mtDNAs can reach 20% (uncorrected p-distance) in marine mussels Mytilus and up to 50% in freshwater mussels [29][30][31]. The latter observation is likely due, in part, to the relative stability and antiquity of the unionoid bivalve F and M genomes, i.e., these two mt genomes have been separately transmitted for .200 my. The newly identified F-and M-ORFs in freshwater mussels are not the only mitochondrial novelties with functional significance in these bivalves. It has been proposed that the C-terminus coding extension of the COX2 protein (Mcox2e), which is unique to freshwater mussel M genomes [32][33][34][35][36], could represent an ''Mspecific label'' for sperm mitochondria that determines their fate in the fertilized eggs (as observed in Mytilus [37,38]). In the marine mussel Mytilus spp., it has been proposed that the primary candidate for sequences that control the mode of inheritance of the two mitochondrial genomes would reside in the first variable domain (VD1) of the control region (CR) [39,40]. The Mytilus CR can be divided in three domains based on indels and nucleotide variation [39,40] (see also Figure 1): the first variable domain (VD1), which is the longest region of the CR, followed by a highly conserved middle domain (CD) and then a second variable domain (VD2), which is the shortest region of the CR. While the average DNA divergence between F and M genomes over the whole molecule may reach ,20%, CD has diverged by only 1.5%, VD2 by about 15%, whereas VD1 is the most divergent part of the entire mt genome with DNA divergences averaging 50% [39][40][41]. Clearly, VD1 is under different, potentially sex-specific selective constraints, suggesting that it could play different roles in the F and M genomes [39,40]. An interesting difference between marine mussels and freshwater mussels is that only in the former group have F mt genomes periodically experienced ''role-reversal events'' and invaded the male route of inheritance, resulting in the formation of new M mt genomes [42][43][44]. These new M mt genomes (known as recently masculinized M types) are essentially recombinants composed of an F genome's coding and control regions with an additional CR from a so-called ''standard M'' genome inserted into the F-type CR [30,[45][46][47]. It has therefore been suggested that the incorporation of sequences from the CR of a standard M genome into an F genome could be responsible for the paternal transmission route of the recombinant mtDNA genome [18,30,39,47,48]. However, attempts to confirm mitochondrial sequences and/or mt encoded proteins that are responsible for the different modes of mtDNA transmission under DUI have not been successful. To date, much of the work on marine mussels, Mytilus spp., has focused on F vs. M-specific mtDNA motifs in the CR as potential features that could determine whether a genome will follow the maternal or the paternal mode of inheritance [39,40,45,46]. Herein, we re-investigated all complete (or nearly complete) F, standard M and recently masculinized M mytilid mt genomes and control regions previously sequenced for the presence of ORFs that could have functional importance in these bivalves. Our results confirm the presence of a novel F genome-specific ORF of significant length (.375 nt), located in VD1, that most likely has functional significance in marine mussels (Mytilus spp.). The identification of this ORF in the relatively closely related Mytilus edulis, M. galloprovincialis and M. trossulus and in the more distantly related, also DUI-containing, M. californianus [49] and M. coruscus (evidence from GenBank) suggests that it has been maintained in the mytilid lineage (subfamily Mytilinae) for ,13 million years [50]. Furthermore, this ORF likely has a homologue in the F mt genome of Musculista senhousia, a DUI-containing mytilid species in the subfamily Crenellinae [51]. Our results offer support for the hypothesis that ''novel mt genome-specific genes'' are involved in key biological functions, such as mtDNA transmission, in bivalve species with DUI. Results and Discussion Identification of open reading frames (ORFs) in the control regions of Mytilus mt genomes To assess whether F and M mitochondrial control regions (CR) could possess ORFs that could have functional importance in Mytilus bivalves, we first investigated complete mtDNAs previously sequenced for the Mytilus edulis species complex (i.e., M. edulis, M. galloprovincialis and M. trossulus of the subfamily Mytilinae). Our results indicate that the VD1 of the F-type CR contains one conserved ORF (F-orf-vd1) of substantial length (.100aa). F-orf-vd1 has complete start and stop codons and is located on the same strand as all other mtDNA-encoded genes ( Figure 1 and Tables 1 and 2). The predicted length of the F-ORF-VD1 protein is 163 amino acids (aa) for M. edulis and M. galloprovincialis and 153aa for M. trossulus. Interestingly, F-ORF-VD1 was also detected in the Ftype CR sequences of the more distantly related species M. californianus (129aa) and M. coruscus (127aa). These results are consistent with the hypothesis that the F-ORF-VD1 region represents a new Mytilus mitochondrial gene with biological significance. In contrast to the F-type CR, conserved ORFs of similar sizes were not found in any of the M-type VD1 regions ( Figure 1 and Tables 1 and 2). Assessing homology using a combination of sequence and position similarity, M-type ORFs of 94aa and 112aa were found in M. edulis VD1, ORFs of 21 and 42aa were found in M. galloprovincialis and ORFs of 73aa and 74aa were found in M. trossulus VD1 (Tables 1 and 2). Notably, ORFs of different sizes (24aa to .100aa) were identified within each species in the other complete M-type VD1 sequences available in GenBank, including the more distantly related M. californianus (30 and 32aa) ( Figure S1). Moreover, analysis of these M-type ORF sequences using the testcode algorithm [52], which recognizes potential protein-coding sequences by evaluating the distribution of nucleotides at the third codon positions within a reading frame, suggests that they are non-coding (Table 3; probability of coding ,30%). The gene-finding program Glim-mer3 [53], which uses an interpolated Markov model scoring algorithm that computes the log-likelihood that a given interval on a DNA sequence was generated by a model of coding versus noncoding DNA, also failed to identify these M-type ORFs as putative protein-coding genes (data not shown). Contrary to the testcode algorithm that does not provide reliable results for sequences ,200 bp [52], Glimmer3 is highly precise and sensitive to find protein-coding genes as small as 90 bp and usually detects .98% of genes in prokaryotic genomes with a limited number of false positive predictions [53]. It is also a very effective gene finder for Figure 1. A typical Mytilus mitochondrial genome. All genes are encoded on the same strand. Gene identities: nd1-6 and nd4l, NADH dehydrogenase subunits 1-6 and 4L; cytb, cytochrome b; cox1-3, cytochrome c oxidase subunits I-III; atp6-8, ATP synthase subunit 6 and 8 (proteincoding genes in white); 12SrRNA and 16SrRNA, small and large subunits of ribosomal RNA (in light gray). Transfer RNA genes are depicted by oneletter amino acid codes (in gray). The red and blue lines at the inner periphery of the ring represent EST sequences for the F and M mt genomes of M. edulis/M. galloprovincialis, respectively. Schematics of the structure of a typical F-type (left) and M-type (right) control regions, which are located between the 16SrRNA and trnY genes, are shown. The F-ORF-VD1 is identified in the F-type control region. CR, control region; CD, conserved domain; VD1, variable domain 1; VD2, variable domain 2 [39,40]. The mean size of each domain of the CR is shown. The ''standard'' F-type CR of M. trossulus, which is a F/M recombinant CR, is not presented. doi:10.1371/journal.pone.0019365.g001 eukaryotic genomes [54], and its accuracy to identify unannotated genes has been convincingly demonstrated by laboratory experiments [53]. For example, Glimmer3 predicted 16 out of 17 new proteins confirmed by protein-based experiments on the archaeon Pyrococcus furiosus [53]. Only the 13 typical mitochondrial proteincoding genes were successfully identified in the Mytilus M genomes using Glimmer3 (no additional ORF were found on the coding strand). These results support previous inferences that the M-type VD1 might function at the DNA or RNA level because of the presence of potential tRNA-like secondary structures in this domain [39,40]. The situation for the F-type VD1 is remarkably different. Specifically, ''full length'' F-ORF-VD1s were found in all M. californianus VD1 (n = 3), in all but 3 (12/15) M. trossulus VD1 and in all but 8 (41/49) of the M. edulis and M. galloprovincialis VD1 that have been completely sequenced to date ( Figure S2) [39,40,45,46,[55][56][57]. It is worth noting, however, that all three ''truncated'' vs. ''full-length'' M. trossulus F-ORF-VD1s (i.e., those with109aa instead of 153aa) and 7 of the 8 truncated M. edulis/M. galloprovincialis F-ORF-VD1 (84 to 144aa instead of 163aa) were found in recombinant CR sequences that consist of both F-type and M-type CR segments or in duplicated F-type CRs [40,46,56]. In M. trossulus, all three truncated ORF sequences were due to a guanine base deletion at position 295 (out of 462 nt) in a segment consisting of a stretch of 5 Gs in ''full length'' M. trossulus F-orf-vd1 sequences ( Figure S2). However, this deletion was absent in all partially sequenced M. trossulus F-type VD1 available in GenBank (n = 156), which correspond to the first 407 nt of the F-orf-vd1 and, when translated, to the first 135aa of the F-ORF-VD1 without any stop codon. These observations raise the possibility that the 3 truncated M. trossulus F-orf-vd1 sequences might represent sequencing errors. In the case of M. edulis/M. galloprovincialis, all ''truncated'' F-orf-vd1 sequences observed in recombinant CR sequences were also found to occur only in sperm, i.e., these haplotypes were consistently absent from females [46], suggesting that they could represent recently-masculinized CR sequences (see below). A unique ''truncated'' F-orf-vd1 sequence, due to a nucleotide insertion in a non-recombinant CR sequence, has been found in the complete M. edulis F genome ( Figure S2), which was obtained from cloning experiments [58][59][60]. Multiple rounds of cloning or a sequencing error could explain this particular exception. Support for identifying F-orf-vd1 as a protein-coding gene Support at the transcriptional level. The maintenance of ''full length'' F-orf-vd1 regions in the closely related Mytilus edulis, M. galloprovincialis and M. trossulus as well as in the more distantly related M. californianus and M. coruscus, which represents ,13 million years of Mytilus mussel evolution [50], strongly argues in favor of functionality for this open reading frame. The hypothesis that the F-type but not the M-type VD1 encodes a protein is consistent with previous observations that F and M VD1 are under different selective regimes and likely explains why the intergenomic DNA divergences between F and M VD1 are the highest for Mytilus mitochondrial genomes [39,40]. As listed in Table 2, the mean DNA divergences between aligned portions of the putative F-orf-vd1 and the M-type VD1 within each species exceed by far those observed for the recently identified, rapidly evolving atp8 gene [61]. The maintenance of a functional ORF only in the F lineage would explain not only the high intergenomic divergences but also support the hypothesis that VD1 has a sexspecific function [39,40,45], specifically, that the F-ORF-VD1 is a novel mitochondrial protein with a F-specific function in Mytilus mussels. In support for such a role, testcode predictions of protein coding function for the F-orf-vd1sequences are, with the exception of M. coruscus (probability of coding = 30%), all very high ( Table 3). The gene-finding program Glimmer3 [53] also predicts the protein coding nature of the F-orf-vd1 sequences. The program attributes a score to each orf, providing a consistent scale to compare coding potential scores of different orfs [53]. For example, the Mytilus edulis F-orf-vd1 presents a higher coding potential score (8.99) than the typical mitochondrial proteincoding genes cox1-cox2-cox3, cob, nd3-nd4l, and atp8 (3.93 to 8.26), and a lower score than nd1-nd2-nd4-nd5-nd6, and atp6 (9.23 to 10.77). The reason why testcode classified the M. coruscus F-orf-vd1 sequence as non-coding and the M. californianus and M. senhousia Forfs as having 77% probability of coding could be explained by the high variability of this putative gene (see below). Indeed, fastevolving genes are often rated as non-coding by the testcode algorithm, presumably because the mechanisms generating diversity are stronger than the ones encouraging consistent codon preference [52]. Interestingly, corroborative evidence for the protein-coding nature of F-orf-vd1 was also obtained from BLASTN searches against dbEST (Expressed Sequence Tag division, EST_others). For example, for the more extensively studied M.edulis/M. galloprovincialis, a total of 366 and 194 ESTs were aligned to the complete F and M mt genomes with nucleotide identity .96%, respectively (out of 24,611 ESTs from .20 different polyadenylated cDNA libraries; [62][63][64][65][66]). The ESTs cover 14,994 bp (89.6%) of the F mt genome and 9,868 bp (59.3%) of the M mt genome. Figure 1 reports the M. edulis/M. galloprovincialis F and M ESTs mapped on the completely sequenced Mytilus mtDNA. The majority of ESTs (n = 168 for the F genome and n = 156 for the M genome) are derived from 16S and 12S rRNAs, suggesting a higher expression level and/or a higher stability compared to other mt genes. Given that ESTs come from multiple cDNA libraries constructed using different methods [62][63][64][65][66], and because it is absent in both F and M mt genomes and thus less likely to be the result of the trimming of low quality sequences or the cloning procedure (e.g., [66,67]), the lack of ESTs corresponding to the F-and M-type nad4L gene suggests that this transcript might be expressed at low levels and/or might be rapidly degraded in M. edulis/M. galloprovincialis. Similarly, the absence of several tRNA-like ESTs for both F and M genomes could be explained by their removal from mature polyadenylated transcripts [67,68]. Furthermore, the typical utilization of Mytilus somatic tissues in cDNA library preparations [62][63][64][65][66] can explain the lower coverage of the M mt genome by M-type ESTs, given that somatic tissues predominantly contain and express the F-type mtDNA [26]. Remarkably, the EST analysis yielded 11 significant hits for M. edulis and M. galloprovincialis F-orf-vd1 sequences whereas neither the second variable domain (VD2) of the F-type control region nor the M-type VD2 was present in ESTs ( Figure 1; the 11 hits are AJ626121, AJ626242, AJ626130, AJ626120, AJ 626205, AJ626443, AJ626444, AJ626129, AJ626131, AJ623360, AJ624518). Because only the last 270 bp of F-orf-vd1 are covered by EST sequences (Figure 1), one could argue that these transcripts represent by-products from an unprocessed polycistronic transcript precursor. However, the observation that 6 other mitochondrial genes (i.e., cox1, cob, nad1, nad2, nad4, and nad5) were not represented by ''full-length ESTs'' containing the entire gene, i.e. only partial transcripts were found in dbEST (data not shown), suggests a reduced enrichment in full-length cDNAs in Mytilus libraries. Moreover, since the VD1-like ESTs have been obtained by oligo-dT priming of mussel mRNA [64,65], they are expected to originate from polyadenylated mature transcripts. Taking together, these findings suggest that the F-type VD1 is expressed in the Mytilus mitochondrial proteome. On the other hand, since we also found significant hits for F-type CD (3 hits) and part of M-type VD1 and CD sequences (2 hits) (see Figure 1), the possibility remains that all of these polyadelynated transcripts function at the RNA level. Polyadenylated transcripts derived from a putative non-coding region have previously been reported in the oyster Crassostrea gigas [67]. The authors hypothesized that this intergenic segment located between the atp6 and nd2 genes could represent the mitochondrial control region, which is polyadenylated at a high level in several mammal species [67][68][69][70]. However, it appears that this intergenic segment in C. gigas actually contains the ''formerly reported as missing'' atp8 gene in these bivalves [61]. Interestingly, the polyadenylated mitochondrial CR sequences observed in mammals have been proposed to be multifunctional molecules serving as primers for mtDNA replication, regulators for replication and translation processes through rRNA binding as well as protein-coding mRNAs [68][69][70][71][72][73]. For example, Nakamichi et al. [73] reported a CR transcript in humans that could code for a peptide of 76 amino acids. The possibility thus remains that CR transcripts in Mytilus function in different ways, i.e., that F-orf-vd1 functions at both RNA and protein levels (or at the protein level only) and that other CR transcripts function at the RNA level. Another hypothesis would be that the M-type VD1 also functions at the protein level but that its function is supported by smaller ORFs such as those found in the M-type VD1 domain of M. californianus. Further data collection and investigation will be essential to clarify the functional role of these CR transcripts and ORFs. Support at the amino acid level. The analysis of the taxonomic distribution of the F-ORF-VD1 in mytilid mussels is one other important step in the assessment of its potential functional role as a protein. To establish whether the F-ORF-VD1 is taxonomically restricted to the genus Mytilus or if it is an evolutionary feature of mytilid mussels, we screened for the presence of F-specific ORFs in the newly sequenced mitochondrial genome of the DUI-containing mytilid Musculista senhousia from the subfamily Crenellinae (Passamonti et al. submitted). Within the Bivalvia, mytilid mussels form a monophyletic group where Musculista and Mytilus are invariably clustered together, while freshwater mussels (Unionoida), including the species Venustaconcha ellipsiformis, are confirmed basal and fully separated from all other autolamellibranchiate lineages, including the Mytilidae ( [31,74]; Plazzi & Passamonti unpublished). Interestingly, we found one relatively large unassigned region specific to the F genome of M. senhousia (UR2 = 543 bp) preceding the control region and containing an ORF of considerable length (121aa). The M. senhousia F-ORF possesses complete start and stop codons, is located on the same strand as all other mtDNA-encoded genes, and has a probability of coding of 77% (Tables 2 and 3). At the amino acid level, comparisons among predicted sequences for Mytilus spp. F-ORF-VD1 and the M. senhousia F-ORF revealed this putative gene as the least conserved in the Mytilus F lineage, with aa sequence identities ,1.5-2.5 times lower than those obtained for the highly variable ATP8 protein, and among the least conserved in mytilid mussels (Table 4). For example, 24% amino acid identities are observed for the F-specific ORFs between the distantly related Mytilus edulis and Musculista senhousia species, whereas 18% amino acid identities are observed for ATP8 (and 74% for COX1). Figure 2 shows the alignment of all mytilid F-specific ORFs. The greatest similarity among all species is principally found within a stretch of 60 residues in the middle of the protein sequence. As expected for a rapidly evolving protein, sequence differences between the F-ORF from M. senhousia and the F-ORF-VD1 from Mytilus spp. are more pronounced (Figure 2). These results suggest that, if they are functionally equivalent proteins, constraints on M. senhousia F-ORF and Mytilus F-ORF-VD1 are imposed at higher levels of protein structure rather than the amino acid sequence level. In support of this hypothesis, the amino acid compositions of the M. senhousia F-ORF and Mytilus F-ORF-VD1 are slightly different ( Figure 3A), but their compositions of chemically equivalent amino acids (i.e., with similar properties) are similar ( Figure 3B) and comparable to what is observed for the fast-evolving mtDNA-encoded protein ATP8 ( Figure 3C & D). In contrast, the translated M-ORF found in the VD1 domain of M genomes shows high variability with regards to amino acid composition ( Figure 3E & F). These results, which can be explained by the presence of frameshifts, premature stop codons and important differences in length in M-ORF sequences, are in agreement with the hypothesis that only the mytilid F-ORF codes for a functional protein. If the F-ORF would not be functional at the protein level, one would expect the presence of within-and between-species ''coding disablements'', such as the numerous frameshift mutations and premature stop-codons seen in M-ORF sequences. Furthermore, amino acid compositional similarities among mytilid species (or even between F-ORF sequences and protein-coding genes within a same genome) would not be expected from non protein-coding sequences since they are not subject to selective pressure to preserve protein structure and function [75]. In this latter case, however, similar mtDNA nucleotide bias and common evolutionary history could at least partly explain the observed results [75][76][77]. For example, it has been demonstrated that amino acid frequencies in proteins or in translations of randomly selected non-coding sequences are changing in response to the genomic change in G+C (or A+T) content, that is GC-rich codons and corresponding amino acids will increase in frequency in proteins and translated non-coding sequences in genomes with increasing G+C content, whereas ATrich codons and corresponding amino acids will increase in frequency in proteins and translated non-coding sequences in genomes with increasing A+T content [75,76]. A+T contents are relatively similar for both F-ORF-VD1 and ATP8 in Mytilus spp. (52-57% for the F-ORF-VD1 and 58-59% for ATP8), whereas Musculista values are higher with 68% A+T for the F-ORF and 68.9% for ATP8. Proportions of AT-rich codons and corresponding FYMINK amino acids [75] are slightly higher in Musculista for the F-ORF (25.6% vs. 13-21.7% for Mytilus) but comparable for ATP8 (26% vs. 26-28.5% for Mytilus). Although preliminary, these results indicate a potential correlation between DNA composition and amino acid compositional similarities of F-ORF sequences but they do not rule out the hypothesized protein-coding function of the F-ORF in mytilid mussels. In addition and of significant relevance here is our observation that a single, conserved predicted transmembrane helix (TMH) is present in the N terminal portion of all mytilid F-ORF proteins (Figure 4), suggesting that this putative gene would assume the underlying DNA composition of the mtDNA to the extent that this does not interfere with the secondary structure and biochemical function of the protein. Like typical animal mitochondrial genes, which all encode TMH proteins of the oxidative phosphorylation system in the inner mitochondrial membrane [4,5], the F-ORF protein could be an element of the electron transport chain or ATP synthase complex in mytilid mussels. However, the recent finding that F-ORF proteins likely play a role in sex determination in unionoid bivalves indicates that the mytilid F-ORF proteins could also have a non-oxidative phosphorylation function [17]. Although our results suggest stabilizing selection on the F-ORF region's amino acid composition and secondary structure and support the protein-coding hypothesis, further protein-based analyses will be necessary to characterize the biological significance of the mytilid ORFs, and to verify if they are functionally equivalent. Additional complete mt genomes, from mytilid and non-mytilid bivalves, are needed to elucidate the number, taxonomic distribution, and evolution of uncharacterized ORFs in this group of molluscs. Support at the nucleotide level. Due to alignment issues, nucleotide-level analyses were performed using the more similar Myilus spp. F-orf-vd1 sequences (i.e., we excluded Musculista senhousia). Comparisons of synonymous substitutions per synonymous site (Ks), nonsynonymous substitutions per nonsynonymous site (Ka) and Ka/Ks ratios within and between Mytilus species also provide evidence that F-orf-vd1 encodes a functional protein: within species Ka/Ks ratios are higher than between species Ka/Ks ratios, which are usually well below 1 and thus indicate purifying selection (Table 5). A lower between-species Ka/Ks is a common finding for mtDNA-encoded protein genes in animals and is explained by the elimination of mildly deleterious polymorphisms from populations before fixation [78][79][80]. Such results would not be expected for non-protein coding sequences. A more exhaustive analysis was undertaken to test the null hypothesis of neutrality and search for the signature of purifying and/or positive selection by calculating the Ka/Ks ratio at each codon site with the SELECTON program using a Bayesian approach [81]. In essence, neutrality is indicated by Ka/Ks = 1, purifying selection by Ka/Ks,1, and positive selection is usually invoked as a possible explanation for rare cases where the pattern Ka/ Ks.1 is observed. The idea is that substitutions at synonymous sites are largely selectively neutral relative to the intensity of selection at nonsynonymous sites and very low proportion of amino acid replacement can be interpreted as a reflection of purifying selection maintaining a functional protein. By contrast, under positive selection, rapid replacement of an amino acid is advantageous to the organism; hence, nonsynonymous mutations are fixed at a rate higher than that of neutral synonymous ones [81,82]. According to our results, the MEC selection model was significantly preferred than the null model M8a (AICc score for MEC is 8195.54 while AICc score for M8a is 13981.56) and suggested 35 putative positively selected residues. For the full alignment and Bayesian Ka/Ks ratios obtained from the models, see Figure S3. Specifically, Mytilus spp. COX1 residues were all found to be under strong purifying selection whereas of the 165 amino acid positions in the F-ORF-VD1 alignment portion, 125 residues (76%) were found under purifying selection, 5 (3%) were under neutral selection and 35 (21%) were found to possess a Ka/ Ks.1, indicating positive selection. However, the inference of positive selection was not considered statistically significant for any particular residue. Most of the unconserved 35 positions were located in the C-terminal portion of the protein (54%), whereas 23% of them were found within the more conserved stretch of 60 residues in the middle of the protein sequence (see Figure 2) and 17% were found in the TMH portion of the Mytilus spp. F-ORF-VD1 (data not shown). Overall, our results suggest that even if most of the sites are subjected to purifying selection, which is suggestive of a functional constraint, the presence of several sites with Ka/Ks.1 indicate that the F-orf-vd1 is a fast evolving gene in the Mytilus F genome. As recently proposed for the highly variable atp8 gene in these species [83], relaxed purifying selection coupled with the compensation-draft feedback process [80] could cause the faster evolution of F-orf-vd1. Specifically, the compensation-draft feedback process postulates that fixation of a mildly deleterious mutation favors compensatory mutations within the same or interacting polypeptides, which in turn can result in fixation of new mildly deleterious mutations by genetic draft due to the linked, non-recombining genes in mitochondrial DNA [80]. The compensation-draft feedback process could have been initiated by selection for an F-specific function for the F-ORF-VD1. Interestingly, the newly discovered F and M lineage-specific proteins in freshwater mussel species are also among the fastest evolving proteins coded by freshwater mussel mitochondrial genomes [11,17]. These findings suggest that mt lineage-specific genes or DNA regions are potential targets for positive selection and thus they might play an important role in bivalve speciation (i.e, mitochondrial populations of the same species could quickly diverge, and possibly become reproductively isolated because of mitochondrial-nuclear incompatibilities; [84]). Novel mtDNA-encoded genes in bivalve species with DUI Assuming a single origin of DUI [31,85,86], the F lineage specific ORFs in both marine and freshwater mussels could represent homologous genes. However, as is the case for the novel, F-and M-specific, mtDNA-encoded proteins in freshwater mussels [11,17], the precise function of the F-ORF-VD1 protein in marine mussels remains unclear. In freshwater mussels, no significant amino acid sequence similarity with known proteins was found for the F-ORF using BLAST Tools, but the estimated tertiary structure of the F-ORF from the species Venustaconcha ellipsiformis is consistent with involvement of this novel mitochondrial protein in DNA replication and/or DNA binding [17]. In the present study, sequence similarity searches for Mytilus F-ORF-VD1 using PSI-BLAST [87] against non-redundant protein sequences and SWISSPROT databases also failed to detect significant sequence similarity with known proteins. However, searching against the Protein Data Bank (PDB; [88]) revealed that the M. edulis and M. galloprovincialis F-ORF-VD1 exhibit relatively weak sequence similarity (E-value of 0.004) to an archaebacterial DNA helicase, suggesting that it could be a DNA-binding protein involved in regulation of mitochondrial DNA replication and/or transcription as might be the case for the F-ORF in freshwater mussels. To our knowledge, helicase genes have never been reported in animal mitochondrial DNA [3,89]. However, a putative helicase has been reported in the mitochondrial genome of the plant Marchantia polymorpha [90]. Moreover, the possibility of open reading frames in the mitochondrial control region playing a role in the replication and/or transcription process has been previously reported in some mammals [6] and Paramecium [91]. It is also worth noting that many of the ''unusual'' protein-coding genes discovered in invertebrate mitochondrial genomes contain amino acid patterns characteristic of interaction with DNA [3,8,92,93]. However, at this moment, it is not possible to confirm the hypothesis that the F-specific ORFs in marine and freshwater mussels are homologous due to their highly divergent nature and incomplete knowledge regarding their phylogenetic distribution. Irrespective of a common vs. independent origins for the Fspecific ORFs in marine and freshwater mussels, there are at least three possibilities for their source: (i) a gene homologous to ancestral bacterial protein-coding genes, (ii) a duplicated and diverged mitochondrial gene or (iii) a transfer from the nucleus to the mitochondrion [93,94]. Again, because of their relatively fast evolutionary rate, the F-specific ORF sequences have probably changed to such an extent that their historical antecedents are no longer recognizable at the aa sequence level. Based on currently available data, the F-ORF in freshwater mussels has persisted for .200 my [11,17]. As is the case for F-ORF-VD1 in Mytilus spp. and F-ORF in M. senhousia, the sequence similarity is low among distantly related freshwater mussels species, but selection has maintained at least one aspect of the secondary structure of the protein: one predicted TMH in the N-terminal portion of the protein [11,17]. However, the amino acid divergences and differences in amino acid composition between the mytilid ORFs and the F-ORF protein of the freshwater mussel species Venustaconcha ellipsiformis are much more pronounced than between the Mytilus spp. F-ORF-VD1 and the M. senhousia F-ORF ( Figure S4). Again, further protein-based analyses will be necessary to characterize the biological significance and critically evaluate the hypothesized functional equivalence of the F-specific ORFs in bivalves. The study of masculinized genomes to identify sequences responsible for mitochondrial transmission mode As mentioned earlier, a phenomenon that characterizes marine mussels is that female-transmitted mt genomes have periodically experienced ''role-reversal events'' and invaded the male route of inheritance, resulting in the formation of new M mt genomes [42][43][44]. Previous sequencing studies have demonstrated that all ''recently-masculinized'' or RM-mitotypes examined to date in mytilid mussels are recombinants composed of an F genome's genes and CR plus an additional M-type CR [30,[45][46][47]. Consequently, it has been hypothesized that recombination with the introduction of a ''Standard Male'' or SM-type CR into an otherwise female type mt genome could be the first step in the masculinization process [48]. However, to establish that a genome is masculinized, one needs to demonstrate that the genome is the exclusive mtDNA molecule in the sperm of the male from which it was extracted [40,47]. Indeed, RM-type sequences obtained from male gonad DNA extractions could be artifacts due to somatic tissue contamination. This logic makes the fully sequenced ''C genome'' of M. galloprovincialis, which was extracted from spermatozoa that were forced to swim through a Percoll TM solution to remove of any debris from somatic cells, the only verified masculinized genome sequenced to date [47]. We will thus mainly refer to this sequence in the section below. Because the first variable domain VD1 has been identified as the most likely site for sequences that could control the mode of inheritance of the mitochondrial genome [39,40], examination of RM-type VD1 sequences is essential to address the hypothesis that these sequences could determine maternal vs. paternal inheritance. The control region of the recently masculinized, male-transmitted ''C genome'' of M. galloprovincialis is composed of an F-type VD1 followed by an M-type CD, an M-type VD2 and a truncated Mtype VD1 (i.e., VD1 F /CD M /VD2 M /DVD1 M ) [40,47]. After VD1 F , the segment ''CD M /VD2 M /DVD1 M '' is repeated tandemly three times. The third repeat unit is followed by one complete CD M and one F/M recombinant VD2 ( Figure 5). From an ''ORF point of view'', the M. galloprovincialis ''C genome'' is different from the standard F-type mtDNA of the same species in having a truncated (at the C-terminus) F-ORF-VD1 protein (139aa instead of 163aa; Figure 5). It has to be noted that the deleted protein region does not involve the conserved 60aa stretch observed among mytilid species. We have shown that standard F genomes consistently possess a ''full length'' F-ORF-VD1 (with one exception that could represent a sequencing error), suggesting the hypothesis that a ''full length'' F-ORF-VD1 is (i) necessary for maternal transmission and/or (ii) its presence could interfere with male transmission. Further evidence in support of this hypothesis is that, except for the singular sequence mentioned above, all truncated F-ORF-VD1 sequences have been found in recombinant CR sequences and are consistently absent from females (i.e., these haplotypes occur only in sperm and could represent recently masculinized CR sequences). The presence of a truncated F-ORF-VD1 in recently masculinized genomes thus suggests that, to enable paternal transmission, a standard F genome must gain a new M functionality (i.e., gain M-type CR sequences) as well as lose an F functionality (i.e., the disruption of F-ORF-VD1). One hypothesis is that the disruption of F-ORF-VD1 occurs first and subsequently this disruption would facilitate F/M recombination, which has been proposed as the first step in the masculinization process [48]. An alternative hypothesis would be that the deletions in the F-ORF-VD1 regions of RM genomes are a consequence of masculinization. If the F-ORF-VD1 is required only for F mt function, its presence in a paternally-transmitted M genome would allow its degeneration. However, we still do not know enough about the recombination process and developmental genetics of DUI to speculate on where and when recombination is likely occurring, though it might happen during spermatogenesis when the five so-called ''mega-mitochondria'' form in the midpiece of a spermatozoon by fusion of several smaller mitochondria [37,95]. Because of the apparently dynamic nature of DUI in marine mussels, the ''Standard Male'' genomes in all Mytilus examined to date are likely the product of previous role reversal events. To reject the hypothesized primacy of F-ORF-VD1 disruption in the masculinization process, one would need to look for paternally transmitted mt genomes containing a complete F-ORF-VD1. The data available to date, however, show that F-ORF-VD1 is eventually lost in ''Standard Male'' genomes, reinforcing the hypothesis that this gene has an F genome-specific function. Although speculative, we propose that F-ORF-VD1 has been maintained in mytilid mitochondrial CRs to participate in the regulation of mt transmission and/or the regulation of F genome replication and transcription. This hypothesis is consistent with the recent suggestions that the RM-type genome sequenced for Mytilus trossulus, which was inferred to be a recently masculinized genome because of its extraction from a male gonad [30], would be in fact the F genome of M. trossulus [40,83]. This particular genome indeed contains a complete F-ORF-VD1, which is consistent with the hypothesis of maternal transmission. However, further data collection and analyses will be essential to clarify the functional role of this putative F-specific protein and to elucidate the mechanisms of mt genome-specific mtDNA transmission in bivalve species with DUI. Conclusion A fundamental question regarding doubly uniparental inheritance of mtDNA in bivalves is whether there are F-and/or Mspecific mtDNA sequences that control the mode of inheritance of Figure 5. Schematics of the structure of a typical F-type (above), recently-masculinized (center, according to Venetis et al. [47]) and typical M-type (below) control regions. The F-ORF-VD1 is identified in the F-type control region. The amino acids that constitute the putative transmembrane helix are indicated in boldface type and bigger characters. The stretch of ,60 residues showing the greatest similarity among the species is underlined. Dashes (-) denote the missing amino acid residues in the truncated F-ORF-VD1. CD, conserved domain; VD1, variable domain 1; VD2, variable domain 2 [39,40]. The ''standard'' F-type CR of M. trossulus, which is a F/M recombinant CR, is not presented. doi:10.1371/journal.pone.0019365.g005 the mitochondrial genomes. Our results demonstrate that there is a systematic difference between maternally and paternally transmitted mytilid genomes: a fourteenth mtDNA-encoded protein, i.e., F-ORF-VD1, is likely present in the former but absent in the latter. Interestingly, this putative additional protein has been found in the first variable domain VD1 of the mitochondrial control region, which is the portion of the CR that was previously suspected to contain the elements responsible for the differing modes of mt transmission in DUI-containing bivalves [39,40]. We present multiple lines of evidence suggesting that a functional protein is coded for by F-ORF-VD1: (i) the gene region has been maintained in the Mytilus lineage (subfamily Mytilinae) for at least 13 million years and our results suggest that a mytiline F-ORF homologue is present in Musculista senhousia (subfamily Crenellinae), (ii) the gene region has been classified as coding by testcode and Glimmer analyses, (iii) the gene region is actively transcribed in Mytilus, (iv) the putative protein's secondary structure has been conserved, (v) the putative protein's amino acid composition are relatively similar and (vi) the gene region's Ka/Ks ratios indicate relaxed purifying selection, which would not be expected for a non-protein coding sequence. Although it is admittedly speculative, we propose that F-ORF-VD1 is essential for the maternal transmission of the F mitochondrial genome in mytilid mussels. Despite the fact that the function(s) of the F-ORF-VD1 protein remains to be determined, our findings suggest that the functional repertoire of animal mitochondrial genomes is greater than previously thought and that novel mitochondrial ORFs, with key biological functions, await discovery in other animal groups. Materials and Methods Complete mitochondrial genome sequences used in this study are listed in Table 1. Complete F-and M-type CR sequences of M. californianus (AF090831 [55]; AY515226-27 and EU826123-24 [40]) and the F-type CR sequence of M. coruscus (AF315574; Barna and Showman unpublished) have also been used. Because the complete M. trossulus F and M mt genomes sequenced by Zbawicka et al. [96] are introgressed mtDNAs from M. edulis, we used the genomes more recently sequenced by Zbawicka et al. [97] as the ''ancestral M. trossulus'' F and M mtDNAs. Examination of ORFs was performed with ORF Finder (http://www.ncbi.nlm.nih.gov/projects/gorf/) using the invertebrate mitochondrial genetic code. Sequence similarity searches were performed in GenBank using BLASTX and PSI-BLAST [87] against the following databases on September 2010 (GenBank release 179.0): (i) non-redundant protein sequences, (ii) SWISSPROT (SWISSPROT release 2010_09), (iii) protein data bank and (iv) environmental samples. We also performed sequence similarity searches using BLASTN against expressed sequence tags (EST others) [98]. T-COFFEE version 8.93 [99] was used to align amino acid sequences and aa alignments were used as a template to align the corresponding codons. Graphical presentation of conserved positions in the alignment was done using Jalview [100]. The coding potential of ORFs was examined using Fickett's testcode algorithm [52] and Glimmer 3 [53]. Transmembrane helices as well as other ORF features of were characterized using HMMTOP [101], TMpred [102] and PredictProtein [103]. Hydropathy profiles were calculated using the method of Kyte and Doolittle [104]. Protein structure and function predictions were made using I-TASSER, a state-of the-art hierarchical protein structure modeling approach that is based on the secondarystructure enhanced profile-profile threading alignment [105,106]. Amino acid composition for chemically equivalent amino acids was obtained following Taylor [107]: acidic amino acids (D and E); aromatic (H, F, W and Y); basic (R, H, and K); charged (R, D, E, H and K); hydrophilic (D, E, K, N, Q and R); hydrophobic (A, C, F, I, L, M, V, W and Y); neutral (G, Q, H, S and T); non-polar (A, C, G, I, L, M, F, P, V, W and Y); and polar (R, N, D, E, Q, H, K, S and T). MEGA 4.0 [108] was used to estimate nucleotide and amino acid divergences among putative ORFs. The number of synonymous substitutions per synonymous site (Ks) and the number of nonsynonymous substitutions per nonsynonymous site (Ka) for the entire F-orf-vd1 sequences within and between Mytilus spp. were also calculated using MEGA 4.0. Site-specific selection, i.e. the estimation of Ka/Ks ratios at each codon site, was studied with the SELECTON server 2.4 (http://selecton.tau.ac.il/index. html) using a Bayesian inference approach [81]. Specifically, the analysis was performed by means of a comparison between a null model assuming no positive selection (M8a; [109]) and a model that allows positive selection (MEC, which treats amino-acid replacements differently by allowing a position with radical replacements to obtain higher Ka value than a position with more moderate replacement; [110]). As data sources we used the codon-aligned partial cox1 and complete F-orf-vd1 Mytilus spp. sequences (M. senhousia was excluded due to alignment issues) and the inferred ML phylogenetic tree. For our models (MEC vs. M8a), likelihood was tested by Akaike Information Criterion (AICc) score comparison [111]. The MEC model is considered the more justified if its AICc score is lower than the score of the alternative model. For each position, a confidence interval defined by the 5th and 95th percentiles of the posterior distributions inferred for the position was estimated. For positions with an inferred Ka/Ks.1, the inference of positive selection is considered reliable when the lower bound of the confidence interval is larger than 1.	2014-10-01T00:00:00.000Z	2011-04-27T00:00:00.000	{ "year": 2011, "sha1": "2d5b8bf62be7996d815edbfe955c4f244cbca59b", "oa_license": "CCBY", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0019365&type=printable", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "2d5b8bf62be7996d815edbfe955c4f244cbca59b", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
10299218	pes2o/s2orc	v3-fos-license	A Stochastic-Variational Model for Soft Mumford-Shah Segmentation In contemporary image and vision analysis, stochastic approaches demonstrate great flexibility in representing and modeling complex phenomena, while variational-PDE methods gain enormous computational advantages over Monte Carlo or other stochastic algorithms. In combination, the two can lead to much more powerful novel models and efficient algorithms. In the current work, we propose a stochastic-variational model for soft (or fuzzy) Mumford-Shah segmentation of mixture image patterns. Unlike the classical hard Mumford-Shah segmentation, the new model allows each pixel to belong to each image pattern with some probability. Soft segmentation could lead to hard segmentation, and hence is more general. The modeling procedure, mathematical analysis on the existence of optimal solutions, and computational implementation of the new model are explored in detail, and numerical examples of both synthetic and natural images are presented. INTRODUCTION: SOFT VERSUS HARD SEGMENTATION Segmentation is the key step towards high-level vision modeling and analysis, including object characterization, detection, and classification. There have been some recent developments indicating that certain high-level visual tasks such as global scene interpretation might be able to bypass segmentation [1,2]. Nevertheless, segmentation still remains perhaps the most important and inspiring task to date in lowor middle-level vision analysis and image processing. The segmentation problem can be formulated as follows. Given an image I ∈ L 2 (Ω) on a 2-dimensional (2D) domain Ω (assumed to be bounded, smooth, and open), one seeks out a closed "edge set" Γ, and all the connected components Ω 1 , . . . , Ω K of Ω \ Γ, such that by certain suitable visual measure (e.g., textural or photometric), the image I is discontinuous along Γ while smooth or homogeneous on each segment Ω i . Each image patch I i = I\| Ωi is also called a pattern, and Ω i is its support. We will call this most common practice "hard" segmentation. A hard segmentation partitions the image domain Ω along a definitive edge set Γ, and outputs nonoverlapping pattern supports Ω 1 , . . . , Ω K . The present work introduces the notion of "soft" segmentation. Mathematically, a hard segmentation amounts to the partition of the unit using indicator functions: 1 Ωi (x), a.e. (in Lebesgue) x = x 1 , x 2 ∈ Ω. (1) A soft segmentation seeks out instead a softer partition of the unit: where p i 's are continuous or smoother functions. Formally, each p i could be considered as the mollified version of 1 Ωi (x). In the stochastic literature of image analysis and modeling, the above notion of soft segmentation is closely connected to mixture image models (e.g., [3]). Suppose a given image I is composed of K unknown patterns: where ω denotes the pattern label variable. At each pixel x ∈ Ω, ω(x) ∈ {1, . . . , K} becomes a random variable. Then the 2 International Journal of Biomedical Imaging Figure 1: Natural images often do not have clear-cut "hard" boundaries between different patterns. Along the arrow, for example, one only observes that the sand pattern gradually becomes a grass pattern. Such a "soft" view is the stochastic view on the segmentation problem. p i 's in (2) carry the natural stochastic interpretation: For this reason, each p i will be called the ownership of pattern i, following Jepson and Black [3]. (Some authors also prefer to call it the membership [4].) Instead of the repulsive ownership in a hard segmentation, a soft one allows each pattern to "own" a pixel with some likelihood. Soft segmentation is more general since it can lead to natural hard segmentation under the maximum-likelihood (ML) principle. Given a soft segmentation {p i (x) : i = 1 : K}, one can define for each pixel x ∈ Ω its unique owner ω * (x) by ω * (x) = arg max ω∈1:K p ω (x), (5) and if the maxima are nonunique, accept the largest index from the arg max pool. The segments are then defined by which leads to a natural hard segmentation. Formula (5) and (6) are called the hardening formulae. Soft segmentation has been motivated by practical analysis of natural images. Patterns in natural scenes often do not have clear-cut boundaries. In Figure 1, for example, there does not seem to exist a "hard" boundary between the grass and sand areas. If one draws an oriented line as shown in the figure, it makes more sense to state that along the arrow, the pattern transits from being "more" sand-like to being "more" grass-like. Such consideration favors the following stochastic view that along the arrow, the ownership Prob ω(x) = grass increases, while Prob ω(x) = sand decreases. In the present work, we propose a new stochastic-variational soft segmentation model for the following celebrated Mumford-Shah model [5,6]: where H 1 stands for the 1D Hausdorff measure [7], which is simply the length when Γ is regular enough. For notational conciseness, the default area-element symbol dx = dx 1 dx 2 will be omitted in most integral formulae. As stated in the abstract, the stochastic softness induces more flexibility and universality in modeling, while the variational-PDE approach facilitates rigorous mathematical analysis as well as more efficient computational implementations compared with purely stochastic approaches including, for example, the Monte Carlo method or Gibbs' sampling [8][9][10][11]. The paper has been organized as follows. Section 2 builds up the soft Mumford-Shah (SMS) model under the Bayesian rationale and the MAP estimator [12,13], which are the formal stochastic foundations of the present model. In Section 3, the prior energy on the ownerships p i 's is developed based on the celebrated work of Modica and Mortola [14] on phase-field modeling and Γ-convergence approximation in material sciences and phase transitions. In Section 4, we analyze the main mathematical properties of the proposed SMS model, including the admissible space, hidden symmetry and symmetry breaking via weak supervision, and the existence theorems. In Section 5, we then derive the system of Euler-Lagrange equations of the SMS model for which the role of the probability simplex constraint is discussed in detail. Section 5 also introduces the alternating-minimization algorithm to compute the Euler-Lagrange equations. Finally, the numerical performance of the SMS model is demonstrated in Section 6 via both synthetic and natural test images that are sufficiently representative and generic. Throughout the manuscript, the notation F[X, Y \| Z] in the deterministic setting always denotes a quantity (often a functional, or an energy) F that depends on X, Y , and Z but with Z given or fixed. Similarly, that is often unimportant as far as the optimization on X (given Y and Z) is concerned. These notations therefore have been inspired by conditional probabilities in the stochastic setting (formally under the Gibbs' correspondence: Bayesian rationale Segmentation can be done in some feature spaces such as gradient-like highpass features or Gabor features (e.g., [11,15,16]). The Mumford-Shah model easily extends to such general features (e.g., [15]), even though it was originally Jianhong (Jackie) Shen 3 formulated only for intensity fields. For maximal clarity in exposing the core ideas of the current work, we will also focus only on the latter, while leaving as canonical exercises to adapt the new model for any given feature distribution. Let K be the total number of intended patterns. As in [10,11], K could also be treated as an unknown to be optimally estimated, which however does not add much to the most significant contribution (i.e., the modeling and computation of the "softening" procedure) of the present work. Given an image input I = I(x) on a bounded, regular, and open domain Ω, the primary goal of soft segmentation is to compute the ownerships Define P(x) = (p 1 (x), p 2 (x), . . . , p K (x)), and where the ( e i \| i = 1 : K) denotes the canonical Cartesian basis of R K . Δ K−1 is often called the canonical (K − 1)-simplex, or the probability simplex in R K . Then meaning that the total ownerships always add up to 100% at any pixel x ∈ Ω. Associated with each pattern label, ω = i is a smooth function u i (x) ∈ H 1 (Ω), similar to the original Mumford-Shah model. Here the Sobolev space H 1 (Ω) is defined by [17] Define U(x) = (u 1 (x), u 2 (x), . . . , u K (x)). Then the goal of soft segmentation is to estimate the optimal vectorial pair of ownerships and patterns given an image I: Prob P, U \| I . By the Bayesian formula [12,13], the posterior given I is expressible via Prob P, U \| I = Prob I \| P, U Prob(P) Prob(U) Prob(I) , (14) assuming that the mixture patterns U and the mixture rules P are independent (as two vectorial random fields). We will call the first term a "mixture generation" model, since it reveals how the image data should look like given the information of the patterns and their ownerships. By taking logarithmic likelihood E[·] = − log Prob(·), or the formally Gibbs' energy in statistical mechanics [18,19], one attains the soft segmentation model in its "energy" form: Assuming that all the pattern channels are independent of each other, one has That is, we assume that U as a random vector field has independent scalar components. It has been motivated by the facts that 2D images are the optical projections of 3D scenes and that different objects in 3D are independently positioned in different ranges or depths. For Sobolev-regular patterns, that is, functions whose gradients are square integrable, one may impose the homogeneous Sobolev energies: for some scalar weight α that models the visual sensitivity to intensity roughness. Unlike the original Mumford-Shah model, the energy for each channel has been defined on the entire image domain Ω instead of on each "hard-cut" patch Ω i . Thus the energy form (17) must carry out extrapolation for practical applications. Long-range extrapolations are, however, often unimportant after being weighed down by their negligible ownerships p i 's. Gaussian mixture with smooth mean fields In this section we discuss the mixture generation model Assume that the patterns are all Gaussian with mean fields u 1 , u 2 , . . . , u K . For simplicity, also assume that they share the same variance σ 2 (which readily generalizes to the more general case with variations). Then at any given pixel x ∈ Ω, Define the Gaussian probability density function (pdf) The pdf of the mixture image I at any pixel x is given by Thus ideally the "energy" for the mixture generation model should be given by for some μ > 0, (21) provided that given two fields P and U on Ω, for any two disjoint and finite sets of pixels X and Y , Here (We also must emphasize that the above derivation should be considered as motivational rather than rigorous, due to the continuum setting.) In the current work, we will adopt a reduced form of the complex formula (21), which is simpler and easier to manage both in theory and for computation. Assume that each soft where the additive constant only depends on σ and K. This suggests the following convenient energy form for the mixture generation model: which amounts to a weighted least-square energy [20]. The weight λ reflects visual sensitivity to synthesis errors. In combination of (15), (17), and (24), the new soft segmentation model takes the form of minimizing Notice that here the ownership distribution P "softens" the "hard" segmentation boundary Γ in the original Mumford-Shah model (8). To complete the modeling process, it suffices to properly define the prior or regularity energy E[P], which is the main task of the next section. MODICA-MORTOLA'S PHASE-FIELD MODEL FOR OWNERSHIP ENERGY To generalize but not to deviate too far from classical hard segmentation, it is natural to impose the following two constraints: (a) each pattern ownership p i (x) has almost only two phases: on (corresponding to p i = 1) and off (to p i = 0), and the transition band in between is narrow; (b) the soft boundaries, or equivalently the transition bands, are regular, instead of being zigzag. In combination, one imposes the following Modica-Mortola type of energy with a double-well potential [14]: Here ε 1 controls the transition bandwidth. Since ε 1, the second term necessarily demands p i 0 or 1 to lower the energy, which well resonates with the expectation in (a). The first term, weighted by the small parameter ε, amounts to a regularity condition on each p i , which meets the requirement in (b). Energies in the form of (26) are very common in material sciences, including the theories of liquid crystals and phase transitions [21,22]. Mathematically, they have been well studied in the framework of Γ-convergence [23], which we now give a brief introduction in the present context. We also refer the reader to the works of Ambrosio and Tortorelli [24,25] on the Γ-convergence approximation to the classical Mumford-Shah segmentation model. Recall that for any q(x) ∈ L 1 (Ω), its total variation as a Radon measure is defined by [7,26,27] where B 2 stands for the unit disk centered at the origin in R 2 . (The TV measure was first introduced into image processing by Rudin et al. [28].) Define that for any q ∈ L 1 (Ω), As a result, a finite energy E 0 [q] necessarily implies that q has two phases only, and to be a subspace of L 1 (Ω) (as a metric space). Then Modica and Mortola's well-known results in [14] readily lead to the following theorem. That is, Jianhong (Jackie) Shen We refer the reader to Modica and Mortola [14] for a proof (with some necessary modification). Here we only point out that the "tight" sequence (q * ε \| ε) in (ii) can be constructed using a smooth sigmoid transition across the hard boundary of a given two-phase function q. Recall as in the theory of neural networks [29] that a sigmoid transition between 0 and 1 is achieved by The scaling parameter ε participates in the transition by the form of σ(t/(3ε)). In particular, ε indeed corresponds to the width of the transition band when t is a distance function. This theorem reveals the close connection of the particular choice of E ε [p i ] in (26) with the original Mumford-Shah model. Proposition 1. Suppose that p ε 's "optimally" (i.e., by the above sigmoidal transition) converge to a given 2-phase pattern Similar results have appeared in the earlier influential works of Ambrosio and Tortorelli [24,25] on the Γconvergence approximation to the Mumford-Shah model. The technique has also been extensively applied in image computation and modeling [30][31][32][33][34][35] to overcome the difficulty in representing and computing the free boundary Γ. To summarize this section, we propose the following energy model for the ownership distribution P(x) = (p 1 (x), p 2 (x), . . . , p K (x)): One, however, must realize that different ownerships are not decoupled by this energy though it has appeared so. The energy E ε [P] must be coupled with the constraint of the probability simplex: In particular, for small ε, although (35) implies that each ownership p i tends to polarize to 0 or 1 independently, they have to cooperate with each other under the above simplex constraint to optimally share the ownerships. The model and admission space Combining the preceding two sections, we have developed the complete formula for soft Mumford-Shah segmentation with K patterns, that is, to minimize with the constraint that that is, p i ≥ 0, i = 1 : K, and K i=1 p i = 1. As discussed previously, it is this simplex constraint that induces coupling among different channels into the seemingly decoupled model (37). Besides the simplex constraint, the last term in the energy (37) requires p i ∈ H 1 (Ω) for i = 1 : K. Similarly, the second term requires each pattern u i ∈ H 1 (Ω). Then with the assumption that "the given image I ∈ L 2 (Ω)," E[P, U \| I] is well defined and finite for any admissible patterns U and pattern ownership distribution P: Breaking the hidden symmetry via weak supervision Let S K denote the permutation group of {1, . . . , K}. Each permutation σ ∈ S K is a 1-to-1 map: Theorem 2 (hidden symmetry of SMS). For any σ ∈ S K , In particular, suppose that P * , U * = arg min (P,U)∈admK is an optimal pair. Then for any σ ∈ S K , (P * σ , U * σ ) is a minimizer as well. The proof is straightforward and thus omitted. Such symmetry not only worsens the nonuniqueness of the minima to the nonconvex energy functional in (37), but also potentially jitters intermediate solutions in iterative computational schemes (i.e., hysterical transitions in the admissible space). To break the permutation symmetry, we turn to a weak supervision scheme in which a user specifies K distinct domain patches: and imposes the symmetry-breaking conditions: where δ i j denotes Kronecker's delta. That is, a user requires each given patch Q i to be a "pure" pattern exclusively labelled by i. Computationally, this weak supervision process can be automated based on multiscale patch statistics as in the contemporary works on scene recognition [1,2], or more generally, the learning theory [36,37]. Existence theorems for nonsupervision and supervision In this section, we briefly state the existence theorems for the soft Mumford-Shah segmentation model (37) without or with the supervision (46). The detailed proof has been moved to the appendix, under the suggestion of one of our referees. Skipping this section will cause no serious problem in comprehending or implementing the models. Mathematically, the existence issue has special appeal to model developers, especially for models that are highly nonconvex. Nonconvex variational models normally fail to guarantee the uniqueness of optimal solutions, and existence is hence often the best one can attempt to establish theoretically [38]. Notice that most interesting variational models in contemporary image and visual analysis are nonconvex, which include, for example, the original Mumford-Shah model [6], various image restoration models (e.g., deblurring and dejittering) [7,39], as well as most optical-flow models [38]. The theoretical proof in the appendix, however, does reveal an important behavior of the model (37) which carries practical implications. If certain channel i becomes dumb in the limiting process of the proof (i.e., the limit p * i ≡ 0 for a minimizing sequence), it has often been introduced unnecessarily in the first place, and the associated optimal pattern u * i could be any featureless constant image. The related issue of determining an optimal class number K (i.e., without containing dumb channels nor missing visually important channels) is also intrinsically driven by the complexity theory of natural images, in particular, the multiscale complexity [40]. Theoretically, K could be any integer, ranging from zero to the infinity, as one zooms into the details of a continuum image from the atomic scale to the ordinary observational scales of the naked eyes. Thus ideally, K itself could be introduced as a random variable taking 0, 1, 2, . . ., and becomes part of the model itself. This idea has already been explored in purely stochastic settings, for example, see Tu and Zhu [10]. For the supervised scenario motivated earlier, the following existence theorem still holds. Theorem 4. Suppose that I ∈ L 2 (Ω). Then an optimal pattern-ownership pair must exist to the soft Mumford-Shah segmentation model (37) with supervision (46), assuming that each patch Q i has a positive Lebesgue measure \|Q i \| > 0. The proof is almost identical to the unsupervised case in the appendix, and simplifies substantially by noticing that no channel could become dumb due to supervision. Furthermore, the functions ρ i 's in the previous proof can be directly set to be without the necessity of turning to Lemma 2. Mixture of homogeneous Gaussians When each pattern i is a homogeneous Gaussian N(m i , σ) with a distinct mean value m i , one has Theorem 5. Suppose that I ∈ L 2 (Ω). Then a minimizer pair (P * , m * ) to E[P, m \| I] exists for both the unsupervised and supervised cases. The proof can be derived readily from the previous general cases and is hence left out. When K = 2, a similar model was proposed earlier by Shen [33] under the symmetrization transform: The model (49) Euler-Lagrange equations on (K − 1)-simplex To minimize the energy for the soft Mumford-Shah segmentation one resorts to its gradient-descent flow or Euler-Lagrange equations. In this section, we discuss these equations and their practical computational schemes. The first-order partial variation on U given P leads to, for i = 1 : K, where n stands for the outer normal vector field along ∂Ω. Thus the Euler-Lagrange equations on the patterns are all in the form of linear Poisson equations with variable coefficient fields: with Neumann adiabatic boundary conditions, where the source terms are f i (x) = λp i (x)I(x). The first-order variation on the ownerships P is carried out on the probability (K −1)-simplex Δ K−1 , which is a compact manifold (with border) of codimension 1 embedded in R K . Chan and Shen [42] developed a general framework for modeling and computing image features that "live" on general manifolds, and especially those that are embedded in R K . We will follow the approach there. Without the simplex constraint on the ownerships, for any given U, the first-order variation of the soft energy E under P → P + δP is given by where H 1 is the 1D Hausdorff measure along ∂Ω, and Define V = (V 1 , . . . , V K ) and v = (v 1 , . . . , v K ). Then which holds for any free variation of P in R K , or one writes in the free-gradient form In reality, P ∈ Δ K−1 . Let T P Δ K−1 denote the tangent space of Δ K−1 at any single point P ∈ Δ K , and the orthogonal projection onto the tangent space in R K . Since the normal direction of the tangent plane is given by K, the projection operator is explicitly given by, for any w ∈ T P R K , The constrained gradient of E on Δ K−1 is therefore given by In particular, the system of Euler-Lagrange equations on P given U is given by for i = 1 : K. The coupling among different channels is evident from these two formulae. Proof. This is obtained by direct computation: at any z ∈ ∂Ω, As a result, the boundary conditions in (62) simplify to the ordinary Neumann conditions ∂p i /∂n = 0, i = 1 : K. Combining all the above derivations, we have established the following theorem. (46), the ownerships must satisfy the interpolation conditions: Theorem 6 (Euler-Lagrange equations). The system of Euler-Lagrange equations of E[P, U \| I] is given by or equivalently, the equations on p i 's in (64) hold on Ω \ ( K i=1 Q i ) with Neumann conditions along ∂Ω, and Dirichlet Similarly, one has the following result for the piecewise constant SMS model (49), which carries much lower complexity compared with the full SMS model. with Neumann conditions for all the ownerships p i 's along ∂Ω. Computation of the Euler-Lagrange equations Computationally, as well practiced in multivariate optimization problems, (64) and (66) can be solved via the algorithm of alternating minimization (AM) [30,39]. The AM algorithm is closely connected to the celebrated expectationmaximization (EM) algorithm in statistical estimation problems with hidden variables [3,43]. In the current context, the ownership distributions p i 's could be treated as the hidden variables. Like EM, the AM algorithm is progressive. Given the current (t = n) best estimation of the patterns U n = (u n i \| i = 1 : K), by solving or equivalently, with Neumann boundary conditions, one obtains the current best estimation of the ownerships P n = (p n i \| i = 1 : K). Subsequently, based on P n , by solving or equivalently, with Neumann boundary conditions, one completes a single round of pattern updating U n → U n+1 . The same procedure applies to the piecewise constant SMS equations in (66). Since the system (70) is linear and decoupled, the main computational complexity resides in the integration of (68), which is coupled and nonlinear due to the simplex constraint and the double-well potential in the energy. Define e i (x) = (u i (x) − I(x)) 2 and e = (e i \| i = 1 : K). In order to solve given e and V = V(P, U) = V(P, e) (see (55)), first notice that since K i=1 p i = 1 and Δ( K i=1 p i ) = 0. We also split the double-potential force in (71) by In combination, the nonlinear equation (71) can then be solved iteratively: by the following linearization procedure: with Neumann adiabatic boundary conditions for all the channels i = 1 : K. This system of linear Poisson equations can be conveniently integrated using any elliptic solvers. The detailed numerical analysis on the convergence behavior of the entire algorithm above, however, is still an open problem and well deserves some systematic investigation. Figure 3: Synthetic image of a T-junction: hard segmentation from the SMS model via "hardening" formulae (5) and (6). The 120degree regularization behavior at the junction point is also well known in the classical Mumford-Shah model [6]. COMPUTATIONAL EXAMPLES In this section, we present the computational results of the proposed soft Mumford-Shah model. Notice that the extension of the above SMS models to color images is straightforward by having the gray values u i 's replaced by RGB vectors. (We, however, must remind the reader that perceptually RGB may not be the most ideal representation of colors compared with other nonlinear approaches, e.g., brightnesschromaticity [42] and HSV [44].) Figures 3 and 4 illustrate the performance of the SMS model on two synthetic images with multiple phases. Figure 3 shows a typical T-junction and Figure 4 shows a 3phase image with a narrow bottleneck. Plotted in the figures are the hard segments obtained from the SMS model via the hardening formulae (5) and (6). Plotted in Figures 5 and 6 are the hardened segments of two MRI brain images computed by the soft Mumford-Shah segmentation model via formulae (5) and (6). For this application, a user specifies three small patches (three rectangles in both examples) Q 1 , Q 2 , and Q 3 , and the SMS model proceeds with the extra interpolation conditions in (46) for the ownerships. Notice in the second example that the detailed branching of the complex boundary is well resolved by the model. In Figure 7, another example of a natural image is segmented via the SMS model and the "hardening" formulae (5) and (6). A user supervises with three patches Q 1 , Q 2 , and Q 3 , and designates the two on the body to a pattern ownership Figure 4: Synthetic image of a narrow bottleneck: hard segmentation from the SMS model via "hardening" formulae (5) and (6). The thickening regularization at the bottleneck junction can be explained similarly by the classical Mumford-Shah model for which minimum-surface or "soap-foam" behavior arises due to the surface tension energy. Also, see the recent work by Kohn and Slastikov [45] for the singularity analysis of a similar problem arising from micromagnetism. p body and the third (from the ocean) to p ocean . If the three are treated as distinct patterns, the SMS model still works, but one needs an extra step of high-level vision processing (e.g., based on Grenander's graph models [46]) to group the skin-tone and the purple-shirt patterns in order to capture the entire body faithfully. Finally, plotted in Figures 8 and 9 are the ownerships from the SMS model based on the 3-phase and 4-phase supervisions separately in Figure 2. The stochastic nature of the outcomes (i.e., the softly transiting ownerships p i 's instead of hard segmentation) is closer to the way a human subject may perceive such a natural scene. In particular, the SMS model seems to be consistent with the most recent theory that hard pattern segments may not be absolutely necessary for natural scene recognition [1,2]. CONCLUSION In this work, we have improved the celebrated Mumford-Shah segmentation to allow stochastic fuzziness of individual patterns. The proposed model outputs the ownership (or membership) probability distributions for all the patterns, from which the classical hard segmentation can be obtained based on stochastic decision rules such as the principle of maximum likelihood or the Bayesian classifier. Figure 6: A brain image with low noise: hard segmentation from the SMS model via "hardening" formulae (5) and (6). Notice how the detailed branching of the gray matter has been successfully resolved by the model. The key component of the new model is an ensemble of regularized double-well potentials inspired by the literature of material sciences and variational calculus. The model is nonconvex and the existence of optimal soft segmentation has been proven. A preliminary algorithm has been proposed and implemented, but without convergence analysis. Several generic numerical examples have demonstrated the flexibility and performance of the new model. Our future work will mainly focus on (1) automating the weak supervision process based on statistical patch analysis, as inspired by the recent work of Li and Perona [1], and (2) developing a comprehensive framework for the effective computation of such a nonconvex and multivariate variational model (with Alan Yuille). PROOF OF THE EXISTENCE THEOREM 3 We will need the following lemma for the proof. Proof. Denote the Lebesgue measure of a measurable set W by \|W \|. Since p * ≥ 0 and Ω p * > 0, there must exist some c > 0, such that V = x ∈ Ω \| p * > 2c has a finite but positive measure. In particular, there exists some N, such that for any n > N, p n > c on W. Define Original noisy image (5) and (6), based on a 2-phase supervision. Denote the two rectangles on the body by Q 1 and Q 2 , and the third by Q 3 . Supervision provides the ownership interpolation condition: p body = 1 on Q 1 ∪ Q 2 and 0 on Q 3 , while p ocean = 1 on Q 3 and 0 on Q 1 ∪ Q 2 . Patch selection can also be automated based on multiscale patch statistics (e.g., see Li and Perona [1]). Then Ω ρ = 1, and for any n > N, Thus by the Schwarz inequality (or E[X] 2 ≤ E[X 2 ] in probability theory), The lemma holds if one defines We are ready to prove Theorem 3. Proof. Take the special pattern distribution: Thus the infimum of the energy must be finite. Let (P n , U n \| n) ⊆ adm K (see (40)) be a minimizing sequence for the soft Mumford-Shah energy (37). That is, E[P n , U n \| I] converges to inf P,U E[P, U \| I]. Due to the third term in the energy and the simplex constraint, for each channel i, (p n i \| n) must be bounded in H 1 (Ω). By the L 2 -weak compactness, there must exist some P * ∈ L 2 (Ω, R K ), and a subsequence of (P n \| n), which after relabelling will still be denoted by (P n \| n) for convenience, such that Then by the L 2 lower semicontinuity of Sobolev measures, Furthermore, with possibly another round of subsequence refinement, one can assume that P n (x) −→ P * (x), a.e. x ∈ Ω, n → ∞. (A.10) Since the probability simplex Δ k−1 is closed and P n (x) ∈ Δ K−1 , one concludes that And by Fatou's lemma [47,48], one has (A.12) (In fact, the equality holds by Lebesgue's dominated convergence [48].) After the above subsequence selection on P n 's, one naturally has an associated subsequence of (U n \| n), which for convenience is still denoted by (U n \| n) after relabelling. For each specific channel i, we then consider two scenarios separately. Suppose that p * i (x) ≡ 0, a.e. x ∈ Ω. We then define for that channel Such a channel is called a "dumb" channel. Otherwise, one must have Ω p * i > 0, and from the first term in (37), Figure 2(a). Plotted here are the three ownership distributions p 1 (x), p 2 (x), and p 3 (x). Due to "under"-supervision, namely the number K of specified patterns is less than that of the visually meaningful ones, the grass pattern has "absorbed" the ocean pattern due to the greenish color they happen to share. Since Ω I 2 p n i ≤ Ω I 2 , by the triangle inequality, for some constant C i independent of n. Then by the generalized Poincaré inequality [48,49] on Ω, w − w, ρ i L 2 ≤ A i ∇w L 2 , (A. 18) where A i = A i (ρ i , Ω) is independent of w ∈ H 1 (Ω), one concludes that for some constant D i . As a result, (u n i \| n) must be bounded in H 1 (Ω). By the L 2 -weak compactness of bounded H 1sequences, there is a subsequence of (u n i \| n), for convenience still denoted by (u n i \| n) after relabelling, such that u n i −→ u * i ∈ L 2 (Ω), n −→ ∞, (A.20) converging in the sense of both L 2 and almost everywhere. Then by the lower semicontinuity, Combining both cases just analyzed above, we have established that and hence (P * , U * ) must be a minimizer. (We must caution our reader that since index relabelling has been performed for a couple of times to simplify notations, this last sequence (P n , U n ) is not the one we have started with originally.) This completes the proof. ACKNOWLEDGMENTS The author is very grateful to Professor Alan Yuille for an enlightening discussion after the current work was first presented. For their generous teaching and continual inspiration, the author is always profoundly indebted to Professors Gil Strang, Tony Chan, Stan Osher, David Mumford, Jean-Michel Morel, and Stu Geman. The author must thank his wonderful former teacher, Professor Dan Kerstan at the Psychology Department of the University of Minnesota, for his first introduction on mixture image models and stochastic visual processing several years ago. The author also thanks the Institute of Mathematics and its Applications (IMA) and the Institute of Pure and Applied Mathematics (IPAM) for their persistent role in supporting this new emerging field. Finally, the author would like to dedicate this paper to his dear friends Yingnian Wu and Song-Chun Zhu for the unique friendship cultivated by the intellectually rich soil of vision and cognitive sciences. The generous help from our referees is also enormous. This work has been partially supported by the NSF (USA) under Grant no. DMS-0202565.	2014-10-01T00:00:00.000Z	2005-10-23T00:00:00.000	{ "year": 2006, "sha1": "644f0dc5a6d6458add684257c6d9d3f53528298b", "oa_license": "CCBY", "oa_url": "http://downloads.hindawi.com/journals/ijbi/2006/092329.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "ee1aaf81ed46dc574874eee33a1b896f73ff1208", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Computer Science", "Mathematics", "Medicine" ] }
7469817	pes2o/s2orc	v3-fos-license	On Non-Point Invertible Transformations of Difference and Differential-Difference Equations Non-point invertible transformations are completely described for difference equations on the quad-graph and for their differential-difference analogues. As an illustration, these transformations are used to construct new examples of integrable equations and autotransformations of the Hietarinta equation. Introduction The present paper is devoted to invertible transformations for both discrete equations of the form u i+1,j+1 = F (u i,j , u i+1,j , u i,j+1 ), and "semi-discrete" chains of the differential equations (u i+1 ) x = F (x, u i , u i+1 , (u i ) x ). Here i and j are integers, x is a continuous variable, u is a function of i, j and i, x for the first and the second equation, respectively. From now on, we shall omit i and j for brevity and, in particular, write the above equations in the form u 1,1 = F (u, u 1,0 , u 0,1 ) (1.1) and (u 1 ) x = F (x, u, u 1 , u x ). (1.2) We assume that F u F u 1,0 F u 1,0 = 0 for equation (1.1) and F ux = 0 for equation (1.2). These conditions allows us to rewrite equation (1.1) in any of the following forms Therefore, all "mixed shifts" u m,n := u i+m,j+n (for both positive and negative non-zero n and m) can be expressed in terms of dynamical variables u k,0 , u 0,l by virtue of equations (1.1), (1.3)-(1.5). (A more detailed explanation of the dynamical variables, the notation u m,n and the recursive procedure of the mixed shift elimination can be found, for example, in [11,10].) Analogously, u (n) m := ∂ n u i+m /∂x n for any non-zero m ∈ Z and n ∈ N can be expressed in terms of x and dynamical variables u l := u i+l , u (k) := ∂ k u i /∂x k by virtue of equations (1.2), (1.6). The notation g [u] means that the function g depends on a finite number of the dynamical variables (and x if we consider equation (1.2)). The considerations in this paper are local (for example, we use the local implicit function theorem to obtain (1.3)-(1.6)) and, for simplicity, all functions are assumed to be locally analytical. In addition to the point transformations v = g(u), some of the equations (1.1) and (1.2) admit non-point transformations v = g [u] which are invertible in the sense of [16]. For example, the differential substitutions v = u x − sin u 2 (1.7) maps solutions of the differential-difference sine-Gordon equation [7,14] (u 1 ) x − sin u 1 = u x + sin u (1. 8) into solutions of the equation which is a semi-discrete version of the complex sine-Gordon equation. Here the sign of the righthand side of equation (1.9) coincides with the sign of the cos u value 1 . Indeed, v 1 = (u x +sin u)/2 follows from equation (1.8) and, together with (1.7), gives us The inverse transformation can be found in [12]: the formula u = π 2 ± (arcsin(v 1 − v) − π 2 ) maps any real solution of equation (1.9) into a solution of equation (1.8). This example belongs to the following class of non-point invertible transformations introduced in [17]. Let functions ϕ(x, y, z), ψ(x, y, z) satisfy the condition ϕ y ψ z − ϕ z ψ y = 0 and equation (1.2) can be written in the form (1.10) Then we rewrite (1.10) in the form of the system express u, u x in terms of v, v 1 from (1.11) and obtain (1.12) The system (1.12) is equivalent to the equation where D x denotes the total derivative with respect to x. The substitution v = ϕ(x, u, u x ) maps solutions of (1.10) into solutions of (1.13) and the transformation u = p(x, v, v 1 ) maps solutions of (1.13) back into solutions of (1.10). It is easy to see that the same scheme can be applied to the pure discrete equations of the form ϕ(u 0,1 , u 1,1 ) = ψ(u, u 1,0 ), (1.14) where ϕ(y, z) and ψ(y, z) are functionally independent. Indeed, expressing u and u 1, we obtain and rewrite (1.16) in the form of the equivalent equation Thus, the transformation v = ϕ(u, u 1,0 ) maps solutions of (1.14) into solutions of (1.17) and the inverse transformation u = p(v, v 0,1 ) maps solutions of (1.17) back into solutions of (1.14). The transformations (1.14)-(1.17) were, in fact, used in [18] without explicit formulation of the above scheme. The invertible transformations allow us to obtain objects associated with integrability of equations (1.13), (1.17) (such as conservation laws and higher symmetries) from the corresponding objects of equations (1.10), (1.14) because we can express shifts and derivatives of u in terms of shifts and derivatives of v. Therefore, the invertible transformations may be useful for constructing new examples of integrable equations of the form (1.1), (1.2). To illustrate this, in Section 4 we construct Darboux integrable equations related via invertible transformations to difference and differential-difference analogues of the Liouville equation. In addition, an example of constructing an equation possessing the higher symmetries is contained at the end of Section 2. In this section we also demonstrate that the scheme (1.14)-(1.17) generates autotransformations of the Hietarinta equation. Invertible transformations of discrete equations We let T i and T j denote the operators of the forward shifts in i and j by virtue of equation (1.1). These operators are defined by the following rules: (2.1) is called invertible if any of the dynamical variables u, u k,0 , u 0,l , k, l ∈ Z, can be expressed as a function of a finite subset of the variables We exclude all mixed variables v r,s , rs = 0, from (2.2) because we consider only the cases when the transformation maps (1.1) into an equation of the form and the mixed variables can be expressed in terms of (2.2) by virtue of this equation. It is easy to see that any shift w = v r,s maps equation (2.3) into equation (2.3) again and the composition of the shift and an invertible transformation v = f [u] is invertible too. This leads to the following 3). Then this transformation is equivalent to either a transformation of the form or a transformation of the form Proof . The transformation is equivalent to that of the form v = h(u, u 1,0 , . . . , u k,0 , u 0,1 , . . . , u 0,l ) (2.6) because we can eliminate "negative" variables u r,0 , u 0,s , r, s < 0 from the transformation by shifts of g. We can express u as if the transformation is invertible. Differentiating equation (2.7) with respect to u k+b,0 , we for any negative a and c, and we obtain P v a,0 = P v 0,c = 0 by differentiating equation (2.7) with respect to u a,0 and u 0,c . Therefore, either The latter equality means that either T −1 i (h) =h(u, u 1,0 , . . . , u k−1,0 , u 0,1 , . . . , u 0,l ) or T −1 j (h) =h(u, u 1,0 , . . . , u k,0 , u 0,1 , . . . , u 0,l−1 ), i.e. any invertible transformation of the form (2.6) with kl = 0 is equivalent to a transformatioñ v =h(u, u 1,0 , . . . , uk ,0 , u 0,1 , . . . , u 0,l ) such thatkl < kl. Applying this conclusion several times, we obtain that (2.6) is equivalent to a transformation w = f (u, u 1,0 , . . . , u m,0 , u 0,1 , . . . , u 0,n ) with mn = 0. Proof . If f u = 0 and s is the smallest integer for which f u s,0 = 0, then the equivalent transfor- ) depends on u. Therefore, we can, without loss of generality, assume that f u = 0. We also can write because the transformation is invertible. Here the notation l = 0, m means that l runs over all integers from 0 to m. Differentiating these equalities with respect to u a,0 , we obtain Let c < 0 and s be the biggest negative integer such that (T s This implies T c j (f ) = g(u, u 1,0 , . . . , u m,0 ) and If c ≥ 0, then equations (2.8) holds too, with g = f andd = d. It is not always easy to see whether equation (1.1) can be represented in the form (1.14). For example, at first glance it seems that the equation does not admit an invertible transformation of the form u = ϕ(v, v 1,0 ). But in reality we can rewrite this equation as and relate it to the equation Therefore, it is useful to reformulate our result in the following form. Returning to equations (2.9), (2.10), we note that equation (2.10) was introduced in [13] in a slightly different form. This equation has also been used in [10] as an example of an equation which is inconsistent around the cube (in the sense of [1]) but possesses the higher symmetries. Therefore, we can obtain symmetries of equation ( Because L F (ξ[u]) = 0 by definition of symmetry, we see that v τ = f u 0,1 T j (ξ[u]) + f u ξ[u] (after rewriting in terms of v and its shifts) is a symmetry of equation (2.3). Applying this, for example, to the three-point symmetries The Hietarinta [6] equation 2 u 1,1 (u + β)(u 0,1 + α) = u 0,1 (u + α)(u 1,0 + β) (2.13) is another interesting example. The invertible transformations v = u 1,0 (u + α) u − α, w = βu 0,1 β + u − u 0,1 map this equation into equation (2.13) again. In addition, the Hietarinta equation is linearizable [15]. We note that the above properties of equation (2.13) are similar to those of the continuous equation which was considered in [16]. Here Invertible transformations of dif ferential-dif ference equations i.e. D x is the total derivative with respect to x by virtue of equations (1.2), (1.6). The inverse (backward) shift operator T −1 is defined in the similar way. Definition 5. We say that a transformation v = f [u] maps equation (1.2) into an equation (1.2) is called invertible if any of the dynamical variables u, u k , k ∈ Z, u (l) , l ∈ N can be expressed as a function of a finite subset of the variables It is easy to see that a transformation of the form (3.2) or (3.3) is non-point only if f depends on more than one of the variables u, u 1 , . . . , u m or on at least one of the variables u (1) , . . . , u (n) , respectively. Theorem 2. Let a non-point invertible transformation of the form (3.3) map equation (1.2) into equation (3.1). Then equation (1.2) can be written in the form where ϕ(x, y, z) and ψ(x, y, z) satisfy the condition ϕ y ψ z − ϕ z ψ y = 0, and the transformation is equivalent to the composition of the invertible transformation w = ϕ(x, u, u x ) and an invertible transformation of the form v = h(x, w, w (1) , w (2) , . . . , w (n−1) ). In particular, any non-point invertible transformation of the form v = f (x, u, u x ) is equivalent to the composition of the transformation w = ϕ(x, u, u x ) and a point transformation v = h(x, w). For brevity, we omit the proofs of the above propositions because they are very similar to the proofs for discrete equations. Examples: the transformations of Liouville equation analogues A special class of integrable equations of the form consists of equations for which there exist both a differential substitution of the form v = X(x, y, u x , u xx , . . . ) and a substitution of the form w = Y (x, y, u y , u yy , . . . ) that map (4.1) into the equations v y = 0 and w x = 0, respectively. Such equations are called Darboux integrable or equations of the Liouville type. They not only are C-integrable (in accordance with the term of [3]) but also possess infinitely many symmetries of arbitrary high order [19,20]. The complete classification of the Darboux integrable equations (4.1) has been performed in [20]. Equations with the analogous properties exist among equations of the form (1.1) and (1.2) too, but the classification of such equations is completed for a special case of equation (1.2) only [5]. Therefore, deriving new examples of discrete and semi-discrete Darboux integrable equations from already known equations may be useful (for example, to check the completeness of a future classification). Discrete equations The first example is the discrete Liouville equation from [9]. According to [2], this equation has the integrals maps solutions of the equation into solutions of (4.2). Applying Corollary 1, we see that the other discrete version [8] of the Liouville equation does not admit a non-point invertible transformation. This equation is mapped into (4.2) via the non-invertible transformation u = v 1,0 v 0,1 and has the integrals Differential-difference equations Let us consider the following analogue of the Liouville equation: This equation has the integrals into solutions of (4.7). The above information and some other details about equation (4.7) can be found in [2]. Equation (4.7) can be written as Applying the scheme (1.10)-(1.13), we obtain v = 1 2 of equation (4.10) from the general solution z = α i + β(x) of (4.9), where α i and β(x) are arbitrary. Equation (4.10) was used in [2] as an example of an equation admitting the integrals and the solution (4.11) was constructed in this article by another method (directly form the equation X[v] = ξ(x)). Moreover, equation (4.7) can be represented in the form (1.13) too. Applying the scheme (1.10)-(1.13) in the reverse order, we get and see that the invertible transformation w = u 1 /u maps (4.7) into the equation (w 1 ) x = w x w 1 w 1 + 1 w + 1 . (4.14) The later equation has the integrals	2014-10-01T00:00:00.000Z	2010-10-02T00:00:00.000	{ "year": 2010, "sha1": "034c9293d2550505dd91d41a3fdc68680fb9b625", "oa_license": "CCBYSA", "oa_url": "http://www.emis.de/journals/SIGMA/2010/092/sigma10-092.pdf", "oa_status": "GOLD", "pdf_src": "Arxiv", "pdf_hash": "930b92dd2a51cf78012d26b9c1f4decbb22c110b", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [ "Mathematics", "Physics" ] }
6830343	pes2o/s2orc	v3-fos-license	Quantifying the Impact of Dependent Evolution among Sites in Phylogenetic Inference Abstract Nearly all commonly used methods of phylogenetic inference assume that characters in an alignment evolve independently of one another. This assumption is attractive for simplicity and computational tractability but is not biologically reasonable for RNAs and proteins that have secondary and tertiary structures. Here, we simulate RNA and protein-coding DNA sequence data under a general model of dependence in order to assess the robustness of traditional methods of phylogenetic inference to violation of the assumption of independence among sites. We find that the accuracy of independence-assuming methods is reduced by the dependence among sites; for proteins this reduction is relatively mild, but for RNA this reduction may be substantial. We introduce the concept of effective sequence length and its utility for considering information content in phylogenetics. One of the fundamental assumptions made by most methods of phylogenetic inference is that characters evolve independently. This is of course not the case in reality, and there has in recent years been an effort to develop models that more accurately reflect the various types of dependence among sites that have been observed in a biological context. One kind of dependence is the correlation of rates of substitution at adjacent sites. Yang (1995) and Felsenstein and Churchill (1996) developed methods that allowed the rate of substitution at a given site to depend on the rates of substitution at neighboring sites. But it is important to note that in these models it is only the overall rate of substitution that is correlated among sites; substitutions under the model remain independent at different sites. We will focus on methods in which both the rate and types of changes observed at one nucleotide position are dependent upon the nucleotide observed at another position in the sequence. An example of this kind of dependence, in which adjacent sites can influence not only the rate but also the types of substitutions that occur, is found in the triplet codon structure in protein-coding DNA. Certain substitutions may be less frequent at one site because a change at that site would alter the amino acid encoded by the three sites taken together. Muse and Gaut (1994) and Goldman and Yang (1994) developed codon-based methods to address these concerns, and Nielsen and Yang (1998) expressed the codon model in the form most commonly used today. Dependence can also arise due to the secondary structure of RNA molecules. Particular attention has been paid to develop methods that address the pairing of nucleotides in RNA stem formations (Schöniger and von Haeseler 1994;Tillier 1994;Tillier and Collins 1995). Dependencies due to secondary structure are often more complicated than those at adjacent sites as the depen-dent positions may be quite far from each other in terms of sequence position. It should be noted that these models, like the codon models, are one-substitution-ata-time models. Codon models for protein-coding DNA and doublet models for RNA share in common a general approach for accounting for dependence: They expand the basic evolutionary unit in the model from the nucleotide to the triplet or to the doublet, respectively. Robinson et al. (2003) took this approach to its logical endpoint using the entire protein-coding DNA sequence as the unit of evolution. They considered dependencies resulting from amino acid interactions as well as those resulting from solvent accessibility, and in doing so they allowed the number of other sites on which a given site was dependent to vary across the sequence. Rodrigue et al. (2005) took a similar approach but using only the amino acid interactions, and Kleinman et al. (2006) showed that the model fit is much better when the solvent accessibility is included (see Anisimova and Kosiol 2009 for a review of several of these models of substitution). Error in phylogenetic estimation due to dependent evolution has been detected in recent data set as well. Castoe et al. (2009) identified 13 mitochondrial proteincoding regions in squamates that they believe to be the result of strong nonneutral convergence. They argue that models of evolution that can account for convergence due to negative selection, such as those which consider the structure of a protein, might be useful for detecting similar cases that may otherwise strongly bias phylogenetic estimates. Here we quantify how robust methods of phylogenetic inference are to violation of the assumption of independence. We use an evolutionary model similar to other sequence-based evolutionary models (Robinson et al. 2003;Rodrigue et al. 2005;Yu and Thorne 2006) constraints based on RNA and protein structures, and we evaluate the performance of traditional phylogenetic methods on these simulated data sets. We find that even small amounts of dependence in the data can lead to significant error in estimation of the true topology, and that this is especially true for RNA. General Strategy We are interested in testing whether or not methods of phylogenetic inference that assume independent evolution at each site are robust to violation of that assumption. We are specifically interested in the ability to recover the correct tree topology rather than in accurately estimating branch lengths or other model parameters. The general strategy is as follows: 1) simulate an alignment under a known tree topology and set of branch lengths, with a known model of dependence; 2) estimate the tree from the simulated alignment using standard methods of phylogenetic inference, all of which assume independence of substitutions at different sites; and 3) assess the accuracy of the methods. The methods we will test are maximum likelihood (ML) using the general time-reversible model of substitution with gamma-distributed rate variation (GTR+Γ ;Tavaré 1986;Yang 1993Yang , 1994, neighbor-joining (Saitou and Nei 1987) using GTR+Γ distances, and parsimony as implemented in PAUP* 4.0b10x (Swofford 1998). We will be less interested in comparing these methods with each other than in examining the effect of dependence in the data on all of these methods. The simplest case in which to study the effect of dependent evolution on phylogenetic inference is with the four-taxon tree and has been well-studied previously (Felsenstein 1978;Huelsenbeck and Hillis 1993). We largely focus on the four-taxon case in order to obtain a thorough understanding for how the tree length, tree topology, and varying levels of dependence affect inference. The tree we consider is shown in Figure 1. Two opposing terminal branches share a common length a, whereas the other two terminal branches and the internal branch share a common length b. The proportion a/b will be of great interest to us; when this quantity is larger the inference problem is increasingly difficult. We will also be interested in the total tree length (V), which allows the tree to be expanded or contracted while preserving the branch length proportions. Evolutionary Model Calculation of the likelihood (or the parsimony score) of an alignment is greatly simplified by the independence assumption. If all sites are independent, then the probability of an alignment is simply the product of the probability of each column in the alignment (or the sum of the parsimony scores). To calculate this probability, the substitution process at a particular site is modeled as a continuous-time Markov chain. The process is gov-FIGURE 1. The four-taxon tree. The ratio of the branch lengths a/b and the total tree length V are the parameters of interest. As a/b becomes large, the inference problem becomes increasingly difficult. erned by a rate matrix Q = {q ij } where q ij is the rate of change from state i to state j. This rate of change depends only on the current state i, and does not depend on what states may have been observed in the past (Markov property). Furthermore, the rate q ij is agnostic to what is happening at every other site in the sequence. When dependence among sites is introduced it will not affect the Markov property, but the rate of change at a given site will depend on the state of the process at other sites. The rate matrix Q can take many forms. For RNAcoding sequences, the matrix Q might be described by anything from the Jukes-Cantor model (Jukes and Cantor 1969) to the GTR model (Tavaré 1986), and does not in principle need to be time-reversible. For protein-coding sequences, Q could be described by various codon-based models (Muse and Gaut 1994;Goldman and Yang 1994) with different rates for synonomous/nonsynonomous sites as well as for transitions/transversions. These codon models typically restrict the possible changes from codon i to only those codons j that involve a single nucleotide substitution, and disallow stop codons. Just as codon-based models expand the unit of evolution from the nucleotide to the codon, the model we consider further expands the unit of evolution from the nucleotide to the entire sequence. Consequently, we will be interested in sequence transition probabilities. More formally, we will consider a continuous-time Markov chain in which the state space is the set of all possible sequences of length N nucleotides. Let x and y be two such sequences. Then for all x and y, the matrix of rates of change from x to y can be defined as where q ij is, as described above, the rate of substitution under an independent-sites model for the position in the sequence that is changing, E(x, y) compares the relative structural fitness of sequences x and y and is described in detail below, and u is a rate-scaling factor to ensure that branch lengths are interpretable in terms of average number of substitutions per site. In the simulations of this paper, we specify the underlying {q ij } as follows. For RNA, we assume the K80 model of substitution (Kimura 1980) with the transition/transversion rate ratio κ = 3. For proteins, we assume a codon substitution model (Nielsen and Yang 1998) with transition/transversion rate ratio κ = 1, nonsynonomous/synonomous substitution rate ratio ω = 1 and equal codon frequencies. These relatively simple substitution models were chosen to better examine the effects introduced by the dependencies due to structure, described below. Energy of a Sequence We will utilize a concept borrowed from structure prediction: a sequence folded into a particular structure will have a free energy associated with it. In structure prediction, the sequence is fixed and the structure of lowest energy is sought; here, we invert this problem by conditioning on the structure being fixed. We assume that all sequences share a common fixed structure that must be maintained to preserve functionality. It is this structure that determines the interactions among sites and their relative positions, and therefore determine their evolutionary interdependencies. We will define the energy of a single sequence x as E(x), but we find it useful to conceive of E(x) not as an actual energy, rather as a measure of how well sequence x corresponds to the given structure, or as a kind of "structural fitness." If x could reasonably fold into the given structure, we expect E(x) to be low, ideally negative. We can then calculate E(y) for any sequence y as well. E(x, y) then takes on the meaning of a comparison of the relative structural fitness of the two sequences. The precise form of E(x, y) can in principle vary, and we will define E(x, y) differently for RNA and for proteins. For RNA, what we call energies are folding free energy changes (ΔG) predicted using the current nearest neighbor model of Turner and co-workers (Mathews et al. 2004). These free energy changes are predicted for a given base pairing structure using the efn2 model (Mathews et al. 1999). This approach utilizes information from both the base pairing and the coaxial stacking of nucleotides, allowing the potential to incorporate more information than a simple doublet model that considers doublets to be independent of each other. For RNA, we then define E(x, y) = e (Ez(x)−Ez(y))z where z is a free parameter determining the degree to which the difference in structural fitness affects the rate of substitution. Note that when z = 0, E(x, y) = 1 for all x and y, reducing the model to the independent-sites model specified by the single-site rate matrix Q. For proteins, we adopt the approach of Robinson et al. (2003) in simplifying the constraints governing the structure into two properties: energies due to pairwise interactions of amino acids and to solubility constraints (denoted E p (x) and E s (x), respectively). To do this, we utilize statistical potentials, which are pseudo-energy values associated with plausibilities of some aspect of the structure estimated from protein sequences of known structure. For pairwise interactions of amino acids, we can from the protein structure determine the relative positions of all amino acids in three-dimensional space, and declare two amino acids to be "in contact" if any of their non-hydrogen atoms are less than 4.5Å apart (Bastolla et al. 2001). Pairs of amino acids whose three-dimensional proximity is due to sequential proximity (within three positions or less) are not considered to be in contact. Following Rodrigue et al. (2005), if our sequence is of length N we can describe a contact map as an N × N matrix C where Two aspects of this formulation should be noted. First, unlike RNA where sites can potentially pair, here a single site can be considered in contact with multiple other sites. Second, these interactions are all weighted equally regardless of actual physical distance, as long as they are sufficiently close. It would be straightforward to alter the latter such that the relative distance is preserved and certain interactions are more influential than others. As described by Rodrigue et al. (2005), we can now define the energy of the sequence x with respect to pairwise potentials as the sum of the pair potentials for all pairs of amino acids in contact: where x l and x m are the amino acids of sequence x at positions l and m, respectively, and B = {b x l ,xm } is the pair potential matrix of Bastolla et al. (2001). To model solubility constraints on protein evolution, we follow Robinson et al. (2003), who used an analysis of a large number of proteins to estimate how frequently a particular amino acid is observed at different degrees of solvent accessibility [see also Jones et al. 1992 andJones 1999]. From the protein structure, we determine the solvent accessibility of a particular amino acid position. The energy with respect to solubility of sequence x, E s (x), is then the sum across all sites of the plausibility of seeing the observed amino acid at that accessibility level, where a k is the degree of solvent accessibility of site k and S(x k , a k ) is the statistical potential for observing amino acid x k at such a degree of solvent accessibility (Robinson et al. 2003). We can now define E(x, y) for proteins in a similar form as for RNA: where p and s are, like z in the case of RNA, parameters that control how much the difference in sequence energies affect the rate of substitution for pairwise potentials and solubility, respectively. Note again that when p = s = 0, E(x, y) = 1 for all x, y, reducing the model to one of independence among sites. Simulation Procedure There are a number of ways to simulate data at a single position under an independent-sites model. Some of these are not applicable for simulating data that are context dependent. We will discuss a few of these methods and their applicability. The first method ( Fig. 2a) begins by drawing the nucleotide at the root node of the tree from the stationary distribution π π π = {π A , π C , π G , π T }. If we specify the rate matrix Q and a branch length t, we can calculate the transition probability matrix P(t) = {p ij (t)} = e Qt . This provides, for all i and j, the probability that after a branch length of t, the descendent node is in state j given that our ancestral node was at state i. Each branch will have its own transition probability matrix because branch lengths may differ. We can then work our way up the tree starting from the root, choosing states at each node until we reach the tips. The usage of matrix exponentiation to calculate transition probabilities is attractive because it considers all the possible paths, or character histories, from i to j in time t. However, the matrix exponentiation becomes intractable when the rate matrix is large. This is the case with the dependent-sites model we have described, where the rate matrix R is 4 N × 4 N , and for any reasonable sequence length N the matrix is quite large indeed. Instead of using a transition probability matrix to consider all the possible paths from state i to j over time t simultaneously, we could instead simulate a single character history (Fig. 2b). One of the properties of the continuous-time Markov chain is that if the process is in state i, the waiting time until we leave state i is an exponentially distributed random variable with rate q ii = − j =i q ij . This means we find our root node state i from the stationary distribution as before, but now draw an exponential random variable with rate −q ii . If this time is less than t, we observe a change from i to some other state j. The particular state j is drawn with probability p ij = q ij /−q ii . This procedure is repeated until the sum of the drawn waiting times exceeds the length of the branch t, at which point the state of the process is the state at the descendent node. This character history simulation is performed iteratively up the tree for all branches until we have our states at the tip nodes. This FIGURE 2. Three methods for simulating data under independence. a) Using matrix exponentiation is intractable for dependent data. b) Simulating a character history can be done with context dependency for an entire sequence, but drawing from the stationary distribution at the root node is still problematic. c) Evolve into stationarity by simulating a very long character history before reaching the root, then continuing up the tree as in (b). method of drawing character histories has the benefit that it can be used under the dependent-sites model we have described. This is done by using the full sequence as the unit of evolution and replacing the site rate matrix Q with the sequence rate matrix R, and then drawing a sequence history along the branch. Both of these methods have assumed that we could draw the state at the root of the tree directly from the stationary distribution. This is not trivial under the dependent-sites model as the state space of all possible sequences is quite large (4 N possible sequences) when compared with independence (four possible nucleotides). However, the intuitive meaning of the process being at stationarity at the root is that the process has been underway for a long time before reaching the root of the tree, and we can simulate this directly (Fig. 2c). Under independence, if we pick any state 64 SYSTEMATIC BIOLOGY VOL. 60 i as an ancestral state and then simulate its evolution along an exceedingly long branch before reaching the root, then the probability that we observe a particular state j at the root is the same as having drawn directly from the stationary distribution. This method can be used for the dependent-sites model described as well and is the method employed for all simulations in this study. We begin with an arbitrary sequence, not necessarily the one that would likely be sampled from the true dependent stationary distribution. We then evolve this sequence along a very long root branch under the model of dependence as described above, allowing the sequence to evolve into the one that would be sampled from the true dependent stationary distribution. The intuition should be clear: we need a sequence that corresponds to a fixed structure, so we choose a random sequence and allow it to evolve into the one that corresponds to the structure (directional selection). This yields a sequence at the root of the tree that corresponds to the structure, that can be used as a starting point for the simulation of the tree itself under continued structural constraint (stabilizing selection). Structures Examined In this study, we examine the effect of dependence introduced via structural constrain in both RNA and proteins. For RNA, we will focus on two structures: the Bombyx mori R2 element reverse transcriptase 3' untranslated region, a 300-nucleotide structure previously examined by Mathews et al. (1997), and the eukaryotic 5S rRNA structure (119 nucleotides) examined by Yu and Thorne (2006). Each simulated parameter set using these structures include 400 and 1000 replicates, respectively. For proteins, we will also use two structures: mammalian myoglobin (Physester catodon; PDB code 1MBD; 459 nucleotides) and 6-hydroxymethyl-7-8-dihydroxypterin pyrophosphokinase (Escherichia coli; PDB code 1HKA; 474 nucleotides), both examined by Rodrigue et al. (2005). Simulations using these protein structures consist of 1000 and 500 replicates, respectively. Energy at Stationarity Because we have described the energy of a sequence as measuring how well a sequence fits a structure, we can visually inspect this process of approaching and sampling from the stationary distribution of sequences under the selective constraint by monitoring the energies of the sequences sampled. Figure 3a shows the energies of a sequence, initially sampled at random, evolving continuously under independence. As expected, the sequences sampled have similarly high energies because the vast majority of the 4 N possible sequences will not naturally fit the structure well. Contrast this with Figure 3b, which shows the energies of a sequence, similarly sampled at random originally, but evolving under the model of dependence. The sequences sampled converge to an area of the sequence space with much lower energies and remain there indefinitely. This indicates that the selective constraints of the structure limit the sequences that can be sampled to those that fit the structure reasonably well. That the chain fails to leave this area of the sequence space is an indication that we are in fact sampling sequences from the stationary distribution. In this manner, we can empirically determine the minimum branch FIGURE 3. Energies sampled every 100 substitutions from a continuously evolving sequence. a) Independence among sites. Energies sampled are similar to that initially sampled at random. b) Dependence due to structural constraint. Low energies indicate that sequences sampled are those that fit the structure. The sequence evolves from a randomly sampled starting state of high energy to sample those states of low energy that correspond to the structure. length necessary to sample from the stationary distribution with high probability prior to the simulation of sequences along the trees. Rate Variation among Sites We expect that the constraints imposed by structures will affect among-site rate variation; at stationarity, a site that is tightly constrained will experience a low rate of substitution relative to unconstrained sites. To confirm this, we simulated the evolution of a sequence at stationarity for varying levels of dependence and observed the number of changes occurring at different sites in the sequence. The results are shown in Figure 4. Under independence, RNA stem and loop positions observed similar rates of substitution (Fig. 4a), whereas under dependence (z = 0.01) the rate of substitution at stem position decreased and at loop positions increased (Fig. 4b). Further increasing the level of dependence did not seem to affect the change in substitution rate (data not shown). For proteins, the substitution process at a particular site can depend upon a number of other sites determined by the site's location in the folded protein. Whereas the rate of substitution observed was similar regardless of the number of contacted other sites under (Fig. 4c), under dependence we observed a clear negative correlation between the number of sites upon which a particular site is dependent and the rate of change that the site experienced (Fig. 4d). Similar results were obtained for all RNA and protein structures examined. It is worth noting that the RNA model induces a higher rate of substitution among loop sites than among stem sites. This is quite different from what is observed in alignments of certain RNAs, in which loop regions are often highly conserved. This difference is in part because, whereas this model accounts for the dependencies introduced by the maintenance of the structure of the molecule, the model does not explicitly consider its function. If loop regions are conserved due to functional constraint of binding to another molecule, the dependence of these sites on their binding site is not captured by our model, which looks only at the structure of the single RNA. Clearly, both types of constraint are biologically relevant, and it would be straightforward to imagine expanding the model beyond a single sequence to consider two RNAs (or proteins) that interact, introducing dependencies both within and between the structures. It should also be noted that although our model does not capture stabilizing selection on RNA loop regions, neither do the independence-assuming models typically used for phylogenetic inference. Effect of Dependence on RNA We simulated sequences on the four-taxon tree using the predicted RNA structure of the Bombyx mori R2 element reverse transcriptase 3' UTR (300 nucleotides) previously examined by Mathews et al. (1997). We did this for a constant tree size (V = 1.75) and for a range of branch length proportions at varying levels of dependence, and then estimated the topology from the data assuming independence. Figure 5a shows the accuracy of ML at estimating the true topology for these simulated data sets (400 replicates). The shorter the internal branch, the more difficult is the estimation problem. As expected, ML performs well for all tree shapes on data simulated with no dependence among sites. But as the level of dependence among sites increases the accuracy of ML decreases, particularly when the internal branch is short. Perhaps most striking is the decrease in accuracy resulting from even small levels of dependence in the data (z = 0.1), with accuracy falling to nearly 50% when the internal branch is short. These simulations also provide a sense for just how short the internal branch must be before ML will begin to see a decrease in accuracy resulting from dependent evolution among sites. Whereas it might be encouraging if ML had difficulty only when the internal branch was quite short, this is not the case. Appreciable decreases in accuracy are observed over a wide range of internal branch lengths, indicating that the effects of dependent evolution on phylogenetic inference are not restricted to extreme topological cases. It is important to note that although our structural model does induce rate variation among sites, this is at least partially accounted for in the GTR+Γ model used for analysis. This means that observed decreases in accuracy are more likely to result from differences resulting from the context-dependent nature of the substitution process induced by the model of structural constraint. The decreased accuracy resulting from dependence in the data is not a particular property of ML however. Figures 5b,c show the analysis of the same data using neighbor-joining (with ML distances) and parsimony, respectively. Whereas the baseline expectations of how well the methods will perform when the data are independent differ, the trend is the same for all methods that assume independence: The effect of dependence is to reduce the accuracy of the methods, particularly when the problem is difficult, as is the case when branches differ markedly in length. We might note that neighborjoining appears to do as well as ML in many cases, and in some cases seems to perform better. It would be tempting to attempt to draw broader conclusions from these simulations about the relative performance of these methods, but it must be remembered that we show here only a small portion of the possible parameter space of topologies, branch lengths, model parameters and have only shown a four-taxon case using a single structure. We refrain from drawing any such conclusions, and instead focus on the observation that all of these methods seem to suffer by failing to account for the dependence. To examine whether these results were specific to the structure examined or more general, we simulated data using the eukaryotic 5S rRNA (119 nucleotides) as the reference structure. We did so on a slightly shorter tree length (V = 1.0) over the same range of branch length proportions and levels of dependence (1000 replicates). The analysis of these simulated sequence sets (Fig. 5d-f) are qualitatively consistent with the previous results: Methods that assume independence experience a reduction in accuracy over a wide range of branch length proportions as the level of dependence increases. This suggests that these decreases in accuracy are not specific to a single structure but are a more general property of the effect of dependent evolution in RNA. The performance of phylogenetic methods assuming independence is also affected by the overall length of the underlying tree as well as its topology. Figure 6 shows the effect on accuracy of ML estimation using simulated R2 element RNA sequences over a range of branch length proportions on trees of total length 0.25, 1.0, and 1.75. The effect of a fixed level of dependence (z = 0.5) is to reduce accuracy relative to independence (z = 0.0) as shown before, but the effect is greater when the overall tree length is greater. On a larger tree (Fig. 6c) reductions in accuracy are observed at small branch length proportions, whereas on a small tree (Fig. 6a) the branch length proportion must be larger before reductions in accuracy are observed. This demonstrates how tree length and topology may interact to cause difficulties in estimation on dependence-containing data; dependence seems 2011 NASRALLAH ET AL.-DEPENDENT SUBSTITUTION MODELS 67 FIGURE 5. The accuracy of independence-assuming phylogenetic methods to infer the correct topology using RNA sequences constrained by structure simulated on a tree of total length V = 1.75. As the level of dependence in the data (z) increases, the methods are increasingly unable to infer the correct topology. This is especially true as the branch length ratio (a/b) becomes large and the problem becomes difficult. Structures: Bombyx mori R2 element reverse transcriptase 3' UTR (R2) [300 nucleotides, 400 replicates] and 5S rRNA (5S) [119 nucleotides, 1000 replicates]. Methods: maximum likelihood GTR+Γ (ML), neighbor-joining using ML distances (NJ), parsimony (MP). to have the greatest effect when the tree is very large and the internal branch is short. Results for neighborjoining and parsimony were qualitatively similar, and for brevity we will largely focus the remainder of the four-taxon case discussion on results for ML, which are representative of trends observed using all methods examined. Effect of Dependence on Proteins For proteins, the dependence involves two components: pairwise interactions and solubility constraints. To explore how each of these affect inference, we used the reference structure of mammalian myoglobin (Physeter catodon; PDB code: 1MBD; 459 nucleotides), previously studied by Rodrigue et al. (2005), to simulate data on a tree topology with a/b = 5 and a total length V of 1.3 (Fig. 7a) or 2.08 (Fig. 7b) across a wide range of dependence parameter values (1000 replicates). The larger tree shows the same trend as RNA: increased levels of dependence result in decreased accuracy. However, the effect seems to be less severe, particularly when the level of dependence is small. Furthermore, the dependence due to pairwise interactions has a much greater effect than dependence due to solubility constraints. Importantly, there is very little effect whatsoever observed when the tree length is small until levels of dependence become quite large indeed, even when the topology itself poses a moderately challenging problem. We then repeated these simulations using the reference structure of 6-hydroxymethyl-7-8-dihydroxypterin pyrophosphokinase (Escherichia coli; PDB code: 1HKA; 474 nucleotides), also examined previously by Rodrigue et al. (2005). Whereas the structures of these two proteins are quite different, the results using the two structures FIGURE 7. Accuracy of phylogenetic inference using ML using sequences generated under varying levels of dependence due to protein structure constraints: solubility (s) and pairwise interactions (p). Accuracy is reduced when dependence is strong and tree length is large. All panels represent the same tree topology (a/b = 5). (Fig. 7c,d). There is some additional variance due to fewer replicates (500), but the trend is the same. This suggests that dependence among sites in proteins may have similar effects on phylogenetic inference regardless of the precise nature of the structure. It is encouraging to see that for proteins, unlike RNA, small levels of dependence in the data do not seem to have a strong effect on the accuracy of phylogenetic methods. Estimates of the level of dependence in actual data will be considered below, but another consideration is whether or not the protein model, which reduces protein structure to two parameters of solubility and pairwise interactions, can adequately account for the complexity of actual protein structures. Domany (1998, 2000) and Park et al. (2000) have argued that there are limits to the utility of pairwise interaction potentials and hydrophobicity constraints in protein structure prediction. It is likely that the structural fitness of a sequence would be more accurately represented by the actual Gibbs free energy of the sequence, but at present this approach is computationally demanding. Although the simplified approach adopted here is well-justified, the conclusions drawn for proteins may not be the final word. Effective Sequence Length How phylogenetic methods behave when the data are neutral and independent may be used as a reference for describing how phylogenetic methods perform when ideal conditions are not met. We may consider the effective sequence length (L e ) as the length of independent neutral sequence that behaves in the same manner (in terms of phylogenetic accuracy) as our dependencecontaining sequences. This is similar in spirit to the concept of an effective population size in population genetics. Because we expect dependence to introduce correlated substitutions, we expect the effective sequence length to be smaller than the actual sequence length (Huelsenbeck and Nielsen 1999). How much smaller is of interest and will depend on several factors including the actual sequence length, the nature of the structural constraints, the relative importance of the dependence, and the topology and length of the underlying tree. Figure 8 quantifies the effective sequence length for one case examined. Each panel represents a different underlying tree topology (a/b) of the same overall tree length (V = 1.75). For each topology, we first simulated under independence sequences of different lengths and assessed the phylogenetic accuracy obtained by using these sequences. Shown in Figure 8 as the curves, these indicate the expected accuracy when using sequences of n independent neutral sites. For each topology, we then simulated RNA sequences of length 300 nucleotides (using the R2 reverse transcriptase structure) under dependence (z = 0.1) and assessed the accuracy using these dependent sequences, indicated by the horizontal lines. Where these observed (dependent) accuracies intersect our expected (independent) curve, we can project to the x-axis to estimate the effective sequence length for these dependence-containing data. The presence of dependence in the data results in a large decrease in effective sequence length, particularly for topologies in which the internal branch is relatively short. One could argue that we might have easily predicted the effective length for RNA by simply considering paired sites to be as informative as a single unpaired site. In the structure used for the simulated RNA sequences, there were 168 stem and 132 loop positions, which by this method would predict an effective sequence length of 216 nucleotides. Alternatively, if all stem positions were considered to be invariable, the effective sequence length would be predicted to be 132 nucleotides. However, what we observe is that the dependence in our data leads to much lower accuracies, and subsequently much lower effective sequence lengths than both of these expectations, observing effective sequence lengths of less than 100 nucleotides. This implies that models simply accounting for covariation in the data are not accounting for all aspects of structural constraint and that these structural constraints lead to greater information loss. Although we suggest that the concept of effective sequence length is useful for thinking about the effect of dependence on data, and is particularly useful for assessing these effects in our simulations, determining the effective sequence length requires knowledge about the true tree and importance of the structural constraints. The practicing systematist would therefore need to make some very strong assumptions in order to use the concept of effective sequence length to explicitly guide analysis. Larger Data Sets In order to understand how dependence among sites might affect phylogenetic inference we have focused on the four-taxon case using a single sequence/structure. This allowed us to thoroughly explore the relevant parameter space and gain some intuition for when we might expect error. However, using only four taxa or such a limited amount of sequence data is hardly something done in practice. It would therefore be useful to understand how the effects we have observed extend when the methods are presented with more taxa or more sequence data. To address the question of how the methods perform on trees containing more than four taxa we simulated sequences on a 22-taxon tree using the R2 element RNA structure. In this tree (Fig. S1, available from http://www.sysbio.oxfordjournals.org/), all terminal branches are of the same length (0.05 expected substitutions per site) and are five times longer than internal branches (0.01 expected substitutions per site). In some sense, this makes for a relatively easy estimation problem: unlike the four-taxon case, all terminal branches are of equal length, and the overall tree length is quite 70 SYSTEMATIC BIOLOGY VOL. 60 FIGURE 8. The effective sequence length (Le) as a means of quantifying the phylogenetic information content of a sequence that contains dependence. All panels represent a fixed tree length (V = 1.75) and level of dependence (z = 0.1). a) a/b = 0.5. b) a/b = 2. c) a/b = 3. d) a/b = 4. e) a/b = 6. f) a/b = 8. The plotted curves indicate the accuracy of ML on these trees using independent data of varying lengths or the expected accuracy if the data were independent. The accuracy of ML on the simulated RNA sequences (R2, actual length = 300 nucleotides) on each topology is shown by the horizontal lines. Where these horizontal lines cross the curve, they drop to the x-axis to estimate the effective sequence length: the length of independent neutral sequence that displays the same amount of error in estimation that the actual dependence-containing sequence displays. small. We simulated 500 RNA data sets on this tree for each of a range of levels of dependence and analyzed these data sets using the same methods used in the four-taxon case. We calculated the Robinson-Foulds metric (Robinson and Foulds 1981) to compare the estimated tree with the true tree, and the results are shown in FIGURE 9. Accuracy of independence-assuming phylogenetic methods for 22-taxon simulations of RNA constrained by structure (R2; 500 replicates). The Robinson-Founds distance metric compares the estimated tree to the true tree for data sets under varying levels of dependence (z). For all methods, small amounts of dependence introduce error in tree estimation. a) ML, b) neighbor-joining, c) parsimony. Figure 9. As expected, dependence in the data increases the amount of topological estimation error, in spite of the estimation problem not being an incredibly difficult one. Notably, even small amounts of dependence are sufficient to cause appreciable decreases in accuracy. We expect that on trees of greater length or containing variance in branch lengths might present more challenging problems and therefore be more sensitive to the effects of dependence. Although these simulations are hardly a thorough exploration of the space of possible trees larger than four taxa, they give a sense for how the problems observed might scale with the number of taxa. Addressing the question of how very long dependence-containing sequences affect the analyses is not as straightforward. This is because one of the limitations of conditioning on an actual fixed structure is that the sequences are constrained to a fixed length. To test this question, we concatenated our R2 element data sets to create three very long (≥ 30, 000 nucleotides) sets of sequences. We similarly concatenated 5S sequences to create two sets of sequences (≥ 45, 000 nucleotides). We opted to use the same structure repeatedly to ensure that the same kind of dependence is introduced as there is no guarantee that different structures will not contain conflicting signal. The results are consistent with what was observed on shorter sequences (Fig. S2, available from http://www.sysbio.oxfordjournals.org/). When the dependence is large (z ≥ 0.5), ML fails to estimate the correct topology when the problem is difficult. When the dependence is small (z = 0.1), ML is able to recover the true tree most of the time. Curiously, the neighborjoining algorithm (using GTR+Γ distances) performs very well for all levels of dependence on all trees. Parsimony behaves qualitatively similar to the results on shorter sequences (Fig. 5c,f), Although these limited number of replicates are hardly conclusive, they give a sense for how these independence-assuming methods might handle a great deal of dependent sequence. Estimates of Dependence Our simulations have shown that failure to account for dependence among sites, such as dependence due to structural constraints, can greatly impair inference of the underlying tree topology. We have shown this for a wide range of levels of dependence, but it would be useful to have a sense for what might be reasonable levels of dependence to expect in actual data. Yu and Thorne (2006) estimated the level of dependence due to secondary structure for a set of eight 5S rRNA sequences to be 0.3661. In our simulations, we observe a significant impact on accuracy at lower levels of dependence than this (z = 0.1; see Fig. 5). This implies that failure to account for secondary structure of RNA may often lead to inaccurate inference of the true topology. However, it is important to note that these kinds of models allow two methods of specifying the importance of structural constraint. One is an explicit level of dependence as specified by the tuning parameters discussed here (z for RNA, p and s for proteins). Another form of constraint is more implicit, namely how much flexibility is allowed in the structure. Here, we have presented a model in which the (implicit) requirements of the structure are strict, but the (explicit) level of dependence has been varied. Yu and Thorne (2006), however, allowed more internal flexibility in the structures they examined. This implies that the explicit level of dependence in a model such as what we have presented might be lower than what Yu and Thorne (2006) presented because the implicit constraint is greater. How much lower is a reasonable question and will be important in determining the level of decreased phylogenetic accuracy to be expected as a result. For proteins, the story is also complicated. Although it is clear that there is dependence due to secondary structure in proteins (Thorne et al. 1996;Goldman et al. 1998), estimates of the level of dependence vary considerably. The model we have described and the model under which these estimates were obtained utilized similar levels of implicit flexibility, so we focus on the estimates themselves. Rodrigue et al. (2005) a model that involved the same pair potentials we employ, but it did not utilize solubility constraints. They estimated levels of dependence due to pairwise interactions to be in the range of 0.36 − 0.70. Robinson et al. (2003) used a model that included both pair potentials and solubility and obtained estimates of pairwise dependence an order of magnitude less than Rodrigue et al. (2005) (0.028 − 0.038) while also estimating the dependence due to solubility (0.88 − 0.95). The large difference in pairwise interaction estimates could be due to differences in the modeling of the pairwise interactions or because the Rodrigue et al. (2005) model lacked solubility constraints. Choi et al. (2007) used the Robinson et al. (2003) model to estimate pairwise and solubility dependence for a wide range of proteins (Choi et al. 2007; Fig. 1), which not surprisingly agree with the Robinson et al. (2003) estimates. The difference between the Robinson et al. (2003) and Choi et al. (2007) estimates and the Rodrigue et al. (2005) estimates is an important one. As we have shown, pairwise interactions of the level Robinson et al. (2003) describe have little effect on our ability to estimate the true topology in spite of our assumptions of independence. If however pairwise dependence is of the level Rodrigue et al. (2005) describe, the impact on phylogenetic estimation is quite large. Use of Energy as Fitness The use of the energy of a sequence on a particular structure is but one possible surrogate for the fitness of a sequence and may have its limitations. It is possible, for example, that a given sequence might be able to fold well into many possible structures; that although a given sequence might have a low energy on the structure of interest, it might have an even lower energy on an alternate structure. This implies that this sequence would in reality spend more time folded in the alternative structure than the one of interest. In this case, we VOL. 60 might argue that the sequence energy itself is not a good proxy for the fitness of the sequence. A better surrogate for fitness in this case might be the probability that a sequence will fold into the desired structure. However, this would involve considering the energy of a sequence on all its possible structures, and as we are allowing the sequence itself to change this becomes computationally prohibitive, particularly as the sequence length increases. Additional Model Limitations The model we have presented is one in which dependence among sites results from the existence of a structure that must be maintained in order to perform some function. One limitation to this is that we do not allow the structure itself to evolve along the tree. This might be reasonable for short phylogenetic distances, but the fact remains that even closely related sequences vary widely in their structural homology across taxa. Accounting for variance in the structural constraints across the tree will be a challenge for future research. Another way in which the kind of model we have described might be developed is to allow for more than one substitution at a time. Huelsenbeck and Nielsen (1999) developed a compound Poisson model that allows for this, and it might be a natural pairing with the type of model described here; evaluating the energy/fitness of a sequence two substitutions away is a straightforward extension. Allowing more than one substitution at a time might be particularly useful when the intrinsic constraints are very strong, enabling sequences to cross fitness valleys more easily. CONCLUSIONS We have shown that failure to account for dependence among sites due to secondary and tertiary structure can lead to inaccurate estimation of the underlying tree topology. This is particularly true when the dependence is strong as may be the case with RNA, when the internal branch is relatively short, and when the overall tree length is large. These findings have direct implications for anyone interested in phylogenetic estimation or analyses dependent thereupon. We have also shown that the effect is stronger than might have been expected under simpler models of dependence, such as considering paired RNA sites as one. This indicates that there is room for improvement in phylogenetic methods by accounting for the nature of the dependencies in the data. We have introduced the concept of an effective sequence length as an intuitive means of quantifying the effects of dependence and have presented a general method of simulating data on phylogenetic trees under complex models of evolution. Although in this paper we have focused on RNA and protein structures to introduce the dependencies among sites, the findings here may extend to the general case in which there may be dependence among characters. Morphological characters, for example, may contain large amounts of dependence, although it may be much more difficult to model the particular nature thereof. But our findings that the presence of dependence in the data, if unaccounted for, may lead to error in phylogenetic estimation should hold regardless of how well we understand the nature of the dependence itself. This suggests that in cases where we may be unable to model the dependence, being able to simply detect the presence of dependence in the data might be valuable.	2018-04-03T00:49:55.435Z	2010-11-15T00:00:00.000	{ "year": 2010, "sha1": "03f174e4de936870344fbf9ac0c38f3f6d542a64", "oa_license": "CCBYNC", "oa_url": "https://academic.oup.com/sysbio/article-pdf/60/1/60/24208042/syq074.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "03f174e4de936870344fbf9ac0c38f3f6d542a64", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
15910349	pes2o/s2orc	v3-fos-license	Synthesis and Cytotoxicity of Cyanoborane Adducts of N 6 -Benzoyladenine and 6-Triphenylphosphonylpurine N 6 -Benzoyladenine-cyanoborane (2), and 6-triphenylphosphonylpurine-cyanoborane (3) were selected for investigation of cytotoxicity in murine and human tumor cell lines, effects on human HL-60 leukemic metabolism and DNA strand scission to determine the feasibility of these compounds as clinical antineoplastic agents. Compounds 2 and 3 both showed effective cytotoxicity based on ED50 values less than 4 μg/ml for L1210, P388, HL-60, Tmolt3, HUT-78, HeLa-S3 uterine, ileum HCT-8, and liver Hepe-2. Compound 2 had activity against ovary 1-A9, while compound 3 was only active against prostate PL and glioma UM. Neither compound was active against the growth of lung 549, breast MCF-7, osteosarcoma HSO, melanoma SK2, KB nasopharynx, and THP-1 acute monocytic leukemia. In mode of action studies in human leukemia HL-60 cells, both compounds demonstrated inhibition of DNA and protein syntheses after 60 min at 100 μM. These compounds inhibited RNA synthesis to a lesser extent. The utilization of the DNA template was suppressed by the compounds as determined by inhibition of the activities of DNA polymerase α, m-RNA polymerase, r-RNA polymerase and t-RNA polymerase, which would cause adequate inhibition of the synthesis of both DNA and RNA. Both compounds markedly inhibited dihydrofolate reductase activity, especially in compound 2. The compounds appeared to have caused cross-linking of the DNA strands after 24 hr at 100 μM in HL-60 cells, which was consistent with the observed increased in ct-DNA viscosity after 24 hr at 100 μM. The compounds had no inhibitory effects on DNA topoisomerase I and II activities or DNA-protein linked breaks. Neither compound interacted with the DNA molecule itself through alkylation of the nucleotide bases nor caused DNA interculation between base pairs. Overall, these antineoplastic agents caused reduction of DNA and protein replication, which would lead to killing of cancer cells. nucleotides have been reported to be even more potent in suppressing the growth of murine and human cancer cells/2/. A common feature of all of these derivatives was that they effectively suppressed DNA synthesis and the activities of enzymes involved in the nucleic acid metabolism. Selected compounds demonstrated DNA strand scission with inhibition of DNA topoisomerase II activity. Based on these previous studies, NCbenzoyladenine-cyanoborane (2), 6-triphenyl-phosphonylpurine-cyanoborane (3) were selected for investigation in human leukemia HL-60 cells for cytotoxicity, effects on metabolic events, and DNA strand scission to determine the feasibility of these compounds as clinical antineoplastic agents. METHODS The cyanoborane adducts of the substituted purines were prepared via a Lewis acid exchange reaction/9/, by using excess (e.g., 4 molar equivalents) triphenylphosphine-cyanoborane (1) and the amine (purine) in dry dimethylfonnamide (DMF) at 55-70C under nitrogen atmosphere (Scheme 2). In the Lewis acid exchange reaction, a weakly basic or bulky amine or phosphine, as its substituted borane (C6H5)3P.HB r + NaBH3C N Reflux N2(g) (C6H5)3P:BH2CN + HBr + H 2 Scheme 1 Synthesis of the Lewis acid exchange reagent, triphenylphosphine-cyanobomne (1). adduct (e.g., Ph3P:BH2CN), is exchanged for a more basic or less bulky amine, (e.g., a substituted purine). These boron exchange reactions must be carried out under anhydrous conditions to avoid coordination of the cyanoborane to water and subsequent degradation to boric acid. The Lewis acid exchange reaction is a general route which has also been used in the preparation of other cyanoborane adducts of aliphatic/3, 9/, aromatic /3,8/, and heterocyclic /8,9/ amines, as well as their borane and carboxyborane adducts /9/. Triphenylphosphine-cyanoborane (_1) was prepared as previously reported/9, 10/(Scheme 1) by refluxing Ph3P HBr and NaBH3CN in dry THF under nitrogen atmosphere. The reaction of 6-chloropurine and triphenylphosphine-cyanoborane produced the unexpected product, 6-triphenylphosphonylpurinecyanoborane (). This resulted when the free triphenylphosphine, formed after the Lewis acid exchange, underwent aromatic nucleophilic substitution displacing the chlorine in the 6 position of the purine ring. Synthesis of Compounds All chemicals and reagents were obtained from Aldrich Chemical Company (Milwaukee, WI) and used as received except for dry solvents which were dried and distilled using standard procedures/11/. TLC was performed using silica gel 60F 254 plates (silica gel on plastic, Aldrich Chemical Company). Melting points 2O 7bnva C.Scarlett et al. Metal-Based Drugs were obtained on a Thomas-Hoover Uni-melt apparatus (capillary method), and were uncorrected. IR spectra were obtained on a Perkin-Elmer 1600 FTIR spectrometer in a potassium chloride liquid cell in CHCI3 or CDC13. NMR spectra were obtained on a 300 MHz Bruker Avance FT-NMR spectrometer using tetramethylsilane as an external standard for H and laC spectra and BF3:OEt2 for IB spectra (6 0 ppm). Preparation of Triphenylphosphine-cyanoborane (1_): To a mixture of 4.02 g (11.7 retool) of triphenylphosphine hydrobromide and 40 mL of dry THF was added 0.93 g (14.8 mmol) of sodium cyanoborohydride. The suspension was stirred under N2 (g) at reflux for 10 hr. The mixture was cooled to RT, filtered and the solid washed with THF. The filtrate and washings were combined and the solvents were removed under reduced pressure. The white solid was washed with cold water, then cold ethyl ether. After air drying, 3 Preparation of N6-Benzoyladenine-cyanoborane (2) To a solution of 10.01 g (33.23 retool, 4 Eq.) of triphenylphosphine-cyanoborane in 30 mL of dry DMF was added 2.03 g (8.49 mmol) of N-benzoyladenine. The solution was stirred under N2 (g) at 70C for 10 days. The solution was cooled to RT, filtered and the solid washed with methanol. The filtrate and washings were combined and silica gel was added until all of the liquid was adsorbed. The solvents were removed under reduced pressure. The product was purified by column chromatography on silica gel using dichloromethane:methanol (95:5, Rf 0.69). A partial yield/12/of 0.0858 g (0.309 retool, 3.6%) of tan solid To a mixture of 15.76 g (52.34 retool, 4 Eq.) of triphenylphosphine-cyanoborane in 30 mL of dry DMF was added 2.02 g (13.1 mmol) of 6-chloropurine. The mixture dissolved upon heating and was stirred under N2 (g) at 70C for 10 days. The solution was cooled to RT and the resulting suspension was filtered and the solid washed with methanol. The filtrate and washings were combined and silica gel was added until all of the liquid was adsorbed. The solvents were removed under reduced pressure. The product was purified by Metal-Based Drugs" Normal fibroblasts 1788 were also used to test cytotoxicity of the agents. The NCI protocol was used to assess the cytotoxicity of the test compounds and standards in each cell line. Values for cytotoxicity were expressed as EDs0 lag/ml, i.e. the concentration of the compound inhibiting 50% of cell growth. EDs0 values were determined by the trypan blue exclusion technique /13/. A value of less than 4 lag/ml was required for significant activity of growth inhibition. Solid tumor cytotoxicity was determined utilizing crystal violet/MeOH and read at 580 nm (Molecular Devices)/14/. Incorporation Studies Incorporation of labeled precursors into 3H-DNA, 3H-RNA and all-protein for 10 6 HL-60 leukemia cells was obtained/15/using a concentration range of 25, 50 and 100 gM of the test agents 2 and 3 over a 60 min incubation. The incorporation of 4C-glycine (53.0 mCi/mmol) into purines /16/ and the incorporation of 4Cformate (53.0 mCi/mmol) into pyrimidines /17/ was determined in a similar manner. Enzyme assays Studies for the inhibition of various enzyme activities were performed by first preparing the appropriate HL-60 leukemia cell homogenates or subcellular fraction, then adding the drug to be tested during the enzyme assay. For the concentration response studies, inhibition of enzyme activity was determined at 25,50 and 100 gM of compounds 2 and 3, after 60 rain incubations. DNA polymerase c activity was determined in cytoplasmic isolated extracts [18]. The polymerase activity for was determined with H-TTP /19/. Messenger-, ribosomal-and transfer-RNA polymerase enzymes were isolated with different concentrations of ammonium sulfate; individual RNA polymerase activities were determined using 3H-UTP /20.21/. Ribonucleoside reductase activity was measured using 4C-CDP with dithioerythritol /22/. The deoxyribonucleotides 14C-dCDP were separated from the ribonucleotides by TLC on PEI plates. Thymidine, TMP and TDP kinase activities were determined using 3H-thymidine (58.3 mCi/mmol) / ct-DNA studies After deoxyribonucleoside triphosphates were extracted/32/, levels were determined by the method of Hunting and Henderson/33/with calf thymus DNA, E. coli DNA polymerase l, non-liniting amounts of the t'ol. 9. Nos. [1][2]2002 Synthesis and C)'totoxiciO, qfQ:anoborane Adducts" three deoxyribonucleoside triphosphates not being assayed, and either 0.4 mCi of (3H-methyl)-dTTP or (5-3H)-dCTP. The effects of compounds 2 and 3 on DNA strand scission was determined by the methods of tCi thymidine hnethyl-3H, 84.0 Ci/mmol/for 24 hr at 37C. HL-60 cells (107) were harvested and then centrifuged at 600 g X 10 min in PBS. They were later washed and suspended in ml of PBS. Lysis buffer (0.5 ml; 0.5 M NaOH, 0.02 M EDTA, 0.01% Triton X-100 and 2.5% sucrose) was layered onto a 5-20% alkaline-sucrose gradient (5 ml; 0.3 M NaOH, 0.7 KC1 and 0.01 M EDTA); this was followed by 0.2 ml of the cell preparation. After the gradient was incubated for 2.5 hr at room temperature, it was centrifuged at 12,000 RPM at 20C for 60 min (Beckman rotor SW60). Fractions (0.2 ml) were collected from the bottom of the gradient, neutralized with 0.2 ml of 0.3 N HC1, and measured for radioactivity. Thermal 30 mg/ml bovine serum albumin, mM ATP, 10 mM MgCI2 and 150 mM KCI. After 30 rain incubation at 37 C the reaction was terminated with 1% SDS and mg/ml proteinase K (v/v). After an additional hour of incubation, aliquots were applied to a 0.8% agarose TBE gel (v/v) containing 0.5 mg/ml ethidium bromide and 1% SDS (w/v). Following overnight electrophoresis at 30 v (constant), the gel was destained and photographed using a U.V-transilluminator and Polaroid film. Topoisomerase activity inhibition was assayed by a similar method. The enzyme reaction consisted of test drugs, 0.5 units of human topoisomerase [TopoGen, Inc., Columbus, OH], 0.5 lag of supercoiled PBR322 DNA in 50 mM Tris-HCl, pH 8.0, 100 mM KCI, 10raM MgC12, 2 mM 2-mercaptoethanol, 30 lug/ml nuclease-free BSA. Statistic Analysis Data is displayed in tables and figures as the means + standard deviations of the mean expressed as a percentage of the control value. N is the number of samples per group. The Student's "t"-test was used to determine the probable level of significance (p) between test samples and control samples. Compound 2 was examined for its mode of action in HL-60 leukemia cells (Table 2). DNA and RNA synthesis after 60 minutes was slightly inhibited by 35% and 25% at 100 laM. Protein synthesis after 60 minutes at 100 laM inhibited 55% at 100 .tM. Utilization of the DNA template showed that the agent inhibited DNA polymerase ct activity by 50% at 100 tM,.mRNA polymerase 41%, rRNA polymerase 37%, and tRNA polymerase 52%. A number of enzyme activities were slightly reduced but were not significantly different from the control. Ribonucleotide reductase activity after 60 minutes was inhibited only 12%, while de novo purine synthesis was inhibited 18%. Compound 2 mildly suppressed PRPP amido transferase activity at 100 IuM by only 6% with an 11% reduction of IMP dehydrogenase activity. Carbamyl phosphate synthase and aspartate transcarbanylase activities were slightly inhibited 14% and 31%. While thymidylate synthase and thymidine kinase activities were increased by 1% and 35%, TMP and TDP kinase was slightly inhibited 22% and 31%. Dihydrofolate reductase activity was markedly inhibited 85%. Studies with ct-DNA showed that compound 2 had no effect on ct-DNA ultraviolet absorption between 220 and 340nm. HL-60 DNA strand scission studies after 24h incubation at 100 laM revealed that compound 2 caused DNA cross-linking ( Figure 1). This was consistent with the increase in ct-DNA viscosity after 24 hr at 100 pM. Deoxyribonucleotide levels were all slightly reduced after 60 min incubation at 100 laM. Compound 3 was also examined for its mode of action in HL-60 leukemia cells (Table 3). DNA and RNA synthesis after 60 minutes was slightly inhibited 35% and 10% at 1001aM. Protein synthesis after 60 minutes at 100 laM was inhibited 48% at 100tM. Utilization of the DNA template was moderately inhibited at 100 pM with inhibition of DNA polymerase e activity 20%, mRNA polymerase activity 44%, rRNA polymerase activity 40%, and tRNA polymerase activity 39%. Ribonucleotide reductase activity was inhibited only 25%, while de novo purine synthesis was inhibited 35% after 60 rain. Synthesis and Q,totoxiciv oj'C'yanoborane Adducts Synthesis and C)totoxicit.y of C,yanoborane Adducts transferase activity at 100 laM by only 12% with a 15% reduction of IMP dehydrogenase activity. Carbamyl phosphate synthase activity showed an increase of 14%, while aspartate transcarbanylase activity was inhibited 42%. Only thymidylate synthase activity' was markedly suppressed 76%, with thymidine kinase activity marginally inhibited 28%, TMP kinase activity 43% and TDP kinase activity 46%. Dihydrofolate reductase activity was suppressed 63%. Studies with ct-DNA showed that compound 3 had no effect on ct-DNA ultraviolet absorption between 220 and 340nm. HL-60 DNA strand scission after 24h incubation at 100 7"alrcl C.Scarlett et al. A4etal...Based Drugs laM revealed that compound 3 caused DNA cross linking (Figure 1) which was consistent with the observed increased in ct-DNA viscosity after 24 hr at 100 gM. Deoxyribonucleotide pools were slightly redtced afier 60 min incubation with agents at 100 laM. Human topoisomerase and II activity was not inhibited by compounds 2 or 3 at 100 laM. DISCUSSION N6-Benzoyladenine-cyanoborane (2), and 6-triphenylphosphonylpurine-cyanoborane (3) proved to be cytotoxic in suspended cancer cells. Surprisingly these compounds were also cytotoxic in solid liver Hepe-2 and ileum HCT-8 carcinoma. In mode of action studies in human leukenic HL-60 cells, both compounds demonstrated inhibition of DNA and protein syntheses after 60 rain at 100 gM. These compounds inhibited RNA synthesis to a lesser extent. The utilization of the DNA template was suppressed by the compounds as determined by inhibition of the activities of DNA polymerase ct, m-RNA polymerase, r-RNA polymerase and t-RNA ploymerase which would cause adequate inhibition of the synthesis of both DNA and RNA. Because the d[NTP] pool levels were slightly reduced after 60 rain further inhibition of DNA synthesis would occur. Both compounds remarkably inhibited dihydrofolate reductase activity, especially compound 2. This would cause the reduction of the one carbon transfer for purine and pyrimidine syntheses/2/. However, the de novo synthesis of purine and pyrimidines was only.marginally affected by the compounds as were their regulatory enzyme activities /2/. Ribonucleotide reductase activity was moderately inhibited which would reduce the amount of ribonucleotide converted to deoxyribonucleotides for DNA synthesis. The reduction of TMP and TDP kinase activities would further reduced thymidine nucleotides levels demonstrated significantly by compound 3. Both compounds appeared to have caused cross-linking of the DNA strands after 24 hr at 100 laM in HL-60 cells, which was consistent with the observed increased in ct-DNA viscosity after 24 hr at 100 gM and lack of inhibition of DNA topoisomerase and II activities with no DNA-protein linked breaks. Neither compounds interacted with the DNA molecule itself through alkylation of the nucleotide bases nor caused DNA interculation between base pairs. Previously studied thymidine, inosine, cytidine, guanosine, and arbinoside cyanborane nucleotides have demonstrated a similar pattern of cytotoxicity on the growth of suspended murine and human tumor cells and solid human tumors. Those nucleoside and nucleotide cyanboranes inhibited DNA and protein synthesis, with a select few of the derivatives reducing RNA synthesis after hr/2/. Mutliple targets of the cyanboranes in DNA synthesis were demonstrated by the compounds. For the nucleoside cyanboranes the major sites of inhibition were IMP dehydrogenase and PRPP amido tranferase activities, suppressing de novo purine synthesis of Tnolh leukemia cells /2/. In contrast, the de novo synthesis of purine, pyrimidine and their regulatory enzyme activities were only marginally suppressed by the current compounds. Although similar nucleoside cyanoboranes inhibited dihydrofolate reductase activity, the current compounds were more potent. The boranated nucleosides cause a reduction of thymidylate synthase activity whereas only compound 3 decreased activity while compound 2 increased activity. However, both types of compounds inhibited TMP and TDP kinase activity and marginally reduced d[NTP] pools. Some of the nucleoside cyanoboranes caused DNA strand scission [thymidine] whereas others ribose and arabinoside] caused DNA cross-linking as the current compounds. However, none of the cyanboranes targeted the DNA molecule itself. I'ol. 9, Nos. I-2. 2002 Synthesis am.t Qtotoxicity of Cyanoborane Adducts CONCLUSION N-Benzoyladenine-cyanoborane (2) and 6-triphenylphosphonylpurine-cyanoborane (3) have been proven to be effective antineoplastic agents in their overall reduction of DNA and protein replication in respect to killing cancer cells. The inhibition of dihydrofolate reductase activity and/or thymidylate synthetase adds to the overall inhibition of DNA and protein synthesis. Even though both compounds showed DNA crosslinking, neither compound interacted with the DNA molecule itself through alkylation of the nucleotide bases nor caused DNA intercalation between base pairs. Sufficient activity was demonstrated by these cyanoborane derivatives to warrant further investigation as potential antineoplastic for clinical use.	2014-10-01T00:00:00.000Z	2002-01-01T00:00:00.000	{ "year": 2002, "sha1": "d312b97046de4285cb6993d7af38bf750b215069", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1155/mbd.2002.19", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "d312b97046de4285cb6993d7af38bf750b215069", "s2fieldsofstudy": [ "Chemistry", "Medicine" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
249525400	pes2o/s2orc	v3-fos-license	Non-linearity of Metabolic Pathways Critically Influences the Choice of Machine Learning Model The use of machine learning (ML) in life sciences has gained wide interest over the past years, as it speeds up the development of high performing models. Important modeling tools in biology have proven their worth for pathway design, such as mechanistic models and metabolic networks, as they allow better understanding of mechanisms involved in the functioning of organisms. However, little has been done on the use of ML to model metabolic pathways, and the degree of non-linearity associated with them is not clear. Here, we report the construction of different metabolic pathways with several linear and non-linear ML models. Different types of data are used; they lead to the prediction of important biological data, such as pathway flux and final product concentration. A comparison reveals that the data features impact model performance and highlight the effectiveness of non-linear models (e.g., QRF: RMSE = 0.021 nmol·min−1 and R2 = 1 vs. Bayesian GLM: RMSE = 1.379 nmol·min−1 R2 = 0.823). It turns out that the greater the degree of non-linearity of the pathway, the better suited a non-linear model will be. Therefore, a decision-making support for pathway modeling is established. These findings generally support the hypothesis that non-linear aspects predominate within the metabolic pathways. This must be taken into account when devising possible applications of these pathways for the identification of biomarkers of diseases (e.g., infections, cancer, neurodegenerative diseases) or the optimization of industrial production processes. INTRODUCTION Machine learning (ML) holds an increasingly prominent place in the field of biology. Indeed, it can lead to better results and has a large range of applications including: drug design using machine leaning algorithms such as the support vector machine (SVM) algorithm to perform structureactivity relationship analysis (Hartwell et al., 1999;Burbidge et al., 2001;Réda et al., 2020); directed protein evolution and enzyme function prediction Wu et al., 2019); reconstruction of biological routes (Kotera et al., 2013;Baranwal et al., 2020) or modeling and optimization of metabolic pathways (Zhang et al., 2019;Kim et al., 2020). With regard to the latter topic, several methods have been developed to analyze complex biological systems (Figure 1): • The knowledge-based model including kinetic models (Chance, 1943;Sel'Kov, 1968;Curto et al., 1997Curto et al., , 1998Hatzimanikatis et al., 1998;Visser and Heijnen, 2003;Liebermeister et al., 2010) and metabolic flux analysis methods (Fell and Small, 1986;Stephanopoulos, 1999); • The data-based model including ML algorithms and ensemble learning (Zelezniak et al., 2018;Ajjolli Nagaraja et al., 2019;Oyetunde et al., 2019); • The hybrid model including combinations of models or modified preceding methods (Cascante et al., 2002;Morgan and Rhodes, 2002). Although, these analyses are conducted on metabolic pathways, few of them are used to predict their fluxes. Among these few works on metabolic fluxes, it is interesting to highlight those of (Ajjolli Nagaraja et al., 2019). For the present work, the method of greatest interest is the data-based model and more precisely, ML. In fact, ML abounds in various methods and is a promising and growing approach that could greatly help to improve existing models, integrate multi-omics data and save researchers' time. Also, a distinction can be made between ML methods: some are linear (ridge and lasso regression, multivariate adaptive regression spline. . . ) and others are non-linear (artificial neural network, k-nearest neighbors, decision tree. . . ). In addition, the non-linearity of metabolic pathway is considered inherent to the pathway, depending on the non-linearity of chemical reaction kinetics and that related to regulatory processes (Song and Ramkrishna, 2013;Yasemi and Jolicoeur, 2021). Reviews on the fundamentals of Metabolic Control Analysis (Heinrich and Rapoport, 1974;Kacser et al., 1995) suggest mathematically that the pathway fluxes are non-linear. Moreover, experiments were done on glycolytic fluxes cells, where intact cells were incubated at different glucose concentrations (Marín-Hernández et al., 2020). The results showed a clearly hyperbolic behavior of the experimental data. Another experimental data used notably in this study indicates that the pattern is non-linear (Moreno-Sánchez et al., 2008;González-Chávez et al., 2015). These experimental data demonstrate that the pathway fluxes are non-linear. However, it has not yet been investigated whether linear or non-linear methods are more efficient in predicting pathway fluxes, and how to choose the appropriate one. Therefore, this study aims to elucidate the most appropriate methods to model three distinct metabolic pathways by designing and comparing five linear and eight non-linear machine learningbased methods (Figure 2): • The lower part of Entamoeba histolytica glycolysis ( Figure 3A), one of the major metabolic pathways of the parasite (Moreno-Sánchez et al., 2008;Muller et al., 2012;Pineda et al., 2015), through the use of a recently developed model (Lo-Thong et al., 2020); • The peroxide detoxification pathway of Trypanosoma cruzi ( Figure 3B) (González-Chávez et al., 2015, 2019; • The industrial-scale penicillin fermentation process of Penicillium chrysogenum ( Figure 3C) (Goldrick et al., 2015). Although these machine-learning approaches have been used to model metabolic pathways, few studies have focused on their usefulness in predicting flux (Wu et al., 2016;Ajjolli Nagaraja et al., 2019). Creating an efficient ML model depends on the availability of a large amount of experimental data (L'Heureux et al., 2017;Schmidt et al., 2019). The measurement of fluxes is cumbersome to carry out experimentally and hinders the possibility of having massive data. Because of the scarcity of these large experimental datasets in the literature, the methodology employed here consists of applying data augmentation to the first two pathways by using hybrid models (Figure 2). These hybrid models, called gray-box models, often predict better results than pure knowledge-based models or data-based models (Wei et al., 2018;Lo-Thong et al., 2020;Pintelas et al., 2020); in this study, the gray-box models consist of metabolic networks that include an adjustment term in one or more kinetic equations. In this study, models are based both on experimental datasets and predicted data coming from the previous graybox model. Here, we show that random forest models are the most effective, with a high predictive capacity starting from predicted and experimental enzyme activities or experimental parameters collected from a bioreactor. Also, two other models stand out as good ways to predict the flux or the final product concentration: XGBoost Linear and Cubist models. This shows the importance of using a non-linear model to design metabolic pathways. Based on these findings, we propose a means of decision support for researchers who wish to use machine learning techniques as a starting or a complementary method for modeling and for improving existing biological pathway models. By greatly increasing the quality of the outputs (flux prediction), machine learning opens the way to better drug target identification within a pathway, efficient disease modeling at molecular level and more efficient optimization for industrial production of metabolites. Experimental Procedures The lower part of glycolysis is reconstituted in-vitro in a reaction assay medium described in a recent work (Moreno-Sánchez et al., 2008), containing different recombinant enzymes (PGAM, ENO and PPDK). The reaction was started by adding 3PG (4 mM). An additional reaction is added, the formation of lactate with lactate dehydrogenase (Figure 3A), in order to follow the flux of the overall pathway by following the rate of NADH oxidation, for more details, see Moreno-Sánchez et al. (2008) works. Concerning the peroxide detoxification pathway (Figure 3B), each enzyme was individually titrated, while keeping the other parameters in the in-vitro system constant. The pathway flux was determined in parallel by observing NADPH oxidation, see González-Chávez et al. (2015) for more information. Finally, the experimental procedures that were followed to obtain penicillin FIGURE 1 \| Classification of metabolic pathway modeling methods according to their complexity and the year of first application in this field. The ellipse size is proportional to the occurrence of the method for pathway modeling in the literature. Three main groups are defined: knowledge-based model (Michaelis and Menten, 1913;Chance, 1943;Shapiro and Shapley, 1965;Garfinkel et al., 1970;Savageau, 1970Savageau, , 1988Fell and Small, 1986;Hatzimanikatis and Bailey, 1997;Curto et al., 1998;Heijnen, 2005;Liebermeister et al., 2010), data-based model (Wu et al., 2016;Cuperlovic-Culf, 2018;Ajjolli Nagaraja et al., 2019;Zampieri et al., 2019;Zhang et al., 2019;Kim et al., 2020) and hybrid model (Wiechert et al., 1997;Drysch et al., 2003;Antoniewicz et al., 2007;Nöh et al., 2007;Leighty and Antoniewicz, 2011;Antoniewicz, 2015;Pan et al., 2017;Yousoff et al., 2017;Heckmann, 2018;Oyetunde et al., 2019;Zampieri et al., 2019;Lo-Thong et al., 2020;Rana et al., 2020). Linear methods are represented in gray and non-linear ones are in blue. Methods in bold and white are those evaluated in this study. production data are described in the studies of Goldrick et al. (2015). Lower Part of Glycolysis Datasets Two datasets are constructed here by applying data augmentation, using a gray-box model detailed in one of the following sections. For the first one, an exploration around the experimental data flux (43 ± 10 nmol·min −1 ) from Moreno-Sánchez et al. (2008) at pH 6 is conducted. In fact, a sample of 2,000 normally distributed enzymatic balances was generated with the sample function on RStudio and resulted in a predicted flux between 0 and 53 nmol·min −1 with the gray-box model. The term balance refers to a set of activities of the enzymes involved in the cascade of reactions. The second dataset is made up of experimental and predicted (gray-box model) data of PGAM, ENO and PPDK activities and pathway flux (J). The experimental data are obtained from plots of Moreno-Sánchez study (Moreno-Sánchez et al., 2008) (only the dots), while the predicted data are obtained with the gray-box model developed in a recent work (Lo-Thong et al., 2020), by varying each enzyme activity from 0 to 1000 mU with a step of 25 mU. These datasets are shown in Supplementary Tables 7, 8 respectively. Peroxide Detoxification Datasets The second studied pathway consisted first of 58 experimental enzymatic balances and their corresponding flux. After applying data augmentation by using a gray-box model of this pathway, a bigger dataset of 1,671 data was obtained. As with the previous dataset, a combination of data normally distributed is generated with the sample function on RStudio, resulting in a predicted flux ranging from 0 to 11.46 nmol·min −1 . The new dataset is a mix of the previous experimental data and new predicted data of enzyme activities (TryR, TXN and TXNPx); final flux and is shown in Supplementary Table 9. FIGURE 2 \| Study workflow. Data from three different metabolic pathways are collected and used to build data-based models. Datasets that contain a small amount of data (n) go through a process of data augmentation, before being separated into two sets: training set and test set. Then, in order to predict the final flux or final product concentration, multiple ML models are built with the training set, while the test set is used to assess the final models. The resulting predictions are compared in a last step to evaluate model reliability. The Gray-Box Models The two following pathways are modeled with an open-source software called COPASI (Version 4.24) (Hoops et al., 2006): the second part of glycolysis and the peroxide detoxification pathway. This software is used for metabolic network design, analysis and optimization. The first gray-box model, representing the lower part of glycolysis, is taken from Lo-Thong et al. (2020) work. It is based on the use of enzyme properties, including kinetic parameters and kinetic equations. To enhance the flux predictions, they suggested adding an adjustment term to the PPDK kinetic equation. The whole process concerning the composition of this term is explained in the previous work (see Methodology part of Lo-Thong et al., 2020 and Supplementary Table 1). The second gray-box model represents the peroxide detoxification pathway and is built specifically for this study. It contains kinetic parameters and equations of three enzymes: TryR, TXN and TXNPx (Table 1). Also, we proposed to add two adjustment terms in TryR and TXNPx equations to improve flux predictions ( Table 1). These are determined in the same way as the terms used for the glycolysis pathway. In fact, a first model was provided by González-Chávez et al. (2019) and could predict the final flux quite well when TryR and TXN activities were varied. However, it overestimated the flux when TryR activity was varied and underestimated it when TXNPx activity was varied. Therefore, we suggest adding a first adjustment term α(V f − V f 0 ) in order to increase TryR rate and a second adjustment term β(V f − V f 0 ) to decrease TXNPx rate. In these adjustment terms, α and β are defined numbers selected as the best for flux prediction from a tested range, V f is TryR (or TXNPx) maximum rate in the forward direction in the model and V f 0 TryR (or TXNPx) maximum rate in the forward direction used in the in vitro reconstitution. Also, as V f of TryR (or TXNPx) is equal to V f 0 when TXN's/TXNPx's (or TryR's/TXN's) activity is varied, we multiplied α (or β) by V f − V f 0 , so that the adjustment term would be zero when V f = V f 0 and the flux predictions are not modified in these cases mentioned above. Also, residual values are determined to evaluate how accurate the gray-box model is, and calculated as follows (1): where e is the residual, y is the observed value andŷ the corresponding predicted value. Data Augmentation For the datasets with <100 data, a process called data augmentation is performed. It consists of using models that accurately predict the experimental data to generate a new bigger dataset. Two different gray-box models are used in this study for the lower part of glycolysis pathway, retrieved from a recent study (Lo-Thong et al., 2020), and for the peroxide detoxification pathway (built for the present work). The gray-box models built on COPASI is set up to predict the variation of the final product concentration over the first hour for a given set of enzyme activities; then the COPASI outputs are processed to obtain the final flux of the studied metabolic pathway. Also, the overall process from the one-hour simulation for each enzymatic balances to the determination of the final flux is then automatized and applied to a range of enzymatic balances detailed in the previous subparts (Lower Part of Glycolysis Datasets and Peroxide Detoxification Datasets). (González-Chávez et al., 2015). Enzyme Kinetic equations a A, B and K mA , K mB are respectively the concentrations and Km of the substrates NADPH and TS 2 ; P, Q and K mP , K mQ are the concentrations and Km of the products NADP + and T(SH) 2 ; α(V f − V f0 ) is the adjustment term with α, a defined number, V f0 , TryR maximum rate in the forward direction used in the in vitro reconstitution and V f is TryR maximum rate in the forward direction in the model. b A, B and K mA , K mB are respectively the concentrations and Km of the substrates T(SH) 2 and TXNox; P, Q and K mP , K mQ are the concentrations and Km of the products TS 2 and TXN red . c β(V f − V f0 ) is the adjustment term with β, a defined number, V f0 , TXNPx maximum rate in the forward direction used in the in vitro reconstitution and V f is TXNPx maximum rate in the forward direction in the model. Dataset Analysis and Non-linearity Assessment A brief analysis of the datasets is performed, including an examination of data distribution and the calculation of linear correlations between the input and output variables. The determination of linear correlation between the inputs and output variables allows the assessment of the non-linearity for each studied metabolic pathway. As a rule of thumb, we consider that the non-linearity is high when one or more inputs has a linear correlation lower than 0.6. The lower the linear correlation, the greater the degree of non-linearity of the pathway. Machine Learning Models Building and Selection To model the metabolic pathway, different machine learning models are developed on RStudio (Version 1.2.5001), with the help of Classification And Regression Training (caret, Version 6.0-86) (Kuhn, 2020). The datasets are split into 80/20 for the training and test sets, and a k-fold cross-validation (with k = 10 for Dataset 1, 2 and k = 3 for Dataset 3) is performed on the models with the training set. After this, the best models are selected based on: The root-mean-square error (RMSE): with Y i andŶ i being respectively the observed and predicted values, n being the total number of values and i = 1, 2. . . n; the coefficient of determination (R 2 ): with Y i andŶ i respectively the observed and predicted values, n being the total number of values and i = 1, 2. . . n. Also, a calculator was used for modeling the metabolic pathways, which has the following characteristics: cluster 2x Intel Xeon E5-2630v4 Broadwell-EP @ 2.20GHz 10 cores, 8x 16GB of RAM, 2400MHz, DDR4, ECC. RESULTS As previously mentioned, ML models could have different applications in biology, including the identification of biomarkers, i.e., a valuable, quantitative component (metabolites, proteins, enzymes. . . ), within a metabolic pathway for health purposes (diseases diagnosis, treatment) or the optimization of a valuable production pathway. Therefore, we have targeted three different datasets based on these two applications. The first one concerns the lower part of glycolysis in Entamoeba histolytica ( Figure 3A) and contains a set of enzyme activities for which the final flux has been measured (Moreno-Sánchez et al., 2008). The second pathway is the tryparedoxin-dependent hydroperoxide detoxification pathway in Trypanosoma cruzi ( Figure 3B), which provides the same type of data as in the previous dataset (González-Chávez et al., 2015). It is important to consider how essential these two previous pathways are, as they play a significant role in the survival of these parasites. Given the small size of the experimental dataset, we use two gray-box models: one developed recently (Lo-Thong et al., 2020) and the other developed in this study, to generate a larger dataset for these two pathways (Datasets 1 and 2) before building the ML models (Figure 2). The last metabolic pathway modeled here is the penicillin fermentation process in Penicillium chrysogenum ( Figure 3C). This dataset did not need to be enlarged (Dataset 3), and we used it to build different ML models (Figure 2). Example 1: The Lower Part of Entamoeba histolytica Glycolysis The Gray-Box Model Allows the Building of Huge Datasets Since the amount of experimental data is limited, the first step here is to build a robust model to generate more data. As explained in the Methods section, the gray-box model developed in a previous work contains all kinetic parameters and kinetic equations of PGAM, ENO and PPDK (Lo-Thong et al., 2020). In order to improve the flux prediction, the first Figures 1A-C). The calculation of residuals shows a defined pattern that is the same for PGAM and ENO. It reveals a general trend of the model to underestimate the flux for low enzyme activity values, and overestimate it for high enzyme activity values (Supplementary Figures 1D,E). Concerning PPDK, the graybox model tends instead to underestimate the final flux when the enzyme activity is varied, with an exception for the last point (at 232.13 mU), which is overestimated (Supplementary Figure 1F). The model is quite accurate to predict the pathway flux and presents low residuals between−3.4-4.7 nmol·min −1 . The next step of this work consists of using the in-silico model for generating larger datasets, a process we call data augmentation. The first new dataset contains 2,000 enzyme balances evolving around the experimental ones (see Supplementary Table 6 In fact, the predicted fluxes count with the highest representation are within the experimental data of the reconstituted pathway (Moreno-Sánchez et al., 2008) and in vivo pathway fluxes in live parasites (Pineda et al., 2015). In order to compare the models, a second dataset (Dataset 1) is generated and includes 68,950 data for which all enzyme activity is varied between 0 and 1,000 mU (see Supplementary Tables 2, 7 We then plotted the final flux in function of the enzyme activity for the largest dataset (Supplementary Table 7) and obtained the same type of curve as we did previously (Supplementary Figures 2, 5). Indeed, variations of PGAM activity have a great impact on the final flux, while those of ENO and PPDK have a lesser impact on the pathway flux (Supplementary Figure 5). It should also be noticed that the experimental fluxes are in the lower part of the predicted flux values. The insets show a gap between the experimental flux values and the dataset flux values; this difference is due to the intervals between two values, used in the two cases, with the interval being smaller for the experimental dots (7-85 mU) than for the predicted data (25 mU). Following this initial analysis of the data, we assessed the correlation between the various variables. The table of correlation shows that the enzymes and the final flux are correlated to varying degrees, with the highest correlation coefficient for PGAM, followed by ENO, and the lowest coefficient for PPDK ( Table 2). These linear correlation coefficients provide insight into the degree of non-linearity of this metabolic pathway. Here, only PPDK has a linear coefficient lower than 0.6 indicating that the lower part of glycolysis has a large degree of non-linearity. Also, even if the mean value of the correlations is above 0.5 (Table 2), we observe a weak linear correlation for many ranges of enzyme activity (Figure 4) when one of the enzymes is varied over the three, for example for PPDK when PGAM varies between 0 and 625 mU and ENO between 0 and 1,000 mU ( Figure 4C). These results indicate significant non-linearity in the metabolic pathway, particularly for PPDK and ENO. In addition, these results lead to the same conclusions as those from flux control coefficient calculations (Lo-Thong et al., 2020): the enzyme exerting the greatest flux control is PGAM, followed by ENO, and PPDK has the weakest control of the pathway flux. Good quality augmented datasets having been generated; they are used to test different ML approaches in the following section. Non-linear Machine Learning Methods for Metabolic Pathway Modeling Outperform Rborist Based on the preceding data, we also investigate whether we can build a good predictive model by using linear and nonlinear ML methods. In the study cited previously, Artificial Neural Networks (ANN) were used to predict the flux (Lo- Thong et al., 2020). Here, only one ANN model is developed and proves to be one of the best models obtained ( Table 3 and Figure 5E). Among the designed models and for the first dataset (Supplementary Table 7), the random forest models stand out, with better flux prediction for the training set with the model built with Rborist package: cvRMSE = 0.883 nmol·min −1 and cvR 2 = 0.995, than the QRF model: cvRMSE = 0.931 nmol·min −1 and cvR 2 =0.994 (Supplementary Table 3, Supplementary Figures 6B,D). As for the test set, the QRF model outperforms the Rborist model, with RMSE = 0.076 nmol·min −1 and R 2 = 1. Another good model, also nonlinear, is the XGBoost Linear method, with cvRMSE = 0.833 nmol·min −1 and cvR 2 = 0.995 (Supplementary Table 3, Supplementary Figure 6A). Moreover, the results obtained with Bayesian GLM, Lasso, Ridge, Spike-and-slab and the PLS model indicate that a linear model is not really adequate to describe this metabolic pathway. In fact, the PLS model gives the highest value for cvRMSE and the lowest value for cvR 2 (Supplementary Table 3); also, we can see that the flux predictions are not very good (Supplementary Figure 6M). For the second dataset (Supplementary Table 7), we obtained almost the same results: first with the Cubist model (cvRMSE = 0.215 nmol·min −1 and cvR 2 =1), then the two random forest models ( Table 3). This time, better results are obtained with the QRF model: cvRMSE = 0.572 nmol·min −1 and cvR 2 = 1, than with the Rborist model: cvRMSE = 0.647 nmol·min −1 and cvR 2 = 1 for the training set (Table 3 and Figures 5A-C). The XGBoost Linear method also gives good flux predictions, with cvRMSE = 0.489 nmol·min −1 and cvR 2 = 1 (Table 3 and Figure 5D). If the SVM Radial method gives almost good results (Table 3 and Figure 5F), it is no longer the case for the last two non-linear models (SVM Poly and bagEarth GCV) which present worse results in predicting flux, with much higher RMSE (Table 3 and Figures 5G,H). For the same reasons stated above, all linear models show poor results in predicting flux starting from enzyme activities, and are therefore not adequate to model the lower part of glycolysis here (Figures 5I-M). Overall and for Dataset 1, the Cubist model has the best generalization capability, with a lower RMSE = 0.154 nmol·min −1 and a higher R 2 = 1 for the test set (Table 3). These results show that the non-linear models, such as random forests, Cubist and XGBoost Linear, are able to indicate the final flux of the pathway by using the predicted data. We look at modeling the second metabolic pathway, which can also be used for drug design purposes. In the gray-box model developed here around this second dataset, the first and third enzymes employ a modified kinetic equation including two different adjustment terms: α =23 and ß = 8 ( Table 1). The determination of these parameters is detailed in the Methods section. We obtained a relatively good model of flux prediction (R 2 ≈ 0.67 and RMSE = 4.668 nmol·min −1 ) when enzyme activities are varied (Supplementary Figure 7). However, the model still overestimates the flux when TryR activity is varied and when TXNPx activity is higher than 698.35 mU. The new dataset contains 1,671 enzyme balances evolving around the experimental ones (Dataset 2, see Supplementary Table 8). The predicted final fluxes vary between 0 and 11.46 nmol·min −1 ; the dataset's distribution is shown in Supplementary Figure 8, Supplementary Table 4. It is important to note that we could not go below 16.1 mU and 57.6 mU for TryR and TXNPx activity. The reason is that the gray-box model is not able to predict the flux below these values. Also, an analysis of the correlation between the different variable shows that TXN has the highest correlation coefficient, followed by TXNPx and lastly TryR ( Figure 6A). Here, these linear correlation coefficients point out the predominantly non-linear character of this metabolic pathway, when TryR orTXNPx activities is varied. The nonlinear aspect of the peroxide detoxification pathway is certainly not to be negligeable, since the coefficient average, when all enzyme activities are varied, is lower than 0.6. These results support those obtained by González-Chávez et al. (2015, 2019 which demonstrate that TXN and TXNPx exert the greatest control on the pathway's flux, while TryR exerts very little control on the flux. The augmented dataset is now used to test different ML approaches, as described in the following section. Non-linear Machine Learning Methods Are Efficient for Flux Prediction We built different ML models and evaluated their performance. Of the thirteen models built, only five predict well the flux for both training and test sets: the random forest (QRF and Rborist), XGBoost Linear, Cubist and ANN (Figures 6B-D, 7A-E). These models have a cvRMSE range of 0.128-0.186 nmol·min −1 and cvR 2 of 0.996-0.998 for the training set, and RMSE range of 0.022-0.098 nmol·min −1 and R 2 of 0.999-1 for the test set ( Table 3). The following three models (SVM Radial, SVM Poly and bagEarth GCV) predict moderately well the flux of peroxide detoxification (Figures 6B-D, 7F-H), with cvRMSE between 0.349 and 0.956 nmol·min −1 , and cvR 2 between 0.916 and 0.989 (Table 3). With the test set, their performance is slightly lower, with RMSE between 0.233 and 0.964 nmol·min −1 and R 2 between 0.914 and 0.996 (Table 3). In contrast, the last five models can hardly predict the flux from enzymatic activities for both training and test sets, particularly for flux below 7.5 nmol·min −1 which is within the physiological and experimentally determined value (Figures 6E-G, 7I-M). These models present higher RSME and lower R 2 values for the training set (cvRMSE range of 1.44-1.581 nmol·min −1 and cvR 2 range of 0.765-0.805) and test set (RMSE between 1.379 and 1.55 nmol·min −1 and R 2 range of 0.777-0.823), confirming their poorer performance not only in terms of learning but also in terms of generalization, in making robust predictions on new data ( Table 3). We also observe that models Bayesian GLM, Spike-and-slab and Ridge give comparable results ( Table 3 and Figures 7I-K). These results, together with those in example 1, allow us to confirm that non-linear models are more appropriate to predict the flux of a metabolic pathway than linear ones. Moreover, it should be noted that our gray-box models, built with COPASI, are non-linear models and that the data of Datasets 1 and 2 are mostly obtained with these non-linear kinetic models. To ensure that the preceding results are not influenced by the kinetic model used to generate the data, we use a new raw dataset from experimental records of a bioreactor. Example 3: The Industrial-Scale Penicillin Fermentation Process of Penicillium chrysogenum In addition, another type of metabolic pathway we can examine is the production pathways; their modeling would allow the development of an optimized overall process. In fact, another study revealed that ML methods can accelerate the optimization of chemical synthesis (Hein, 2021). As stated before, we do not need to enlarge this dataset, which is composed of records of the various parameters of an industrial-scale penicillin fermentation process. The use of this dataset made of only experimental data will ensure the reliability or not of the ML models for metabolic pathway prediction. It is important to consider that the inputs of our models are no longer the enzymatic activities, but different variables such as: batch time, oil flow, aeration rate, vessel volume and weight, carbon evolution rate and CO 2 percentage in off-gas. A slight variation of CO 2 in off-gas is recorded (Supplementary Table 9); this can be explained by the implementation of a system, by the operators, allowing corrective measures to be taken when the CO 2 level is too high, thus avoiding the detrimental effect of an accumulation of CO 2 on the growth of Penicillium chrysogenum and the production of penicillin. As the percentage of CO 2 in off-gas is maintained at a certain level, it is not surprising that the carbon evolution rate does not vary much either and presents a low standard deviation (Supplementary Table 5). Also, the output we are interested in is not the pathway flux, but the final concentration of penicillin ( Figure 3C). As regards the correlation coefficient between the variables, we note that it is generally high between the parameters and the final penicillin concentration ( Table 4); this correlation can be positive (e.g., time) or negative (e.g., oil flow). These correlation coefficients reveal the linear nature of the fermentation process studied in Dataset 3. Table 3 for the statistical measurements of each model. Frontiers in Artificial Intelligence \| www.frontiersin.org Table 3 for the statistical measurements of each model. Non-linear Machine Learning Methods Predict the Fermentation Process Better Than Linear Methods The results of penicillin concentration predictions reveal that Random Forest models effectively predict experimental concentrations, with cvRMSE = 0.814/0.877 g·L −1 and cvR 2 = 0.993/0.992 (QRF/Rborist) for the training set and RMSE = 0.134/0.319 g·L −1 and R 2 = 1/0.999 (QRF/Rborist) for the test set (Table 3 and Figures 8A,B). We can then separate the rest of the models into two groups, based on their performance on the test set. The first one, which predicts the penicillin concentration fairly well, has RMSE between 1.097 and 2.418 g·L −1 , and R 2 between 0.941 and 0.988 (Table 3 and Figures 8C-H). By contrast, we found that the predictions of the second group are considerably worse, with many more outliers (Figures 8I-M), and with RSME higher than 3.5 g·L −1 and R 2 lower than 0.9 for the test set ( Table 3). As noted in the previous dataset, we also found many models that give the same results, namely: Bayesian GLM, Spike-and-slab, Ridge and Lasso (Table 3 and Figures 8I-L). Here also, Lasso and PLS were the worst in terms of predictions. Interestingly, compared to the preceding results, Dataset 3 gives the best results for linear models (lowest RMSE and highest R 2 values for the training and test sets); this could be explained by the largely linear nature of the penicillin concentration used with respect to the parameters used. These results support the previous ones and confirm that non-linear models surpass linear models for the prediction of penicillin concentration through the fermentation process. Performance Comparison of All Models After showing that non-linear ML methods are more suitable for modeling metabolic pathways, we performed a comparison of the performance of all models. At first glance, the plots further confirm the preceding results and display higher RMSE values and lower R 2 values for the linear models compared to nonlinear models (Figure 9). In addition, regardless of the number and/or type of data, we observe that Spike-and-slab, Ridge, Lasso and Bayesian GLM models give almost the same results (Figure 9 and Table 3). Also, it appears that some non-linear models work less well with large datasets; this is the case for ANN, bagEarth GCV, SVM Poly and SVM Radial (Figure 9). Moreover, it appears that random forest models (QRF and Rborist) are the best suited for metabolic pathway modeling, as they give the best results in term of RMSE and R 2 whatever dataset was used. Furthermore, we can evaluate the impact of the degree of non-linearity of the pathway on the predictions. Indeed, the pathway that has a high non-linear structure (Dataset 1) gives worse results for linear models than the pathway that presents a less non-linear structure (Dataset 3), which also gives good results with non-linear models ( Figure 9A and Table 3 Table 3 for the statistical measurements of each model. Table 3 for the statistical measurements of each model. R 2 = 0.716, than Dataset 3, which performs well with the same model, with RMSE = 3.579 nmol·min −1 and R 2 = 0.87. Besides, with a view to applying these methods at an industrial level, we perform a comparison of model error prediction and time of processing among the different datasets (Figure 10). The results confirm the previous findings, where random forest models have the best performance for metabolic pathway flux prediction. We noted that Rborist model presents a better RMSE -time of processing ratio than QRF model. However, even if QRF models have a processing time higher than 1h, we obtain an RMSE gain of about 96 %, when comparing it with PLS model, which could be of considerable significance for the industrial level. In view of the considerable gain of using this method compared to a linear one, non-linear methods could be more beneficial at the industrial level, where a gain of 1% is colossal. Spike-and-slab, Ridge, Lasso and Bayesian GLM models result in comparable performance in terms of RMSE and time of processing. At least, these results show a better RMSE -time of processing ratio for non-linear methods than for linear ones. We did not add the ANN models in the results, as they were not performed using parallelization process compared to the other methods. Furthermore, we assess the impact of the amount of training data on ML model performance to have a desired level of performance (Supplementary Figure 9). We observe that the results are roughly the same for the datasets when they are predicted with linear models (Supplementary Figures 9B,D,E,H,I), thus the amount of data required to obtain a strong linear model can be higher than 80,000 data, as long as the studied pathway does not have a high degree of non-linearity. When it comes to non-linear models, we find that using a dataset smaller than 40,000 data is sufficient to obtain a good ML model (Supplementary Figures 9A,C,F,G,J-L). Using a dataset higher than 40,000 data leads to non-linear models that are efficient only in case of random forests (QRF and Rborist), Cubist and XGboost Linear methods, for which RMSE is low. We could also consider making an ablation of our datasets to examine the impact of amount of training set data on the ML model performance. Comparison and Applicability of Knowledge-Based and Data-Driven Approaches The first objective of this study is to determine what sort of data-driven model could better simulate the biological pathways studied. By using different datasets, we build several models with the enzyme balances or parameters collected from a bioreactor and reveal that Random Forests (QRF and Rborist), Cubist and XGBoost Linear are three good methods to predict the final flux or concentration of a final product. This works is part of a larger study about the applicability of either a knowledge-based or a data-driven approach. Indeed, in other fields such as fault detection and diagnosis, a comparison of these two methods demonstrates that they both have comparable performance and can be used (Alzghoul et al., 2014;Yang and Rizzoni, 2016). In biological system modeling, as is the case here, we demonstrated that in instances where little knowledge is available and difficult to obtain on a large scale basis (e.g., kinetic parameters k cat and K m of an enzyme, pathway fluxes), or when complex feedback regulation mechanisms take place, a data-driven method can be a good alternative for modeling a metabolic pathway, as many authors have shown before (Ramachandran et al., 2011;Hou et al., 2016). By comparison, the knowledge-based method can be laborious and long, due to data mining from the literature or wet laboratory experiments, whereas there is ease and speed of building models with the data-driven method (Kadarmideen, 2016). Another criterion that we considered was the degree of nonlinearity of the pathway. As mentioned above, it is generally admit that metabolic systems have an inherent non-linear behavior (Koza et al., 2001;Song and Ramkrishna, 2013;Yasemi and Jolicoeur, 2021). However, there is no formal demonstration of the non-linear structure of metabolic pathways. According to Song and Ramkrishna, this non-linear behavior would be due to: (i) the non-linearity of the chemical reactions forming the pathway and (ii) the regulatory processes that added nonlinearity to the system (Song and Ramkrishna, 2013). Also, it is expected that pathway fluxes are non-linear, because they are controlled by enzymes and the activities of metabolic enzymes are saturable by their ligands. Besides, when the fluxes are measured in intact cells, they give a non-linear behavior and flux variation appears as hyperbolic or even sigmoidal. If the measured fluxes appear linear, it might be because the saturation point is not reached. Furthermore, according to the Metabolic Control Analysis, the fluxes are hyperbolic or non-linear because always exist one or two flux-controlling steps which utlimately determine the pathway flux (Fell, 1992). The determination of linear correlation coefficients of the different variables of the datasets gives us insights into the degree of non-linearity of the studied metabolic pathways and provides a method to evaluate the non-linearity of metabolic pathways. We found that all metabolic pathways studied here have a notable nonlinear structure, with Dataset 1 having the highest degree of non-linearity, then Dataset 2 and lastly Dataset 3. These results generally comfort the main hypothesis that metabolic pathways are predominantly non-linear. The determination of the degree of non-linearity is therefore important for selecting and applying of a ML technique when modeling a metabolic pathway. Moreover, the suitability of using either method relies on the quantity and quality of the knowledge or the data. Here, to illustrate this point, we simulate two datasets: the first one consisting of an exploration of the experimental data (2,000 data) and the second one composed of enzyme activities from 0 to 1,000 mU (68,950 data). The largest one gives better predictions for the three best models (Random forests, Cubist and XGBoost Linear) than the other dataset, and shows us the importance of having a large dataset before using machine learning methods. In fact, the size of the training set has been shown to be a major driving factor of prediction accuracy (Somarathna et al., 2017). However, we used two datasets made up of a mix of experimental and predicted data to build the models, and even if predicted from a good quality model, they remain mostly predicted data and are not comparable to a fully experimental dataset, which is also difficult to obtain. Thus, it would be worth considering methods using only experimental data, when sufficient data are available to build the models. Interestingly, a data-driven approach is often used to discover biological pathways or unravel pathways that are not well understood. Thus, combined with the knowledge-based approach, this can quickly make clear the complexity of biological systems modeling. Another possibility would be to test ML models on experimental data from E. coli or yeast, which can present a larger degree of non-linearity and are easily found in the literature. This issue will be addressed in our next study. Surprisingly, model performance was weaker for the largest dataset from the bioreactor records than for the smaller datasets. The reason for this result may lie in the choice of input variables. Several studies have highlighted the need for variable selection in order to have better predictions (Camacho et al., 2018;Awan et al., 2019;Genuer et al.). Indeed, variable selection allows the use of the most informative variables to predict the output variable(s) and reduce the time of computing. Unlike the knowledge-based model, a diversity of variables for data-based models does not always mean better performance. This is one of the limitations of our study, since only one combination of input variables was tested during the work. It would be interesting for a future study to compare, for the same dataset, models using different sets of input variables, and to analyze their impact on model effectiveness. Interpretability of Machine-Learning Approaches Another major issue facing users of machine learning approaches is the interpretability of these models. Even if, at this time, we do not have a common general definition of this term, many researchers, such as Schmidt et al. (2019), define a model's interpretability based on two aspects: (a) intrinsic interpretability (or transparency): the ability to understand the inner mechanism of the model in the context of the study (e.g., identification of variables most involved in the predictions), and (b) post hoc interpretability: the ability to extract new information from the model or provide new insights into the relationships discovered during the process (e.g., the effect of a variable on another one) (Murdoch et al., 2019;Schmidt et al., 2019;Pintelas et al., 2020). Although some ML methods, such as decision trees or linear regression models, are easily interpretable; this is not the case for most of the models developed here (e.g., XGBoost Linear, bagging MARS, ANN). Nevertheless, using the variables that are directly related to the variable to be predicted, as we do here, allows us to gain some understanding of how the model works and the types of relationships that are revealed, enabling us to rely on the models. Furthermore, while we identified Random forests as one of the best methods for predicting final flux or product concentration, Pintelas et al. (2020) classifies it as a model that is hard to interpret. Therefore, it would be interesting to compute variable importance or to apply different techniques to explain the model in order to increase its interpretability (Zhou et al., 2019;Azodi et al., 2020). Besides, knowing that models based on decision trees are among the simplest to interpret, we support the idea of Schmidt et al. that RF models are more accessible than others from an interpretability point of view (Schmidt et al., 2019). An alternate solution would be to develop simpler models, but this would certainly reduce their overall performance. Moreover, one of the key factors in the interpretability of the models is linked to the equations used. In fact, compared to knowledge-based models that use well-defined equations with a biological significance, ML models are governed by other equations, which sometimes are "outside our understanding" as Schmidt et al. (2019) observed in their study of the applications of ML in solid-state materials science. This raises a real problem of confidence in the prediction results obtained with such methods. As these authors point out, the fact that these models were not based on physical principles in their studies, or on biological principles in ours, could result in wrong predictions in completely unexpected cases, while providing great results overall. And in the present case where the models are used in the context of biomarker identification or optimization of an industrial bioreactor, we cannot risk obtaining such results from our models in these specific situations. Far from hindering us in the use of ML models, awareness of these problems allows us to formulate several recommendations for future research. These include the combination of interpretable models, e.g., knowledge-based kinetic models with ML models, e.g., random forests models; the prediction of a new set of experimental data with unexpected values. In this latter instance, this would require experimentally testing a range of "extreme" data that would be found in the parasites studied, or recording the bioreactor data even during failures of the penicillin production. Strengths and Weaknesses of the Modeling Methods After analyzing the interpretability of the different modeling methods, it is worthwhile to note some advantages and disadvantages of their use in flux and concentration prediction. One of the best methods in our case is the random forest (QRF and Rborist). Many studies report the use of random forest in the biological field for the prediction of: protein interaction (Qi et al., 2006), drug response based on protein markers (Ma et al., 2006) and in vitro drug sensitivity (Riddick et al., 2011). Also, Riddick et al. used SVM and random forest to predict the flux of N 2 O emissions, and found that random forest achieves the best performances among the built models (Villa-Vialaneix et al., 2010). They highlighted that these models offered the advantage of having a low computational cost, compared to the SVM method. However, in our case, we notice that random forest is the least accurate predictability model compared to SVM methods, with the highest computation time for almost all datasets. Moreover, among the random forest packages developed on R, Rborist is quite a recent implementation, designed for multicore hardware, which minimizes data movement within memory to increase the performance and decrease the processing time (Wright and Ziegler, 2017). Surprisingly, here, Rborist package is the one that has the longest time of computation and is more efficient on big datasets compared to other methods. It would be of interest to create variant models combining the random forest method and other methods, as in previous studies (Chen et al., 2018;Zampieri et al., 2019). An existing variant of random forests is the quantile regression forest (QRF) method, which has the capability of establishing prediction intervals that cover uncertainties, useful in the prediction of possible new data (Meinshausen, 2006). Francke et al. demonstrated in their work that this method had the advantage of calculating uncertainties associated with the predicted sediment yields, through the calculation of confidence intervals (Francke et al., 2008). But they also stated that the model predictions will always be within the range of observations, which prevents implausible values but inhibits prediction outside the range of values learned from the training set. We saw here that, overall, QRF models have a good generalization capability; additional prediction of new experimental data, with data separated by a larger stepsize (>25), would be beneficial to confirm or invalidate this capability. This could be useful for the study of metabolic pathways in extremotolerant organisms. This leads us to note one of the advantages not only of the QRF method but also of other ensemble learning methods, such as XGBoost Linear: prediction from high-dimensional data. Indeed, these models are among the best we have, with any starting dataset we have, from the simplest to the most complex with several types of variables. Remarkably, compared to other models, XGBoost Linear is better ranked for small datasets. This is confirmed by the work of Yang et al. (2010) which propose that ensemble methods have the advantage of reducing the potential for overfitting in small sample size problem. Another strength of XGBoost Linear compared to its peers is the combination of high accuracy and a short time of processing. However, despite the great accuracy of these models, they are often more complex and less interpretable, and present a higher computational intensity. Moreover, Cubist, a model based on modified regression tree theory, has the advantage of analyzing big data with high speed (Xu et al., 2018). This was confirmed by our results, which show that Cubist is one of our best models (e.g., for Dataset 1, Cubist: 2.49 min and QRF: 1.76 hr). However, we noted that the performance was better for the small datasets than for the bigger one. Another advantage that Das et al. noticed is the fact that the Cubist model is easy to interpret and is a suitable method for beginners (Zhou et al., 2019;Das et al., 2020). The PLS method turned out not to be appropriate here to model these pathways and predict the final flux starting from enzyme activities, or the final product concentration starting from parameters of a bioreactor. This may be due to the inherent limitation of the PLS method to capture the non-linearities of the metabolic pathways. However, it performs better when we have a smaller dataset, as it has also been noted in a previous study on gluconeogenic flux prediction (Antoniewicz et al., 2006). But these results contradict those obtained with the PLS model for the prediction of limonene and isopentenol synthesis. In fact, in this work, results showed that the model performed well when the dataset was larger (lower RMSE, better predictions) (Costello and Martin, 2018). Also, one big advantage of the PLS technique remains that it has the shortest calculation time for modeling. It is relevant to observe that the model implementation will differ depending on varying levels of data. In fact, a ML model will be more difficult to implement, if the available data is limited. In this case, a significant additional time is required. Among the various studied models, the difficulty to implement the model could also be based on the higher number of parameters to adjust during the training time. Our findings generally support the idea that non-linear models are more suitable than linear ones for modeling metabolic pathways. Furthermore, it would be interesting to apply these ML models on genome-scale metabolic networks for which the literature abounds in data. Recently, hybrid models coupling a genome-scale model and ML model have been found to be effective for different purposes such as the prediction of individual amino acid concentration in culture medium (Schinn et al., 2021) and identification of prognostic metabolic biomarkers in cancer studies (Lewis and Kemp, 2021). One of the benefits that ML models could bring is the integration of multi-omics data as genomic, transcriptomic, metabolomic and proteomic data. This topic will be addressed in an upcoming study. As far as we know, genome-scale models have a predominant place in the field of metabolic networks for the identification of key-molecules in the metabolism. This study allows us to consider the machine learning methods as performant models to predict metabolic pathways. Indeed, their ability to take over large datasets makes them applicable techniques to efficiently predict larger metabolic pathways (e.g., E. coli). While flux balance analysis (FBA) based methods, as used in the genomescale models, need information about the pathway in a given condition as they are hypothesis-driven, machine learning models could predict the metabolic pathways without needing to clearly understand the underlying biological mechanisms of the pathways. Also, constraint-based model (e.g., FBA) are not able to predict metabolite concentrations, while the machine learning methods can consider these predictions. We can thus envisage a FIGURE 11 \| Decision-making support for the construction of metabolic pathway models using machine learning methods. Frontiers in Artificial Intelligence \| www.frontiersin.org hybrid method using both machine learning and FBA methods for metabolic pathway modeling (Zampieri et al., 2019). Decision-Making Support for Pathway Modeling Given the many different methods that exist and continue to emerge, one can struggle with the choice of a model to build from a dataset. Faced with this decision, we can choose to build simple models or to use models being used in the same field of study and giving good results (Camacho et al., 2018;Cifuentes et al., 2020). In view of this, it would be useful to review and define some basic rules for building a decision-making support for future studies on modeling metabolic pathways. The first feature to consider is the quality of the biological dataset (Figure 11). Do we have an initial dataset of good quality? Data quality can highly impact the model predictions. If the model is not of good quality, it would be better to build a new dataset and generate good quality experimental data. When the dataset is of good quality but small in size, it is useful to do data augmentation, as we did in this study; if this is not possible, we can use an ensemble model to build the metabolic pathway, since such models can deal with small datasets. Another useful criterion we can investigate is the number of variables. If the dataset presents a high number of variables, we can consider doing variable selection before building the model, or we have the option of building the model by using ensemble modeling that gives good accuracy with several input variables. Also, one key factor is the non-linearity of the studied metabolic pathway; do we have a non-linear or a linear process? If our pathway is linear, we can design a battery of linear models which will give a high performance. But if our study involves a pathway that is non-linear, then it is preferable to use a non-linear model. After building our model, an evaluation of its accuracy is necessary to validate it. In case the performance of the model is not suitable, we can plan to refine it, for example by tuning the hyperparameters (Chicco, 2017), or simply to replace it and build a new one. Non-linear machine learning methods enable us to model metabolic pathways by identifying key-molecules, which are important for the drug-design process, improving disease diagnosis (cancer, viral/parasitic/bacterial infections, neurodegenerative diseases) by highlighting the differences between healthy and pathological situations, or even optimizing industrial production processes. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. The custom codes for the data analysis used in this study are available from the corresponding author in the Github repository: https://github.com/ophelielt/Lo-Thong_et_al._Non-linearity_ of_metabolic_pathways_influences_the_choice_of_ML.git.	2022-06-10T14:04:54.774Z	2022-06-10T00:00:00.000	{ "year": 2022, "sha1": "0aeaadd95564a2b2d7e6243daf034e544f6256aa", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Frontier", "pdf_hash": "0aeaadd95564a2b2d7e6243daf034e544f6256aa", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Computer Science" ] }
257199980	pes2o/s2orc	v3-fos-license	Clinical description and treatment outcomes of Paederus dermatitis in Phuentsholing, Bhutan in 2021: A cross‐sectional study Abstract An increasing number of beetle population and outbreaks of irritant contact dermatitis are reported from newer geographic locations. Bhutan is one such country that witnessed an outbreak of Paederus dermatitis (PD) in Phuentsholing sub‐district in 2021. This study describes the clinical symptoms, skin lesions and treatment outcomes of PD in Bhutan. This was a descriptive cross‐sectional study of an outbreak of Paederus‐related contact dermatitis in Phuentsholing, Bhutan. Clinical symptoms, skin lesions, duration of illness, recovery time and response to treatment were recorded. Of the 81 patients with PD, the males constituted 54% (44) and the mean age was 22 years (range: 1–51 years). The commonly affected groups were those aged 11–20 years (40.7%) and school or college students (50.6%). The common symptoms were pain, itching, redness, tenderness and blister formation. The lesions were erythematovesicular (70%), linear (54.3%) and kissing lesions (28.4%). All patients received some form of topical or oral steroid therapy and recovery was 100%. The mean duration from the onset till the recovery of the skin rash was 13 days (SD ± 8.3 days). PD outbreak is a self‐limiting form of contact dermatitis. This is the first report of PD in the sub‐Himalayan region and may be linked to climate change. There is a need for active surveillance and monitoring of such emerging weather/climate‐related agents for appropriate health system response in disease prevention and treatment. \| BACKGROUND The health of human populations is sensitive to shifts in weather patterns and a shifting pattern of disease agents which, in part, are linked to ecological disruptions and climate change. 1 One such example is the spread of beetles into warm and humid places in the tropics leading to disease outbreaks in newer locations. Beetles constitute an order with the highest number of species known and account for roughly 25% of approximately 1.5 million described species. 2 Of these, about 622 species are the Paederus genus belonging to the Staphyllinadae family and Paederinae subfamily. It is distributed worldwide, except in Antarctica. The size of Paederus beetles is around one and a half times that of a mosquito-usually 7-10 mm long and 0.5 mm wide. They live in moist habitats and feed on small insects and plant debris. 3,4 contain the toxic agent known as paederin. Paederus dermatitis (PD) or Blister beetle dermatitis is an irritant contact dermatitis due to the accidental crush of insects belonging to the Paederus family on the skin. 5 This beetle does not bite or sting but accidental crushing or brushing against the skin causes the release of its haemolymph containing paederin that causes painful necrotic blisters. Paederus beetles have been associated with outbreaks of dermatitis in various countries across the world including Sri Lanka and India in the South Asia region. 6 PD is commonly caused by Paederus melampus in India and is reported in Odisha, West Bengal, Punjab, Rajasthan and Tamil Nadu. 7 This condition is commonly seen during or after rainy seasons. An unusually high number of dermatitis cases were observed in Phuentsholing, Bhutan during the peak summer of 2021 and 2022. The disease was characterized by a painful and itchy skin rash that developed overnight and caused significant discomfort and disruption of daily routine activities. The lesions were characterized by poor response to antibiotics and other conventional treatments such as antihistamines. An outbreak of such skin conditions at place and time during the rainy season suggested a possible aetiological agent related to climate and weather. In 2021 in Phuentsholing, Bhutan, the first group of victims with unusual dermatitis was recorded among those living in clusters settlements such as the boarding students at schools and colleges. The second group of victims was among workers at the food auction yard and the temporary shelter in Toorsa where hundreds of people resided while only a few cases originated from elsewhere in Phuentsholing. We have recorded more than 100 such cases; of whom the majority of them had exposure to Paederus beetles that are found inhabiting their locality. In 2022, another outbreak with more than 300 cases of similar skin conditions was reported 8 where more than 80% of the victims were exposed to the same beetles. This outbreak of unusual skin rash is new to the local inhabitants. We also studied and discussed the symptoms and signs of dermatitis, duration of illness, treatment and possible preventive methods and their association with exposure to beetles in their locality among patients who presented to the Phuentsholing General Hospital in 2021. \| Study design and setting This was a descriptive cross-sectional study of an outbreak of PD in Phuentsholing Bhutan, 2021. The study was conducted in the Phuentsholing municipality, Chhukha District where the outbreak occurred. Of the 11 sub-districts in Chhukha, Phuentsholing municipality is densely populated with >40% of the district population. The 2017 Population Housing Census of Bhutan recorded 27 658 individuals in Phuentsholing, excluding non-Bhutanese and tourists. 9 The municipality is the commercial hub of Bhutan with a diverse group of the transient population coming in from and travelling to all other districts. \| Case definition The case was defined in patients living in Phuentsholing between May and August 2021 with skin lesions, flat or linear with surrounding erythema and one of the following characteristics: lesions mimicking a burn with a crusty appearance or a vesicular lesion 3,7 with or without history or contact with a beetle. \| Data collection Based on the above case definition, data were collected from June to August 2021. After obtaining informed written consent, pictures of skin rash were taken and stored for use in this study. When patients sought medical attention at Phuentsholing hospital, a history of contact with the Paederus beetles was recorded. The majority of the patients did not know about the beetle until when shown the pictures of it, they agreed to have noticed or come in contact with the rove beetles. They were also informed to notify the investigators if they came across these beetles in their living places or surrounding. These pictures of the insects that we collected from the patients were sent to the Royal Entomological Society, United Kingdom via email that confirmed the genus Paederus, commonly known as rove beetles. A questionnaire was used to collect sociodemographic information (sex, age, among others), date and place of occurrence of dermatitis cases, What is already known about this topic? � It is known that Paederus beetles can cause irritant contact dermatitis. What does this study add? � This study highlights the occurrence of this condition in newer locations indicating a possible link to climate change. It also gives additional information on common clinical symptoms, signs and treatment of Paederus dermatitis. residence type and clinical manifestations related to the lesions (anatomical location, symptoms, risk factors). Kissing lesions are characterized by the rash appearing near flexures which are mirror images of each other due to contact on limb lesions and, erythematovesicular is characterized by the presence of tiny vesicles in the background of an erythematous rash. The linear rash is those appearing longitudinal or linear with surrounding erythema and, the classic lesions with erythematovesicular and necrotic patterns. The patients who presented with a skin rash that have a doubtful diagnosis or alternative diagnoses and those who did not consent to this study were excluded. We recorded all the medications that were prescribed during the clinical course of dermatitis. We did not undertake any intervention or introduce new treatments apart from the standard treatment prescribed by the local clinicians. A separate investigator was assigned for following up on the cases and measuring the outcome to prevent the investigators' evaluation from being influenced (intentionally or unintentionally) by their personal treatment preferences. Recovery was defined as the resolution of symptoms and a sign of healing. The follow-up for the majority of patients was done at the hospital during review after 5 days of the initial hospital visit and then after every 3 days until recovery. Home visit was done for a small number of patients who could not visit the hospital after contacting them through the telephone. \| Data entry and analysis Data entry was done in Google sheet and the data analysis was carried out in STATA 13.1. Continuous variables are expressed as mean � standard deviation and categorical data are expressed as a percentage. The descriptive analysis of the symptoms and signs, morphology and distribution of skin rash, the average duration of illness and recovery time are presented. \| RESULTS There were 102 suspected cases of beetle dermatitis where 12 did not consent to this study, and 9 accounted for the loss to follow-up. The final sample consisted of 81 patients; there were 44 males (54.0%) and the mean age was 22 years (range: 1-51 years, SD: �11.59 years). There were 80 (98.8%) Bhutanese and one was an Indian national. Those aged between 21 and 30 years (33, 40.7%) and school or college students (41, 50.6%) were the most affected age group. Notably, 61 (75.0%) patients had reported exposure to Paederus beetles before the onset of skin rash. These sociodemographic details of the patients with PD are shown in Table 1. The first case included in this study reported the onset of a rash on 3 June 2021. There was a surge in the number of cases in June and July as shown in Figure 1. The mean duration from the onset of symptoms till the recovery of skin rash was 13 days (SD � 8.3 days) (Figure 2). The symptoms reported by the patients were pain (73, 90.0%), itching (67, 82.7%), redness (48, 59.3%) and tenderness (45, 55.6%). Fever, pus and watery discharge were reported by less than 8.0% of the patients ( the background of an erythematous rash (Figure 3(a, b)). Linear rash accounted for 44 (54.3%) and appeared longitudinal or linear with surrounding erythema (Figure 3 (c-e)). The typical kissing lesions were noted in 23 (28.4%) patients (Figure 4(a, b)). Classic lesions with erythematovesicular and necrotic patterns were observed in 32 (39.5%) patients, annular in 38 (46.9%), papular in 26 (32.1%) and bullous lesions in 17 (21.0%) patients. The details of the description of the skin lesions are shown in Table 2. Common sites of lesions were the upper limb in 30 (37.0%), followed by a rash on at least two separate areas (upper limb, lower limb, body, face and head or neck) 28 (34.6%), and the head and neck 26 (32.1%) ( Table 2). The majority (47, 58.0%) were prescribed local application of hydrocortisone ointment, followed by triamcinolone ointment 26 (32.0%), and the combination of ointment and oral prednisolone ( Table 3). All the patients recruited for this study recovered with or without post-inflammatory pigmentation. \| Aetiological agent Exposure to rove or Paederus beetles before the onset of skin rash was reported by 61 (75.3%) patients and the majority of them did not know that contact with these beetles would cause a skin rash. The beetle was identified as Paederus species by the online entomology information support group (Royal Entomological Society, United Kingdom) ( Figure 5(a, b)). \| DISCUSSION The foothills of the eastern Himalayas across northeastern Indian states, Nepal and Bhutan have reported their hottest July summers, 2.5°C higher than the normal average temperatures of 25.6°C in 2021 and 2022. 10 This corresponded with the sighting of Lesions on at least two of the body parts listed above. of 8 -Paederus beetles and an outbreak of PD in this region. While numerous outbreaks of blister beetle dermatitis were reported worldwide, 4,11,12 some of them have been linked to an increased population of beetles. Beetles are most active during the rainy season, after unusually wet weather patterns, and during hot and humid weather. 13 Humid, wet conditions prevent desiccation of the beetle during flight and movement thereby encouraging higher levels of activity and wider geographic ranges of dispersal. 14,15 The outbreak of PD in the Phuentsholing subdistrict in the years 2021 and 2022 led to the spread of fear and panic among populations living in the subtropical belt of Bhutan. Many patients have had epidemiological linkage with the increased beetle population in their residences or surroundings. In addition, there were many other reports of Paederus beetles associated with outbreaks of dermatitis in many places in the northeastern Indian states of West Bengal, Sikkim and Assam and along many districts in Nepal. In recent years, many countries across the world including Central Africa, Uganda, Sierra Leone, Argentina, Brazil, France, Sri Lanka and India have reported PD. 6,16 The predominant symptoms reported were pain, itching, redness and tenderness in overexposed parts of the body, similar to findings reported in other countries. 17 The majority of lesions were linear erythematous and erythemato-vesicular with a 'burnt' or crusty appearance and a grey necrotic centre. 7,18 In mild cases, the erythema lasts for a couple of days, whereas in moderate cases, the erythema evolves into vesicles and bullae over a few days and is followed by desquamation. Scarring usually does not occur. The lesions are characteristically linear due to smearing the crushed insect across the skin. 19 Severe cases are reported with more extensive blistering and may demonstrate additional symptoms, such as fever, neuralgia, arthralgia and vomiting. 20,21 In the absence of a history of exposure to beetles, there are other differential diagnoses to consider including allergic dermatitis, arthropod bites, herpes zoster, herpes simplex, bullous impetigo, psoriasis and fungal infections. 22 The mean duration of skin rash from the onset until recovery was 13 days. Reports from India and Italy show almost similar duration to recovery with 12-15 days, 21,23 whereas another study from Sierra Leone suggested that healing time ranged from 14 to 28 days and lesions in all the patients healed with residual dyschromia. 20 Although the condition is self-limiting, the symptoms can be distressing for the patients and may affect their daily activities. Complications such as superadded infections are relatively rare. In our study, steroid treatment was prescribed with an understanding that PD is one form of contact dermatitis. 24 Though symptoms and signs can resolve spontaneously, wet compresses, antihistamines and lotions are recommended to alleviate symptoms. 13,25 PD, though self-limiting, is likely to cause a major public health concern given the background of unchecked warming of the climate and an increasing population of rove beetles. This outbreak led to fear and panic among residents in the sub-tropical region and increased healthcare utilization, which is a burden, especially in Bhutan where it is provided free of cost. This also led to temporary disruption of education and work where outbreaks were reported in schools and hostels in northeastern parts of India. 15 Such outbreaks necessitate appropriate preventive measures. It is advised to minimize artificial light sources where these beetles are attracted. During the contact of the beetle with skin, it is advised to avoid crushing the beetle against the skin (blow it off instead of crushing it), close the doors and windows at night during the season of an outbreak, washing the affected body area with soap and clean water, and the application of cold wet compresses. [24][25][26] So far, very little is known regarding the application of insecticide to control the Paederus beetle population. Therefore, the health authorities must place a robust surveillance system to forecast massive outbreaks. People living in tropical regions must be given education and awareness about blister beetle dermatitis and the need to adopt preventive measures. Climate change has particular consequences on the emergence or re-emergence of infectious diseases, and the impact can transcend beyond national boundaries. 27 This outbreak in the foothills of the eastern Himalayas is not an isolated phenomenon. The region is witnessing a northward shift in disease agents which, to make matter worse, are further spread by the increase in travelling within and outside the region. For example, travelling to places with increased beetle populations has resulted in contracting PD among the returning travellers in Italy from Zanzibar island, 22 and in Virginia among travellers who returned from Sierra Leone. 28 This calls for action from health systems across the world to develop mechanisms to detect and monitor the emergence of new disease agents and mechanisms to respond to such outbreaks and importantly to limit their spread and prevent future outbreaks. \| CONCLUSION PD is characterized by a painful, itchy and blistering rash that is predominantly erythematovesicular or linear and is due to exposure to Paederus or rove beetles. The outbreak in Phuentsholing occurred in the background of an unusually hot summer with temperatures higher than the historical records. Emerging infectious and vector-borne diseases will continue to challenge health systems as the climate remains unchecked. \| LIMITATIONS OF THE STUDY The size of the sample might not represent everyone affected by the outbreak of PD in Phuentsholing due to several reasons: some patients with milder symptoms might not have come to the hospital; some of the affected patients with shorter duration of skin rash or with relatively small lesions may have missed. In addition, there might have been selection bias as this was a hospital-based study.	2023-02-26T16:15:17.753Z	2023-02-23T00:00:00.000	{ "year": 2023, "sha1": "7475e60409d3dce5345ff00ea336b27843e90a3e", "oa_license": "CCBY", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/ski2.223", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "28af3d26a16b1cf1dab6cb9d554501a50a550dae", "s2fieldsofstudy": [ "Environmental Science", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
245791511	pes2o/s2orc	v3-fos-license	Effects of Planted Pollinator Habitat On Bee Health And Interspecic Pathogen Detection Shared resources can instigate pathogen spread due to large congregations of individuals in both natural and human modied resources. Of concern is the addition of pollinator habitat in conservation efforts as it attracts bees of various species, potentially instigating interspecic sharing of pathogens. Common pathogens have been documented across a wide variety of pollinators with shared oral resources instigating their spread in some, but not all, cases. To evaluate the impact of augmented pollinator habitat on bee health, we screened samples from eight bee species across three families against a panel of 9 pathogens using RT-qPCR. While we found that some habitat characteristics inuenced pathogen detection, we found no evidence that pathogen detection in one bee species was correlated with pathogen detection in another. These ndings suggest factors other than the habitat itself may be more critical in the dissemination of diseases among bee species. However, we found high levels of gut parasites in some bee species which may be of concern, such as Bombus pensylvanicus. Future monitoring of bee health at augmented pollinator habitat is needed to ensure pathogens do not build up over time to then spread within their communities. or watering Introduction Shared resources can pose health risks to organisms; this is true for naturally occurring resources such as mating grounds or watering holes 1 , but also for human modi ed resources such as supplemental wildlife feed 2 , hunter attractants 3 , and even bird feeders 4 . These shared resources can result in dense congregations of individuals 2 potentially causing them to act as "hotspots," leading to pathogen build up that can then spread throughout the environment 1 . Further, the interspeci c spread within these congregations can intensify if resources are scarce or limited 5 . In some cases, the resource itself can harbor pathogens 6 , increasing pathogen spread within populations 7 . However, interspeci c and intraspeci c pathogen spread depends on the host competency of the individual and the species for each pathogen in question 1 . Rather than acting as a hotspot, an incompetent host in a biologically diverse community at a shared resource could act to dilute the spread of a pathogen 8 . Pollinator population declines 9 have been repeatedly suggested to be driven by factors including agricultural intensi cation, nutritional stress, habitat alteration and fragmentation, and pathogens 10 , all of which can interact synergistically. Habitat loss in particular has arguably received the most attention in recent years. To combat this, augmenting habitat to support pollinators is becoming an increasingly popular conservation tool, especially in agricultural settings. While such habitat has been found to support pollinator abundance and diversity 11,12 , it is being implemented en masse with limited scienti c evidence for best practices 13 . Evaluating the impacts of this habitat on bee populations and bee health is critical to ensure that we are not exacerbating the exact pressures that are intended to be alleviated. Parallels can easily be drawn between human modi ed pollinator habitat to support bees and the shared resource examples of watering holes, supplemental wildlife feed, and bird feeders. There is a great wealth of previous literature exploring the potential for pathogen cross-over among bee species (Tables 1 and 2), particularly because similar pressures are of concern; for example, there is evidence that high abundance of common species can intensify pathogen occurance 14 . Additionally, certain ower species have been found to harbor pathogens 15 ; however, this could be counteracted or ameliorated with increased ower community diversity 16 . As differing results have been documented (Tables 1 and 2), it begs the question: will augmented pollinator habitat act to congregate individuals leading to hotspots of pathogen spread, or will these habitats attract a diverse pollinator community leading to pathogen dilution? And what role do the habitats themselves play in pathogen spread or dilution? To evaluate how pollinator habitat in uences pathogen dynamics within bee communities, we evaluated the pathogen prevalence in eight bee species from three families across 2 years. To do this, we sampled newly established pollinator habitat across North Carolina as part of the North Carolina Department of Agriculture and Consumer Services' (NCDA&CS) mandate titled "Protecting NC Pollinators." We investigated pathogen dynamics within Apis mellifera, within Bombus impatiens, between Apis mellifera and Bombus impatiens, and within six other bee species that have rarely if ever been quanti ed in this context. Sample Collection Samples were collected at established pollinator habitat at 12 sites across North Carolina in 2017 and 2018. Collection events occurred once a month for 4 months during peak bloom at each plot, for a total of four sampling events per locations per year (hereafter referred to as Spring, Early Summer, Late Summer, and Fall), utilizing hand nets for 30 ± 10 minutes along haphazard transects 17 . Focus was placed on the most commonly occurring species to ensure su cient replication. Each individual bee collected was placed into a separate 1.7 ml microcentrifuge tube and transported back to the lab on dry ice where they were then stored at -80 ˚C until further processing. At each station during each sampling event, the ower cover and ower diversity within the plot was documented and categorized into low, medium, or high. All samples were collected in accordance with the guidelines established by the NC Department of Agriculture and NC Fish and Wildlife Service. Table 1 A summary of previous screenings of bees for interspeci cally similar pathogen presence. A two-letter code is used for each country, with a two-letter state code also included for US projects. The total number of bee species tested is shown, followed by what common species are included. Similarly, the total number of pathogens tested is shown, followed by what common pathogens are included. This Seven of the pathogen targets are viruses and were selected because they are some of the most commonly occurring honey bee pathogens that have been shown to negatively affect honey bee health 19 . Although little is known about the true impact of most of these pathogens on native bee health and longevity 14 , interspeci c infection is possible and transmission has been previously suggested for several of these viruses [20][21][22] . The remaining pathogens are gut parasites; these pathogens were selected because they are commonly detected, known to negatively impact Individual 189 189 1 40 -32 8 1 27 26 22 8 Bombus impatiens Individual 201 201 -------68 1 2 Bombus pensylvanicus Halictus poeyi/ligatus Megachile xylocopoides Xylocopa micans Totals 502 733 1 40 0 32 8 1 27 108 29 bee health 23 , interspeci c transmission has been previously documented 24 , and infection of gut parasites has been linked to population losses in some cases 25 . Sample Preparation: Individual bee samples Samples of A. mellifera and B. impatiens were processed as individuals as the sample sizes of these species were the highest in our study. Samples of X. virginica and X. micans were processed as individuals due to their large body size. When processing these individual samples, we removed each specimen from cold storage and kept it on dry ice until crushed, following an adapted protocol from Leite et al. 2012 to ensure successful pulverization and the highest quality RNA due to sample brittleness. We used two Zirconium beads (3.0 mm) for A. mellifera and B. impatiens and three Zirconium beads for X. virginica and X. micans, placing each tube into the Ivoclar Silamat S6 in order to crush the sample. Once completely pulverized, we extracted RNA using the TRIzol® Reagent 27 and the Zymo Direct-zol TM RNA Miniprep Kit, following the Directzol protocol. After extraction, we assessed RNA quantity and quality using the Thermo Scienti c NanoDrop ND-1000 Spectrophotometer and diluted to 200 ng/microliter. All RNA was again stored at -80 ˚C until further analysis. Sample Preparation: Pooled bee samples Due to sample size and low pathogen detection (discussed below), we tested B. pensylvanicus, H. poeyi/ligatus, S. obliqua, and M. xylocopoides in pools of up to ve individuals (depending on how many were collected during each sampling event) using whole bodies (summarized in Table 3). To process pooled samples, we took up to ve bees per sampling event per station out of ultracold storage and immediately placed them into a sanitized ceramic mortar. Su cient liquid nitrogen was immediately added to cover all bee material and allowed to sublimate to ensure that the samples were brittle. We then immediately and quickly pulverized samples using a pestle. Once completely crushed, we lled a new 1.7 microliter tube approximately halfway with the powdered materials so as to leave enough space for the TRIzol® Reagent. As individuals of H. poeyi/ligatus are small, we combined these pooled samples into one new 1.7 tube and crushed them using the Zirconium bead protocol described above for individual samples. RNA extraction of pooled samples followed the same protocol as described above. Sample Testing To determine the concentrations of pathogen infections in the samples, we used a two-step reverse transcriptase quantitative PCR analysis. In step one, we used 1.0 microliter of the extracted RNA to synthesize cDNA using the BioBasic High Reverse Transcriptase Kit (Biobasic, Marhkam, Canada), after which we diluted the cDNA 5-fold. In step two, we performed real time PCR in triplicate on 384-well plates using Life Technologies PowerUp SYBER Green chemistry with a Quant Studio 6 Flex machine. We included standards for absolute quanti cation in each plate, which involved a serial dilution of known quantities of a custom synthesized plasmid containing the targets, with one negative control containing only water also included. We ran each PCR with a reaction volume of 5 microliters with modi ed cycling conditions from the PowerUp SYBER Green protocol. Even though under this protocol each plate completed 40 cycles during the PCR stage, we only included positive results that were within the range of the quanti ed standards. If a sample contained a positive result at a cycle number higher than the positive standards, it was not considered to be biologically relevant. Thus, the cycle number cutoff ranged from 28 -32 cycles, depending on the target and the speci c target's standard's results. We performed analyses using the included Quant Studio software and then normalized results to the reference gene levels using GeNorm 28 . In 2018, we collected a subset of the owers on which the bees were foraging and conducted pathogen screening in order to determine if they contained similarly detectable levels of the pathogens. Five ower heads per sampling event were removed, placed in individual bags, and transported back on dry ice as was done with the bee samples. We screened these owers against our panel of pathogens as detailed above; however, no pathogens were detected (data not included) and as such we did not analyze these data. Statistical Analysis Since there is an overdispersion of zeros in our dataset, we used a Zero In ated Negative Binomial model (ZINB) 29 with a logit link. Detection levels of each pathogen were analyzed in two ways; copy number (standardized to the reference gene), and relative intensity (categorized into non-detect (ND) if zero, and low, medium, or high based on the bottom two, third, and fourth quartiles of the natural log transformed copy numbers of each pathogen, respectively). To explore the pathogen dynamics of A. mellifera and B. impatiens, we included season, ower cover, and ower diversity as independent variables and copy number of each pathogen as dependent variables in a ZINB model. To explore the dynamics between these two species, we used ANOVA in base R and compared A. mellifera presence and relative intensity with the presence and relative intensity of B. impatiens. When constructing our ZINB models, year was not found to signi cantly impact pathogen detection in A. mellifera (all p > 0,34), except for BQCV detections (p < 0.05). As such, year was only included as a random effect in models when analyzing BQCV detection in A. mellifera. Additionally, sampling location was not found to signi cantly impact pathogen detection in A. mellifera (all p > 0.16), except for Nosema spp. detections (p < 0.0001). However, in order to maintain statistical power, it was not included in any of our models. Both year (p < 0.0001) and station (p < 0.005) were found to signi cantly impact Trypanosome spp. detections in B. impatiens, however through an AIC based approach for best model selection these variables were not included in our nal model. Due to low sample size and low pathogen presence, we were not able to conduct further analyses on the pathogen results from the remaining six bee species; however, these ndings are summarized descriptively below. All analyses were conducted in RStudio (version 3.6.2) using base R 30 , the pscl 31 package, and the boot 32 packages. Results We originally collected and screened 616 bee samples; however, we removed 114 samples from analysis as the ampli cation levels of one or both of the reference genes were at an unacceptably low level. As such, we included a total of 502 samples in our analysis-411 individually processed samples and 91 pooled samples ( Figure 1 and Table 3). Apis mellifera was the only bee species in which we detected any of the viruses in our study. The most commonly detected pathogen in A. mellifera was BQCV (40 individuals), followed by DWVa (32) (Table 3). Further, many individuals were found to be simultaneously infected with multiple pathogens, with two individuals infected with four pathogens (Figure 3). We found that LSV had the highest copy number overall, but that BQCV (29.0% of positive detections) and Trypanosome spp. (52.9% of positive detections) more often fell into the high category of relative intensity. Due to low, or no, positive detections, we were unable to analyze ABPV, CBPV, DWVb, and IAPV results for A. mellifera. From the pathogens we were able to analyze, we found that BQCV copy number was signi cantly highest in the spring (logq = -1.34; DF = 11; p < 0.0001, SE ± 1.10), and was lowest at medium ower diversity (p < 0.005, SE ± 2.04). LSV did not signi cantly change across the sampling season or ower diversity (all p-values > 0.18), but we detected the highest copy numbers in low ower cover (logq = -1.07; DF = 17; p < 0.0001, SE ± 1.51). Conversely, we detected the highest Trypanosome spp. copy number at high ower cover (logq = -0.25; DF = 17; p < 0.0005, SE ± 1.20; Figure 2) and when ower diversity was low (p < 0.0001, SE ± 1.27). Additionally, copy number of Trypanosome spp. was highest in late summer (p < 0.01, SE ± 1.71; Figure 3). To analyze the Nosema spp. results, ower diversity was removed from the model. We found that copy number detection level of Nosema spp. was highest in fall (logq = -0.44; DF = 13; p < 0.001, SE ± 1.63) and spring (p < 0.0001, SE ± 1.33) and was not signi cantly impacted by ower cover (p = 0.64). DWVa was not signi cantly in uenced by any of the variables in our model (all p-values > 0.06). We only analyzed Trypanosome spp. copy number within B. impatiens as no viruses were detected in any of our B. impatiens samples and only one individual was detected with Nosema spp. Copy numbers within B. impatiens (68 individuals) were higher than copy numbers in A. mellifera. We found that Trypanosome spp. copy number was signi cantly lowest in the fall (logq = -0.85; DF = 17; p < 0.005, SE ± 0.91; Figure 3) and signi cantly highest with medium ower diversity (p < 0.05, SE ± 0.88) and low (p < 0.01, SE ± 0.70; Figure 2) ower diversity. Trypanosome spp. copy number was not signi cantly in uenced by ower cover (all p-values > 0.24). When exploring pathogen dynamics between A. mellifera and B. impatiens we focused on Trypanosome spp. detections, as this was only pathogen detected in both species with high sample numbers. We did not nd any evidence that pathogen detection of one species was correlated with the pathogen detection of the other. Presence of a positive Trypanosome spp. detection in A. mellifera had no correlation with any relative intensity category in B. impatiens (all p-values > 0.15). Similarly, the relative intensity of Trypanosome spp. in A. mellifera had no correlation to the presence or relative intensity of Trypanosome spp. in B. impatiens (all p-values > 0.18). While we did not nd any positive detections of viruses in the other bee species tested in this study, we did nd gut pathogens. Trypanosome spp. were detected in B. pensylvanicus (6 pools), H. poeyi/ligatus (6), S. obliqua (1), and X. micans (1 individual; Table 3 and Figures 2-3). Nosema spp. was also detected in B. pensylvanicus (5 pools) and S. obliqua (1). Within these gut pathogen results, B. pensylvanicus had the highest copy number detection level for Nosema spp., by an entire order of magnitude, followed by S. obliqua and then A. mellifera. Bombus pensylvanicus had the highest copy number detection level of Trypanosome spp. again followed by S. obliqua,and then B. impatiens. Discussion Apis mellifera was the only pollinator species in which we detected any of the viruses included in our pathogen panel. However, we detected gut pathogens across most of the bee species tested. Some pathogen copy numbers-such as BQCV and Nosema spp. in A. mellifera, and Trypanosome spp. in both A. mellifera and B. impatiens-signi cantly changed across the sampling season, a nding that is similar to previous literature 33 . While other pathogen copy numbers-such as LSV in A. mellifera and Trypanosome spp. in both A. mellifera and B. impatiens-were signi cantly in uenced by ower cover; however, this occurred in opposite directions where LSV was highest at low ower cover and Trypanosome spp. were highest at high ower cover. Similar to previous literature 34 , Trypanosome spp. detection levels were highest in low ower diversity. While Trypanosome spp. detection patterns were similar in A. mellifera and B. impatiens, we found no evidence of correlations between these two species. These results suggest that the habitat is not acting as a pathogen hotspot for interspeci c bee pathogen dynamics but rather some other mechanism may be more critical in pathogen dissemination within bee communities. One explanation could be that even though shared oral resources have been documented as a source of spread for some pathogens 35 , the occurrence may actually be rather rare 36 and its success depends on the bee 15 and ower species in question 36 . It has also been suggested that non-host bees can reduce infection levels through the dilution effect 37 . It is possible that as time progresses and bees continue to utilize these habitats, the pathogens pressures will intensify intraspeci cally. Further long-term testing will be necessary to evaluate this possibility. Gut parasites are currently considered a serious threat to several bee species, especially bumble bees 25 ; of particular concern in North America is the American Bumble Bee (B. pensylvanicus). In our study, B. pensylvanicus had the greatest positive detections of gut parasites out of all the bee species tested, supporting the hypothesis that gut parasites pose a threat to their populations. At the time of writing this paper-but after the period when samples were collected and analyzed-the United States Fish and Wildlife Service (FWS) has announced a 90-day ndings petition for B. pensylvanicus populations in order to inform decisions surrounding its population status, and status reviews are underway in state FWS o ces. Currently in North Carolina, B. pensylvanicus is listed at "W3: Rare but Questionable Documentation" 38 and "Vulnerable/Apparently Secure" 38 , meaning more documentation is needed on this species before making any regulatory decision. Information from this study will be important in making future conservation decisions surrounding this and other species, and data from this study has already been shared with the NCFWS to do so. As gut pathogens are considered a threat to this species' population 39 , monitoring should be continued in future work. However, one consideration is noting the species of gut pathogens being detected. All samples in this study were screened for N. ceranae and preliminary results showed that some samples tested positive for the Nosema spp. primer but did not test positive for N. ceranae. It is possible that native bee populations are facing their own Nosema species which are not being actively monitored for; however, as the results were inconclusive, the data are not included here. Further, it is important to note that detecting a pathogen neither equates to infection nor demonstrates speci c health impacts of the pathogen 40 . For example, it has been suggested that the presence of N. ceranae in B. terrestris may be due to ingested spores passing through the gut rather than true infection 41 . Future research should prioritize evaluating the true infectivity and health impacts of these pathogens on a variety of bee species, taking into consideration the use of species speci c pathogen primers. Many studies have previously found the presence of what are traditionally called 'honey bee' viruses in various native bee species, something this study does not con rm. Given that several other recently published papers have also documented fewer detections than previous research [42][43][44] , the unexpected results require speculation as to why. Unlike most other studies, we collected honey bee samples as individual foragers rather than groups from nest entrances or even inside managed beehives. This could have resulted in lower infection levels in our samples (e.g., heavily infected bees may not live long enough or be su ciently healthy to forage) resulting in reduced pathogen detection and spread. Alternatively, oral diversity has been documented as an important factor for pathogen sharing and infection levels 16,34 . Thus, plant diversity could potentially be used as a tool to intentionally limit pathogen sharing between honey bees and native bees at these augmented habitats. This is something that should be investigated further in future research and taken into consideration when establishing new pollinator habitat. Another factor to consider when comparing the results from this study to previously published work is the techniques used to screen for pathogens. Many previous papers evaluating co-occurrence of pathogens between honey bees and native bees have used traditional PCR or ran reactions at very high cycle numbers, often using pooled samples (summarized in Tables 1 and 2). However, these results could be due to spurious PCR ampli cation, which is known to occur at 30 cycles and above 45 . When our RT-qPCR results from individual samples were re-scored at 35 cycles (the cycle cutoff in 46.3% of previous studies) rather than the cycle number recorded from the serially diluted standards, we saw an 81.3% increase in positive detections across all targets and all bee species (Figure 4). Similarly, we saw a 239.8% increase in positive detections when our samples were re-scored at 40 cycles (the cycle cutoff in 34.1% of previous studies). Within our standardized results, we saw no detection of CBPV in any bee species and only one individual each infected with ABPV and IAPV. When scoring at 35 and 40 cycles, however, all three of these pathogens were detected at much higher levels. While we believe these detections to be spurious, if they are true detections, it begs the question of biological relevance when pathogen infections are present at such low levels. Comparing across studies, therefore, should be done with extreme caution, because the different methodologies make direct comparisons di cult and potentially misleading. As planted habitat for pollinators will likely continue to be used as a tool in pollinator conservation, we should take care to establish this habitat with plant species that provide oral resources while limiting pathogen transmission. We should also prioritize conducting long-term monitoring of the bees within these habitats to ensure it continues to protect pollinator populations and their health over time.	2022-01-07T16:16:01.825Z	2022-01-05T00:00:00.000	{ "year": 2022, "sha1": "592f50bdaf5db5af6617447d19ff214d624efb67", "oa_license": "CCBY", "oa_url": "https://www.researchsquare.com/article/rs-1178773/latest.pdf", "oa_status": "GREEN", "pdf_src": "ScienceParsePlus", "pdf_hash": "78e5973e92b28dadc11460909ded50a9a4ec4742", "s2fieldsofstudy": [ "Environmental Science" ], "extfieldsofstudy": [] }
226628487	pes2o/s2orc	v3-fos-license	Influence of body mass index on variables of brain-stem evoked response audiometry in children aged 2-10 year with delayed development of speech attending Bankura Sammilani Medical College . rainstem evoked respond audiometry (BERA) or brainstem auditory evoked potential (BAEP) parameters are recorded from the scalp as small voltage potentials after passing auditory stimuli through a headphone 1 .There are mainly five BERA waveforms of which waves I, III and V are the most visible and of more significant clinical value.These waves represent the neuro electrical activity that is generated by the neural generators in the auditory pathway between the cochlea and the brainstem.Among the various factors that affect BERA test finding.Age and hearing loss with definitive prolongation of absolute peak latencies are well established in BERA 2 .The effect of sex on BERA parameters like absolute peak latencies (APL) and interpeak latency (IPL) difference showed non-significant differences between 2 and 13 years, whereas between 16 and 45 years, sex differences were evident.These differences can be attributed to hormonal factors, temperature variations and anthropometric components like height, weight and head size 3,4 .In few studies, the effects of BMI and head size was also studied and it was observed that both affected the BERA latencies and IPL differences independently 5,6 .Although the effect of these variables on BERA has been studied extensively outside India, there remains a limitation of data in the Indian context.Further, most studies have not included an important BERA feature, the waveform V:I amplitude ratio, which indicates the involvement of the auditory pathway in hearing loss, whether central or peripheral.In one such study, it was observed that both IV-V/I and IV-V/III amplitude ratios were independent of sex, but the IV-V/I ratio increased significantly with age 7 . BMI is very important parameters that influence BERA test variables.Although the correlation of BMI with BERA parameters has been studied recently in various group of population, there is no common consensus on it.Some studies show a positive correlation in normal population, whereas others report a weaker correlation.A few studies discuss correlation between BMI and BERA over the children with delayed development of speech. This cross sectional study tried to examine any possible association of body mass index on variables of brainstem evoked respond audiometry in children aged 2-10 years with delayed development of speech. Exclusion criteria Gross developmental anomaly, autism, very low IQ, recent ear infection, head injury, systemic disease or drug toxicity that affects ear were excluded from the study. Method of study Written consent was taken from parent or guardian or accompanying caregiver of all participants after explaining purpose and procedure of study.Children were chosen according to the inclusion and exclusion criteria.The participants were made familiar with the equipment and the test was discontinued if they faced any discomfort.History was taken regarding delayed speech/language impairment followed by general health check-up.BERA test was then done. Measurement of anthropometric parameters: Weight was measured using a standardized electronic weighing machine, with light clothes in standing posture without footwear.The height of the subjects was measured to the nearest centimetre using stadiometer, The body mass index (BMI) Recording of BERA: All participants were subjected to BERA testing according to standard techniques on an RMS Portable Aleron EP-Electromyograph machine manufactured by Recorders and Medicare System (RMS) (Chandigarh, India).Recording of BERA was carried out in a quiet and dimly lit room with the participant in the supine position.Participants were briefly informed about the procedure.Restless, irritable and apprehensive participants were allowed to relax for 5-10 min before testing. Patients or their parents/guardians were instructed to take/give shampoo bath to the hair on the day of the investigation so that the scalp is oil free.The non-inverting electrode is placed over the vertex B (CZ) of the head and the inverting electrode is placed over the ear lobes (Ai and Ac) or mastoid prominence.One more electrode known as the earthing electrode was placed over the forehead (ground).Monaural auditory stimulus consisting of rarefaction clicks of 100 μsec pulse were delivered through an electrically shielded earphone at a rate of 11.1/s.The contralateral ear was masked with pure white noise 30 dB below that of the BERA stimulus.A band pass of 10-3000 Hz was used to filter out undesirable frequencies in the surrounding.Responses to 2000 click presentations were live averaged to obtain a single BERA waveform pattern.Waveforms were obtained at 40, 60, 90 and 110 dB in each ear.Data of waveforms obtained at 110 dB were used for analysis.APL of waves I, III, V and IPL of I-III, III-V, I-V wave forms were considered for assessment. Statistical analysis Statistical analysis was carried out using SPSS version 20.0 (IBM SPSS Inc., Chicago, IL, USA).Descriptive analysis included the calculation of Mean with Standard Deviation and range for various parameters.Pearson's correlation coefficient was used for correlation statistics. Results Descriptive statistics of brainstem evoked response audiometry (BERA) parameters among all children are shown in table 1. Unpaired student 't' test was done between the underweight and nonunderweight children.The test showed significant changes (p value <0.05) in APL-I, APL-II and APL-III (Table 1).In underweight children significant correlation were observed between APL-I and V:I ratio with BMI. There is lack of association between APL-II, APL-III, I-III IPL, I-V IPL and III-V IPL with BMI (Table 2). In non-underweight children APL-V, III-V IPL and V:I ratio were significantly correlate with BMI and there is lack of association between APL-I, APL-III, I-III IPL and I-V IPL with BMI (Table 2). Discussion The present study observed the effect of BMI on different parameters of BERA in children aged 2-10 year.So we had already excluded sex factors from the study, it clearly accounts for the effect of only BMI with variables on BERA.There was significant correlation between any of the BERA parameters and BMI. Our study show in underweight children APL-I, APL-II and APL-III significantly lower in compare to non-underweight children whereas, I-III IPL slightly higher and only significant correlation were observed between APL-I and V:I ratio with BMI.In patients with a prolonged I-III inter peak latency and a normal III-V inter peak latency, the auditory dysfunction is assumed to be located between the distal part of CN VIII (near the cochlea) and the superior olivary nucleus, ipsilateral to the ear stimulated 11 . Present study in Non-underweight children APL-V, III-V IPL and V:I ratio were significantly correlate with BMI.In patients with a prolonged III-V inter peak latency and a normal I-III inter peak latency, the auditory conduction defect likely is located between the superior olivary nucleus and the inferior colliculus, ipsilateral to the ear stimulated 11 . We also explored the V/I amplitude ratio, which is an essential measure for delineation of central and peripheral auditory pathway involvement.The literature suggests that if the amplitude ratio is below 0.5, it may be consistent with central impairment and if it is above 3.0, it may be consistent with peripheral auditory pathway impairment 12 .Thus, with wide differences in the normative values obtained in each study and its importance in delineating the central versus the peripheral auditory pathway, further research in this domain with a large sample size is essential.BMI-related observations in BERA are in accordance with a previous study by Solanki et al 6 In a comparison study of BERA parameters between obese participants (BMI >30) and controls (BMI <30) among healthy young adults, significant differences were observed 13 .These findings were corroborative with observations in present study.But both studies have limitations of smaller sample size and hence results can be extrapolated to generalized population.Hence further studies with larger sample size are required to explore association of BMI with BERA parameters.Although, association of BMI with peripheral neuropathies have been well understood but it is association with cranial neuropathies remains unanswered. Conclusion It suggests both APL and IPL difference as more useful tool that affecting the outcome that needs to be considered.The diagnostic utility of BAEPs can be improved by taking into consideration these physiological variables.Utility of BAEP can be optimized by establishing normative data for every setup based on BMI along with age before using it as a clinical tool and needs further elaboration for re enforcement of this observation. 8 was 9 . calculated, then BMI percentile was computed using Centres for Disease Control and Prevention (CDC) growth charts Based on percentiles10 children were grouped into categories as given by Macek and Mitola as (1) underweight (BMI < 5 th percentile) and (2) non-underweight (BMI > 5 th percentile). Table 1 : BERA parameters of all children with delayed development of speech APL-absolute peak latency, IPL-inter peak latency Table 2 : Correlation between BERA parameters and BMI	2020-08-20T10:08:56.877Z	2020-01-01T00:00:00.000	{ "year": 2020, "sha1": "998319ac1888d126d3c7057f1f577415b4355e38", "oa_license": "CCBYSA", "oa_url": "https://doi.org/10.5455/jmas.91264", "oa_status": "GOLD", "pdf_src": "ScienceParsePlus", "pdf_hash": "d8337fff15704af0e6df2604d986311c7f771884", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
226975766	pes2o/s2orc	v3-fos-license	One Stop Career Centre for People with Disabilities Career is a journey of life that made the field of profession or employment options as a way to live. Careers are basis to generate income to sustain the needs of everyday life. Disabled people also need a job and benefit from the job same as a normal person. However, the attitudes of some prejudice community against the disabled people to find a job. The purpose of this paper is to discuss about website namely as one stop career centre for disabled people. This website helps disabled people who graduate from high school to find a job according to their qualifications and skills. Through this website, they can get information related to employment areas that they serve. Therefore, this website will help disabled people to get a job and to reduce the unemployment rate among them in Malaysia. Introduction Employment is essential and needed by human being, human need to work to meet their needs. The same applies to the disabled, they also need to work to fulfill their needs like a normal human. Generally, these people are not too weak to gets sympathy from others. Although they have certain disabilities, it does not mean that they are not able to do activities as a normal people. Although they are not able to do any work, but they are still able to use their mental and mind to make a contribution that can provide progress to this country, regardless of economic or social. Disabled people have difficulty in obtaining productive employment and provide a source of income for their living. The number of disabled people who were employed in the public sector and the private sector is less encouraging. Labour Department in Malaysia data showed that the number of disable people working is lower compared to those who completed school (Madinah, 2012). In the public sector, government have been implemented namely a Service circular 10/1988 which provides the basis of taking 1% disabled configure each agency and the Appointing Authority (AA) must be disabled at least 1% of the number of officers in such agencies subject to the application by the disabled and the appropriateness of its functions and facilities (Persons with Disabilities Act, 2008). However, it does not guarantee disabled people will get 1% for every job that is offered. This is because the circular was still tied to the determination of eligibility and basic conditions. In addition, the provision of 1% either as an opportunity or actual placement is still not enough. There are many challenges to be encountered by disabled people to enter the workforce. The challenges faced by them are a negative perception, fear and lack of confidence by the employer to take disabled people to work. In addition, there are also other issues that affect the opportunities for disabled people to enter the job market such as the issue of accessibility in and out of buildings (built environment), public places, workplaces and public transport. Awareness of employers, especially in the private sector is not much different from the past until now. Although some of the private sector employs Problem Statement Employment is an important asset in a man's life does not matter to normal people or people with disabilities. Employment is the main source of income and they are also responsible for providing and delivering various types of information. In addition, the ability to generate an employment workspace to create social ties between workers themselves and others. Those who do not have jobs will feel something down, especially financially, socially and psychologically. In other words, employment is a basic requirement for most people in the present age, where they think the job is a key determinant of quality of life (Zainida, 2009). For disabled people, employment is a very important thing in their lives, but many people still ignore these things in the life of the disabled. People always impress that disabled people do not need a job to earn a better life like a normal person. They also argue that the aid money given is sufficient to accommodate disabled people's lives. In reality, disabled people are the most creative and productive if given the chance. Physical disability is not an obstacle to work and what's important, employers need to be rational with special offers to those involved. The advantage is that they are more focused than others. They also realized that if they work with half-heartedly, the opportunity to work will be destroyed and to get a new job is not easy. However, negative attitudes of employers who make the wrong impression on the ability of disabled people, such as the production of low productivity, high humility and high accident rate and barriers to employment. Based on statistics from the Organization of the United Nations (2010), Malaysia has a population of approximately 28 million people. Based on these statistics, the number of disabled people in Malaysia is estimated at 2.8 million people. However, disabled people registered with the Social Welfare Department (SWD) of only 283 thousand. Heward (2000) found that there is a 50 % to 70% of people with disabilities who do not have jobs and among of them just received a minimum income. This is closely related to the attitude of some communities that show prejudice to the ability of the disabled to work. Attitudes of employers a still not opens their mind and are not ready to give disabled people the opportunity to work. Employer's refuse to give them jobs from several factors such as disabled workers need a lot of sick leave, higher insurance rates, safety in the workplace is disrupted, require special facilities and perhaps some changes had to be done in the workplace to ensure the safety of workers with disabilities. So, this will involve expenditure if employers employ them. Although disable people receive an education like other normal people, opportunities for them to work are still limited. Thus, the current jobless rate among them a still high while pushing the system of education for the disabled reviewed (Ministry of Human Resources, 2006;Norani et al., 2005). Even facing with the competition in an increasingly challenging job market, disabled people should be trained in vocational skills which meet the requirements and needs. Until now, they remain to be fully accepted in the open job market. The role played by each individual is different. Malaysia has a lot of job opportunities in various sectors such as agriculture, forestry, mining, electronics, wholesale and retail, manufacturing and others. However, opportunities for the disabled to find employment are limited. In general, suitable employment opportunities for disabled people in all sectors and occupations are still limited. Disabled people were not Other goals of the one stop career centre system were to develop an accurate performance assessment of its responsiveness to people with disabilities (Hall & Parker, 2005), and to create meaningful and seamless service delivery between the workforce and disability systems (Cohen, Timmons & Fesko, 2005). Seamless service delivery consists of a streamlined delivery of services by different agencies, and is smooth, coordinated, and efficient, reduces paperwork, avoids duplication, and links multiple programs into one system. Therefore, a person with a disability would not need to navigate multiple services systems in order to access a variety of necessary services, which can be overwhelming (Cohen et al., 2005). Instead, the partner agencies of the one stop career centre negotiate with each other to collaborate and coordinate their services, resulting in improved consumer outcomes and simplification of the process for customers. If the one stop career centre system was successful in achieving the goal of increasing the employment rate of people with disabilities, economic statistics would reflect this, and would likewise, reflect an increase in their employment. Evaluation of the performance of the one stop career centre's employment services, however, revealed that Workforce Investment Act customers with disabilities were typically less likely to enter employment and retain employment when compared to peers without disabilities (Hall & Parker 2005). Holcomb & Barnow (2004) found that, although the program enrolled a number of people with disabilities, only a small proportion of them were actually served. Furthermore, people with disabilities who exited the program had lower employment and earnings than other existing customers (Holcomb & Barnow 2004). In principle WIA's universal access provision along with the non-discrimination requirements of the Americans with Disabilities Act (ADA) addresses the accessibility needs of job seekers with disabilities at one stop. However, research shows that in practice persons with disabilities may still face a number of difficulties when utilizing the one stop system. Research has shown that not only the state workers, but also parents, teachers, and employers are not always aware of all the accessible resources for individuals with disabilities (Glazier & Tillmon, 1999). Some state workers may have more enhanced programs and educated employers while other state workers do not seem to know of all the programs and benefits available for individuals with disabilities. Despite the efforts of the WIA's and the one stop career centre system, only about 37% of people with disabilities are reported in the national workforce (U.S. Census Bureau, 2006). Six per cent of adults Salhazan Nasution, Mohd Hanafi Mohd Yasin, Noraidah Sahari, Nor Syaidah Bahri Jamil Ahmad, Hasnah Toran, Safani Bari, Mohd Mokhtar Tahar Seminar Internasional Pasca Siswazah Pendidikan Khas UKM-UPI Siri 4/2014 aged 16 to 64 with a disability report the presence of a condition that makes it difficult to remain employed or to find a job (U.S. Census Bureau, 2006). The average earnings of workers with disabilities are lower; they are under-represented in the workforce, and experience both higher rates of poverty and limited access to employee benefits (Timmons 2002;U.S. Census Bureau, 2006). The Workforce Investment Act's Standardized Record Data revealed a significant decrease from 2001 to 2003 in percentages of people with a disability successfully exiting the one stop career centre program (National Council on Disability, 2005). There was reluctance by the one stop career centre staff to serve people with disabilities because investment of the additional time and support required to assist those with disabilities could possibly result in inadequate outcomes to meet mandatory performance measures (Bader 2003;Hamner & Timmons 2005). It was found that employers who were aware of one stop career centres and used its services were more likely to be large and medium sized employers (U.S. Government Accountability Office, 2005). One Stop Career Centre Nowadays, with the amazing development and capabilities of information and communication technology is very help people with disabilities to find job. The ability of this technology is connecting people all over the world, store and disseminate information that can be accessed without going anywhere and done anywhere (Halimah, 2009). Thus, a career website for the disabled school graduates is being developed by researchers from the Faculty of Education, UKM. The one stop career centre system was designed to integrate, collaborate, and upgrade relevant community programs and resources, and to provide employers with a larger pool of qualified, skilled workers (Rutgers, 2002). Federally mandated partnerships in one stop career centres provided individuals with meaningful and seamless access to information, services, and opportunities in the world of work. The one-stop career centre system has been described as a "no-wrong door" because it allows individuals with disabilities to choose, receives, and blends a variety of employment and training services through a single door (Dew, McGuire-Kuletz & Alan, 2001). Productive and effective partnerships in one stop career centres also provide individuals with meaningful and seamless access to information, services, and opportunities in the world of work. One stop career centres are obligated to ensure that their facility and services are universally accessible to any individual seeking employment (Rutgers, 2002). So, people with disabilities have a right to use the one stop centre system, and are entitled to reasonable accommodations and modifications when using its services (Hoff, 2002). Unfortunately, many one stop career centres were not equipped to serve people with disabilities, and automatically referred these individuals to the public vocational rehabilitation system (Hoff, 2002). In Malaysia, Ministry of Human Resources has introduced Placement System for Persons with disabilities (SPOKU), a system that was developed to help the disabled and those with special needs to find a job that suits their qualifications and abilities in line with the society that promoted by the government. After registration, the individual will be helped to find a job that fits their criteria and conditions. Not all disabled people can do the job like everyone else. An age, type of disability and no experience is the reasons why people are often rejected from job applications. In this system, employers can register vacancies reserved for disabled workers. As of March 2007, the system has already registered Mohd Hanafi Mohd Yasin, Noraidah Sahari, Nor Syaidah Bahri, Salhazan Nasution, Jamil Ahmad, Hasnah Toran, Safani Bari, Mohd Mokhtar Tahar Seminar Internasional Pasca Siswazah Pendidikan Khas UKM-UPI Siri 4/2014 5 a total of 9,070 and has successfully placed a total of 6,799 people in the employment sector (Ministry of Human Resources, 2009). Although SPOKU system was launched by the government to help people with disabilities to obtain employment, the percentage of those who use or access to the system is still not satisfactory because still not reached its target. Therefore, this study aimed to develop a new system of one stop career centre which can be accessed by disabled students for secondary school graduates to find a job. This system can also be accessed by teachers, school counsellors, parents and caregivers for the disabled to fill students' personal information into the system. The cooperation of the teachers, parents and employers can play an important role for students with disabilities to help them make the transition from school to the working world. This website would be enhanced if more resources school-employer program is created, the involvement of teachers who are more aggressive, advanced training for students and the wider communication of a work between students and employers. Conclusion Disclosure of job opportunities for the disabled people can be done while they are still in school. Before they enter to job environment, the information related to the world of work for disabled can be collected by teachers and parents. In addition, they can get information at the school, internet, newspapers, radio, and television and so on. They can also get the information provided by government departments, NGOs, private agencies and etc. It is hoped that the development of this website will help the disabled people to get a job in line with their qualifications and skills. Indeed, all parties must play a role in opening up wider career opportunities to disabled so that they get the process to work smoothly and orderly. On the bright side, the cooperation from all parties allows for better employment opportunities for persons with disabilities. Giving a job to disabled in the private sector at least can reduce the unemployment rate of people with disabilities in Malaysia.	2020-11-18T02:01:25.888Z	2020-11-17T00:00:00.000	{ "year": 2020, "sha1": "dafffbf28b26db4fb76a95b6631263403f7965a9", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "dafffbf28b26db4fb76a95b6631263403f7965a9", "s2fieldsofstudy": [ "Education", "Sociology" ], "extfieldsofstudy": [ "Computer Science" ] }
247048220	pes2o/s2orc	v3-fos-license	Sepsis in Critically Ill Patients: Epidemiology, Risk Factors, and Role of Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry for Identification of Sepsis-Causing Organisms Background Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is a rapid mass spectrometry technology for species identification. It is a useful, fast, and accurate tool for routine laboratory analysis. This study aimed to investigate the epidemiology of sepsis-causing organisms in patients admitted to tertiary-level intensive care units (ICU), the role of MALDI-TOF MS in species identification, and patients’ clinical outcomes. Methodology A prospective observational study was performed in a tertiary-level ICU for one year. The first blood, urine, and endotracheal (ET) aspirate samples were sent before starting antibiotics. We received the antibiotic sensitivity report within 48 hours or earlier using MALDI-TOF MS. Treatment was modified based on MALDI-TOF MS reports. Response to treatment was monitored, and clinical outcomes were noted based on ICU stay. Patients were followed up until discharge, shifting to referring parent unit, or death. Results This study included 200 patients admitted to ICUs who at the time of admission did not have a fever. The most common organisms were Acinetobacter baumannii, Klebsiella pneumonia, and Escherichia coli in ET aspirates; Candida albicans and Enterococcus faecium in urine; and Pseudomonas aeruginosa, K. pneumonia, and A. baumaniiin blood. Of the 200 patients, 130 (65%) shifted to the parent unit ward, and 70 (35%) patients died, with an ICU stay of 12.89 ± 6.51 days. There was no significant difference in mortality when organisms grew from either ET or urine compared with sterile samples. If organisms resistant to all primary antibiotics grew from ET, mortality was 60.6%. Mortality was 56.8% if isolates were in the blood. Conclusions Early MALDI-TOF MS-based species identification and appropriate antibiotics initiation play a key role in the treatment and care for critically ill patients with sepsis. MALDI-TOF MS has the potential to significantly aid sepsis management. Introduction Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is a rapid mass spectrometry technology developed in the late 1980s. It is a useful, fast, and accurate tool for routine laboratory analysis. Several studies have reported that the introduction of MALDI-TOF MS reduces intensive care unit (ICU) and hospital stays of patients with bacteremia and/or candidemia and markedly decreases hospital costs [1,2]. However, to our knowledge, the impact of MALDI-TOF MS on the clinical outcomes of patients with sepsis, especially critically ill patients requiring admission to an ICU, remains poorly understood. MALDI-TOF MS utilization in a clinical microbiology laboratory has markedly increased over the past 10 years [3]. Over time, platforms have progressively advanced, with significant improvements in the software, interpretive rules, and databases. Consequently, there is limited value in comparing results for any category of organisms using a retrospective literature review. In this study, we hypothesized that rapid 1 2 3 1 1 MALDI-TOF MS-based identification on top of a well-established antimicrobial stewardship program (ASP) would significantly improve antibiotic management compared with an ASP using conventional identification even in a hospital setting with a low prevalence of resistant organisms. Consequently, we conducted this prospective study of a one-year duration investigating the epidemiology, risk factors, and role of MALDI-TOF MS in species identification of sepsis-causing organisms. This study aimed to investigate the epidemiology of sepsis-causing organisms in patients admitted to tertiary-level ICU care, the role of MALDI-TOF MS in species identification, and patients' clinical outcomes. Study setting and design This prospective observational study was conducted in a tertiary-level ICU from September 2019 to August 2020. In this study, we performed convenience non-probability sampling. The sample size was determined using the following equation: N = (r + 1/r) x σ 2 (Zβ + Zα/2) 2 /(difference) 2 , where n is the number needed in each group, r is the ratio of control to cases (valued at 1, an equal number of cases and control), σ is the standard deviation from the reference study, Zβ is the standard normal variate for a set power (0.84 for 80% power), Zα/2 is the standard normal variate for the level of significance (1.96 for 95% confidence interval, CI), and the difference of means in the reference study (d = 0.7) was 196.85. Finally, 200 cases included in the study. The minimum sample size for this study was statistically calculated based on the study by Chatterjee et al. [4]. Inclusion and exclusion criteria We included patients aged 16-60 years who were admitted to the ICU, those with new-onset fever after 48 hours of hospital admission, and those with new-onset neutropenia with fever. Because the study was conducted in adult ICU, pediatrics patients were excluded. As the number of patients above the age of 60 years was low, we included patients between the age group of 16 and 60 years. We excluded patients unwilling to provide consent, those having fever at admission, and those developing fever within 48 hours of admission. Additionally, we excluded all patients who presented with the diagnosis of sepsis at the time of admission. Study procedure The study was performed in a tertiary-level ICU. If a patient met the inclusion criteria and was willing to participate, informed consent was obtained. The first sample was collected at the time of admission. Repeat samples after developing fever with two sets of peripheral blood cultures were collected, each containing at least 8-10 mL. One sample from the central venous line was also collected (if applicable) along with peripheral blood culture. We received an antibiotic sensitivity report within 48 hours or earlier using MALDI-TOF MS. Treatment was initiated based on MALDI-TOF MS and susceptibility test findings. Response to treatment was monitored, and clinical outcomes were documented based on the length of ICU stay. The cases were followed up till the patients were discharged, shifted to a referring parent unit ward, or deceased. Statistical analysis The data were entered in Microsoft Excel and analyzed using SPSS version 23 (IBM Corp., Armonk, NY USA). The Student's t-test was used to analyze parametric data, and the Mann-Whitney U test was used to analyze non-parametric data. Fisher's test was used to analyze categorical data. The confidence interval (CI) was set at 95%, the alpha value was 0.05, the power of 1-ß was 0.8, and the level of statistical significance was set at P-values of <0.05. Results In this study, the most common isolated organisms were Acinetobacter baumannii, Klebsiella pneumoniae, Escherichia coli, and Pseudomonas aeruginosa in endotracheal (ET) aspirates; Candida albicans, Enterococcus faecium, and C. glabrata in urine; and P. aeruginosa, A. baumannii, and K. pneumoniae in the blood ( Table 1). This study was conducted in a government-aided hospital where we received referred patients with sepsis-induced multiorgan dysfunction syndrome (MODS) with multidrug-resistant infections. Crossinfection may be one of the causes for the high incidence of multidrug-resistant organisms. Because we only included patients who developed a fever after admission, it may be one of the causes of high incidence. ET: endotracheal If microorganisms growing in ET aspirates were resistant to all primary antibiotics, mortality was 60.6% ( Table 8). Discussion Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection [5]. The Surviving Sepsis Campaign recommends attempting to prove the presence of an infection by methods such as microbiological testing [6]. The gold standard to detect bloodstream infections is blood culture in a liquid media [7], followed by cultural and biochemical identification of the culprit pathogen. Thus, rapid blood culture testing with pathogen identification is essential to establish a diagnosis and enable efficient therapy. Severe sepsis and septic shock are the most common causes of morbidity and mortality in critically ill patients. Angus et al. [7] reported severe sepsis incidence of 2.26 cases per 100 hospital discharges, with 51.1% requiring intensive care. The overall mortality rate was 28.6%, which increased in patients with comorbidities. The overall mortality in gram-negative bacteremia is 25%. When septic shock develops, the mortality increases to 50-60% [8]. However, despite dramatic improvements in our knowledge of the pathogenesis, diagnosis, and therapeutic and supportive care, the mortality in septic patients remains unacceptably high, ranging from 30% to 50% in severe sepsis, and increasing to 50-87% in septic shock patients [9]. In the study, we found that the most common organism isolated from ET aspirates (n = 186) was A. baumanii (91), followed by K. pneumoniae (14) and E. coli (9). In urine, out of 197 samples, the most common organisms were C. albicans (10) and E. faecium (9). In blood (n = 197), the most common organisms were P. aeuroginosa (10), K. pneumoniae (9), and A. baumanii (9). This study was conducted in a government-aided hospital where we received referred patients with sepsis-induced MODS with multidrug-resistant infections. Cross-infection can be a cause for the high incidence of multidrug-resistant organisms. As we included patients who developed a fever after admission that may be one of the causes of high incidence. Our results are consistent with the increasing trend of gram-negative septicemia in ICU [8]. Thus, there is a need for early diagnosis and species identification to initiate species-specific antibiotics and early shifting from empirical to a specific therapy. Traditional techniques for culture and antibiotic susceptibility usually take four to five days to provide results and delay the initiation of appropriate antibiotics. Our findings were comparable to the study by Khwannimit and Bhurayanontachai [9] who reported that the respiratory tract was the most common site for both community and hospital-acquired infections. In our study, 10% of patients had diabetes, followed by hypertension (7.5%), chronic obstructive pulmonary disorder (4.5%), COVID-19 (3.5%), and HCV (3.0%), whereas 72% did not have any of the above risk factors. Our findings were consistent with those of Zhong et al. [10] who reported diabetes mellitus, chronic obstructive pulmonary disease, and chronic hepatic insufficiency as the major risk factors for bacterial bloodstream infections. About 72% of our patients did not have any comorbidity; therefore, the data are inconclusive to draw results. A major advantage of using MALDI-TOF MS was that it reduced the time between blood drawing and receiving the results about organisms causing sepsis [11]. Our findings were consistent with the previous study proving the efficacy of MALDI-TOF MS in critically ill patients [3]. We collected hospital data from our ICU to see any mortality benefit. In 2018, mortality was 35.36% (325/919), and in 2019, it was 35.34% (328/928). Thus, in our study, it was about 35%, although mortality is multifactorial and does not depend on one factor. Hence, we cannot conclude that MALDI-TOF MS has mortality benefits in sepsis patients, but it has an established role in early detection, early initiation of specific antibiotics, early de-escalation of antibiotics, and reduced ICU stay. Considering the antibiotics susceptibility of the organisms isolated from ET aspirates, A. baumanii was most susceptible to tetracycline, tobramycin, levofloxacin, and meropenem. In blood, P. aeuroginosa was the most common isolated organism and was susceptible to piperacillin-tazobactam and gentamycin. In urine, isolated Candida spp. was susceptible to both fluconazole and voriconazole. Patients who developed symptoms or neutropenia were treated with fluconazole. We observed clinical outcomes of the patients regarding ICU stay. Of a total of 200 patients, 130 (65%) patients shifted to the referring parent unit, and 70 (35%) patients died, with an ICU stay of 12.89 ± 6.51 days. There was no significant difference in mortality when microorganisms grew either in ET or urine compared with sterile samples. If microorganisms resistant to all primary antibiotics grew from ET, mortality was 60.6%. Mortality was 56.8% if the isolates were in the blood. Out of 200 patients, 186 patients were mechanically ventilated, with mortality of 37%, and the rest 14 patients were transferred to the parent referring unit. Therefore, mechanical ventilation is an independent risk factor of mortality in ICU. Zhong et al. [10] reported that in 117 patients, the mean ICU stay was 14 days. Further 28-day mortality was 35.0%, whereas 60-day mortality was 39.3%, and in-hospital mortality was 42.7%. Su et al. [1] also reported on the clinical impact of patients with bloodstream infection with different groups of viridians streptococci by using MALDI-TOF MS, concluding that the mean ICU stay was significantly higher in patients who survived than those who died (P < 0.01). The present study has some limitations. Because it was a single-center study of one ICU, antibiogram and organisms may not reflect the actual burden. Moreover, there was no data to compare the results of MALDI-TOF MS with the traditional methods. Due to the SARS-CoV-2 infection, results became incomparable to the past data. The sample size was small to demonstrate any mortality benefit of MALDI-TOF MS and the inability to discriminate between related species. Additionally, MALDI-TOF MS is currently unable to differentiate E. coli from Shigella. Conclusions As the incidence of gram-negative organisms is rapidly increasing in sepsis patients, early recognition and species identification can be achieved by MALDI-TOF MS, and initiation of appropriate antibiotics plays an important role in the treatment and care of patients. MALDI-TOF MS has the potential to revolutionize sepsis management. A major advantage of using MALDI-TOF MS was that it reduced the time between blood drawing and receiving the results regarding organisms causing sepsis. Although the time required for the identification by MALDI-TOF MS is 10 minutes/strain, we received the final sensitivity reports within 48 hours, which earlier used to take four to five days. MALDI-TOF MS is quick and accurate, providing species identification within two to six hours from positive growth to sensitivity testing and reporting. Our findings show that the use of MALDI-TOF MS has the potential to replace traditional methods of species identification, and routine use may have mortality benefits in patients with sepsis as sepsis management is time-specific. The tool has a role in the early initiation of specific therapy and early de-escalation of treatment. A larger randomized control trial is needed to establish mortality benefits. Additional Information Disclosures Human subjects: Consent was obtained or waived by all participants in this study. Institutional Ethics Committee (King George's Medical University, Lucknow, India) issued approval 102 ECM II B-THESIS/P79. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.	2022-02-23T16:05:15.455Z	2022-02-01T00:00:00.000	{ "year": 2022, "sha1": "bc9a91b21f65194445ed0a83e15a35aad47675ea", "oa_license": "CCBY", "oa_url": "https://www.cureus.com/articles/83129-sepsis-in-critically-ill-patients-epidemiology-risk-factors-and-role-of-matrix-assisted-laser-desorptionionization-time-of-flight-mass-spectrometry-for-identification-of-sepsis-causing-organisms.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "22c9f1aadc8c1756a1d8affeac3e8ac0ccfffc40", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
174810263	pes2o/s2orc	v3-fos-license	An Electrospun Preparation of the NC/GAP/Nano-LLM-105 Nanofiber and Its Properties In this work, an energetic composite fiber, in which 2,6-diamino-3,5-dinitropyrazine-1-oxide (LLM-105) nanoparticles intimately incorporated with a nitrocellulose/glycidyl azide polymer (NC/GAP) fiber, was prepared by the electrospinning method. The morphology and structure of the nanofiber was characterized by scanning electron microscopy (SEM), energy dispersive X-Ray (EDX), fourier transform infrared spectroscopy (IR), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and Brunauer–Emmett–Teller (BET). The nanofibers possessed a three-dimensional (3D) net structure and a large specific surface area. Thermal analysis, energetic performance, and sensitivities were investigated, and they were compared with NC/GAP and LLM-105 nanoparticles. The NC/GAP/nano-LLM-105 nanofibers show higher decomposition rates and lower decomposition temperatures. The NC/GAP/nano-LLM-105 decomposed to CO2, CO, H2O, N2O, and a few NO, -CH2O-, and -CH- fragments, in the thermal-infrared spectrometry online (TG-IR) measurement. The NC/GAP/nano-LLM-105 nanofibers demonstrated a higher standard specific impulse (Isp), a higher combustion chamber temperature (Tc), and a higher specialty height (H50). The introduction of nano-LLM-105 in the NC/GAP matrix results in an improvement in energetic performance and safety. Electrospinning is a universal technology that is used to obtain multifarious nanocomposite [13][14][15][16]. The as-spun 3D nanofibers, with high specific surface areas and porosities are desired carrier for supporting nanoparticles [17][18][19]. However, the application of electrospinning technology in composite energetic materials is rarely performed [20]. For instance, nitrocellulose/aluminum-cupric oxide (NC/Al-CuO) and nitrocellulose/2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane (NC/CL- 20) nanofibers with high burning rates were obtained via electrospinning [21,22]. Li also fabricated nanoboron/nitrocellulose (B/NC) electrospun nanofibers with excellent thermostability [23]. Similarly, researchers selected single NC as electrospinning matrix. However, simplex NC has a relatively large viscosity, which affects the morphology of the nanofiber, and results in low loading of the nano-explosive. In addition, a high spinning voltage could generate electric sparks and bring great safety risks. The composite energetic matrix can compensate these deficiencies. GAP is a high-energy prepolymer with low viscosity and high density. Moreover, it has more flexible segments, a lower glass transition temperature (T g ), and higher mechanical properties than NC. Currently, there is no report on using GAP as a matrix to load explosives with electrospinning [24][25][26]. In this work, ball milling nano-LLM-105 is assembled onto a NC/GAP composite matrix by electrospun technology, to create a new type of energetic 3D nanocomposite. In fact, it is not dangerous to prepare explosive materials by electrospinning and ball milling. This is because that the energetic material is very stable at a normal temperature and pressure, and under the protection of solvent. At this time, they are no different from inert materials. Further experiments suggest that the NC/GAP/nano-LLM-105 nanofibers possess lower sensitivity and remarkable thermal decomposition and energy performance, which makes the nanofibers have application potentials in the field of solid propellants. Fabrication of Nanofibers Firstly, nano-LLM-105 was prepared by the high-energy ball milling method. The ingredients, including 200 g balls, 6 g LLM-105, 60 mL deionized water, and 60 mL ethanol, are added into a mill jar. The four jars are sealed and immobilized on the ball mill. The mill rotates at 300 rpm for 6 hr. Then 0.3 g nano-LLM-105 is dissolved in 4.4 g acetone to get nano-LLM-105 suspension. Exactly 0.45 g NC and 0.45 g GAP were added into 4.4 g acetone to obtain a NC/GAP solution. The above-prepared nano-LLM-105 suspension and NC/GAP solutions were blended to obtain a NC/GAP/nano-LLM-105 precursor (12 wt %). The mass ratio of NC, GAP, and nano-LLM-105, respectively was set to 3:3:2. As a contrast, the NC/GAP precursor solution (12 wt %) was obtained by dissolving 0.6 g NC and 0.6 g GAP into 8.8 g acetone. The mass ratio of NC and GAP is set to 1:1. For the electrospinning process of these two nanofibers, the inner diameter of the stainless steel needle is 0.8 mm. The ambient humidity was controlled at 40-50%. The applied voltage was maintained at 12-18 kV. Additionally, the flow rate was fixed at 3-5 mL·hr −1 . Aluminum foil was used to collect the fibers, which were placed 12 cm away from the needle. The preparation scheme is described in Figure 1. Characterization The analyses of SEM, EDS, IR, XRD, and XPS were performed in order to investigate the morphology and structure of NC/GAP, nano-LLM-105, and NC/GAP/nano-LLM-105. Scanning electron microscopy (SEM) was performed on a Hitachi SU8010. The diameters of particles and fibers were measured by Nano Measurer 1.2 software. X-ray diffraction (XRD) analysis was performed on a DX-2700 X-ray diffractometer (Hao yuan) with Cu Kα radiation. The IR spectrum was obtained on an infrared spectrometer (American Thermo Fisher Scientific Nicolet 6700). XPS was conducted with X-ray photoelectron spectroscopy (XPS) and a PHI5000 Versa-Probe (ULVAC-PHI). The BET measurements of NC/GAP and NC/GAP/nano-LLM-105 were performed, utilizing nitrogen adsorption with a Micromeritics ASAP 2010 instrument. Thermal analyses for NC/GAP, nano-LLM-105, and NC/GAP/nano-LLM-105 were conducted on a differential scanning calorimeter (DSC, TA Model Q600) at heating rates of 5, 10, 15, and 20 °C/min. thermal-infrared spectrometry online (TG-IR) analyses of NC/GAP and NC/GAP/nano-LLM-105 were performed on a thermal analyzer system (TG/DSC, Mettler Toledo) coupled with a Fourier transform infrared spectrometer in a nitrogen atmosphere. The temperature range that we considered was 50 °C to 400 °C. The impact sensitivity was tested by using HGZ-1 impact equipment. In each test, 25 drop tests were carried out to calculate the H50, and each portion was performed three times to obtain a mean value and a standard deviation. Morphology and Structure Figure 2a,b reveals that there are some weaker agglomerates rather than hard agglomerates for LLM-105 nanoparticles, and there are no bridge between the particles. The particle diameter distribution is obtained by measuring a diameter of ~100 particles, and the results are displayed in Figure 2c-d. We acquired the volume curve by integrating the frequency curve. The mean diameter calculated from the frequency curve is 152 nm, which is same as the median diameter (d50 = 152 nm) calculated from the volume curve. For the SEM images of NC/GAP and NC/GAP/nano-LLM-105 (Figure 3a,b), it is clearly observed that both of the two nanofibers reveal 3D reticulate structures. The surface of the NC/GAP nanofiber is smooth and uniform. On the contrary, the surface of the NC/GAP/nano-LLM-105 nanofiber is rough and uneven. The difference in morphology is primarily caused by two factors. The addition of LLM-105 nanoparticles results in an inhomogeneity of precursor solution. Furthermore, partial LLM-105 nanoparticles are agglomerated during the electrospinning process. From Figure 3c-f, for NC/GAP nanofibers, the mean diameter and median diameter are 469 nm and 478 nm, respectively. By comparison, the mean diameter and median diameter of NC/GAP/nano-LLM-105 are 758 nm and 764 nm, respectively. It is apparent that the Characterization The analyses of SEM, EDS, IR, XRD, and XPS were performed in order to investigate the morphology and structure of NC/GAP, nano-LLM-105, and NC/GAP/nano-LLM-105. Scanning electron microscopy (SEM) was performed on a Hitachi SU8010. The diameters of particles and fibers were measured by Nano Measurer 1.2 software. X-ray diffraction (XRD) analysis was performed on a DX-2700 X-ray diffractometer (Hao yuan) with Cu Kα radiation. The IR spectrum was obtained on an infrared spectrometer (American Thermo Fisher Scientific Nicolet 6700). XPS was conducted with X-ray photoelectron spectroscopy (XPS) and a PHI5000 Versa-Probe (ULVAC-PHI). The BET measurements of NC/GAP and NC/GAP/nano-LLM-105 were performed, utilizing nitrogen adsorption with a Micromeritics ASAP 2010 instrument. Thermal analyses for NC/GAP, nano-LLM-105, and NC/GAP/nano-LLM-105 were conducted on a differential scanning calorimeter (DSC, TA Model Q600) at heating rates of 5, 10, 15, and 20 • C/min. thermal-infrared spectrometry online (TG-IR) analyses of NC/GAP and NC/GAP/nano-LLM-105 were performed on a thermal analyzer system (TG/DSC, Mettler Toledo) coupled with a Fourier transform infrared spectrometer in a nitrogen atmosphere. The temperature range that we considered was 50 • C to 400 • C. The impact sensitivity was tested by using HGZ-1 impact equipment. In each test, 25 drop tests were carried out to calculate the H 50 , and each portion was performed three times to obtain a mean value and a standard deviation. Figure 2a,b reveals that there are some weaker agglomerates rather than hard agglomerates for LLM-105 nanoparticles, and there are no bridge between the particles. The particle diameter distribution is obtained by measuring a diameter of~100 particles, and the results are displayed in Figure 2c-d. We acquired the volume curve by integrating the frequency curve. The mean diameter calculated from the frequency curve is 152 nm, which is same as the median diameter (d 50 = 152 nm) calculated from the volume curve. For the SEM images of NC/GAP and NC/GAP/nano-LLM-105 (Figure 3a,b), it is clearly observed that both of the two nanofibers reveal 3D reticulate structures. The surface of the NC/GAP nanofiber is smooth and uniform. On the contrary, the surface of the NC/GAP/nano-LLM-105 nanofiber is rough and uneven. The difference in morphology is primarily caused by two factors. The addition of LLM-105 nanoparticles results in an inhomogeneity of precursor solution. Furthermore, partial LLM-105 nanoparticles are agglomerated during the electrospinning process. From Figure 3c 764 nm, respectively. It is apparent that the diameter of NC/GAP/nano-LLM-105 is larger than that of the NC/GAP nanofibers. The difference in mean diameter of these two nanofibers is due to the fact that the LLM-105 nanoparticles are loaded onto the surface of NC/GAP. EDS analyses were performed to probe the surface elements of the nanofibers; the results are exhibited in Figure 4. From Figure 4a,b, the peaks at about 2 Kv belonged to the gold element sprayed during the test. There were only O, C, and N elements that were presented on the surfaces of those two fibers, illustrating that impurities were not introduced in the process of ball milling and electrospinning. The theoretical elemental contents are tabulated in Table 1. After the addition of the LLM-105 nanoparticles, the O content hardly varied, the N content appreciably increased, and the C content declined. The IR spectra of NC/GAP, NC/GAP/nano-LLM-105, and nano-LLM-105 are contrasted in Figure 5a. For NC/GAP/nano-LLM-105, the peaks at 3437, 3405, 3285 3233, and 1648 cm −1 respectively indicated symmetric, anti-symmetric stretching vibrations and deformation vibrations of -NH2 in nano-LLM-105; two strong absorption peaks located at 1480 and 1448 cm −1 corresponded to the stretching vibrations of the C=C skeleton in the ring of nano-LLM-105; the peak at 1577 cm −1 indicated anti-symmetric stretching vibrations of -NO2 in nano-LLM-105; the peaks at 1351 cm −1 and 1383 cm −1 corresponded to symmetric stretching vibrations of -NO2 in nano-LLM-105 [27]; the peak at 2101 cm −1 was ascribed to the stretching vibrations of -N3 from the GAP that was present [28]; the peaks at 1280 and 1648 cm −1 reflected the symmetric and anti-symmetric stretching vibrations of -ONO2 in NC, respectively [29]; the peak at 1075 cm −1 corresponded to the out-of-plane bending vibrations of C-H. Hence, the functional groups for NC/GAP/nano-LLM-105 were in accord with nano-LLM-105 and NC/GAP, indicating that the LLM-105 nanoparticles were well-combined with NC/GAP. These peaks were the same as the published report. Overall, the molecular structures of the nano-LLM-105 and NC/GAP do not alter in the process of electrospinning. There were no new EDS analyses were performed to probe the surface elements of the nanofibers; the results are exhibited in Figure 4. From Figure 4a,b, the peaks at about 2 Kv belonged to the gold element sprayed during the test. There were only O, C, and N elements that were presented on the surfaces of those two fibers, illustrating that impurities were not introduced in the process of ball milling and electrospinning. The theoretical elemental contents are tabulated in Table 1. After the addition of the LLM-105 nanoparticles, the O content hardly varied, the N content appreciably increased, and the C content declined. EDS analyses were performed to probe the surface elements of the nanofibers; the results are exhibited in Figure 4. From Figure 4a,b, the peaks at about 2 Kv belonged to the gold element sprayed during the test. There were only O, C, and N elements that were presented on the surfaces of those two fibers, illustrating that impurities were not introduced in the process of ball milling and electrospinning. The theoretical elemental contents are tabulated in Table 1. After the addition of the LLM-105 nanoparticles, the O content hardly varied, the N content appreciably increased, and the C content declined. The IR spectra of NC/GAP, NC/GAP/nano-LLM-105, and nano-LLM-105 are contrasted in Figure 5a. For NC/GAP/nano-LLM-105, the peaks at 3437, 3405, 3285 3233, and 1648 cm −1 respectively indicated symmetric, anti-symmetric stretching vibrations and deformation vibrations of -NH2 in nano-LLM-105; two strong absorption peaks located at 1480 and 1448 cm −1 corresponded to the stretching vibrations of the C=C skeleton in the ring of nano-LLM-105; the peak at 1577 cm −1 indicated anti-symmetric stretching vibrations of -NO2 in nano-LLM-105; the peaks at 1351 cm −1 and 1383 cm −1 corresponded to symmetric stretching vibrations of -NO2 in nano-LLM-105 [27]; the peak at 2101 cm −1 was ascribed to the stretching vibrations of -N3 from the GAP that was present [28]; the peaks at 1280 and 1648 cm −1 reflected the symmetric and anti-symmetric stretching vibrations of -ONO2 in NC, respectively [29]; the peak at 1075 cm −1 corresponded to the out-of-plane bending vibrations of C-H. Hence, the functional groups for NC/GAP/nano-LLM-105 were in accord with nano-LLM-105 and NC/GAP, indicating that the LLM-105 nanoparticles were well-combined with NC/GAP. These peaks were the same as the published report. Overall, the molecular structures of the nano-LLM-105 and NC/GAP do not alter in the process of electrospinning. There were no new The IR spectra of NC/GAP, NC/GAP/nano-LLM-105, and nano-LLM-105 are contrasted in Figure 5a. For NC/GAP/nano-LLM-105, the peaks at 3437, 3405, 3285 3233, and 1648 cm −1 respectively indicated symmetric, anti-symmetric stretching vibrations and deformation vibrations of -NH 2 in nano-LLM-105; two strong absorption peaks located at 1480 and 1448 cm −1 corresponded to the stretching vibrations of the C=C skeleton in the ring of nano-LLM-105; the peak at 1577 cm −1 indicated anti-symmetric stretching vibrations of -NO 2 in nano-LLM-105; the peaks at 1351 cm −1 and 1383 cm −1 corresponded to symmetric stretching vibrations of -NO 2 in nano-LLM-105 [27]; the peak at 2101 cm −1 was ascribed to the stretching vibrations of -N 3 from the GAP that was present [28]; the peaks at 1280 and 1648 cm −1 reflected the symmetric and anti-symmetric stretching vibrations of -ONO 2 in NC, respectively [29]; the peak at 1075 cm −1 corresponded to the out-of-plane bending vibrations of C-H. Hence, the functional groups for NC/GAP/nano-LLM-105 were in accord with nano-LLM-105 and NC/GAP, indicating that the LLM-105 nanoparticles were well-combined with NC/GAP. These peaks were the same as the published report. Overall, the molecular structures of the nano-LLM-105 and NC/GAP do not alter in the process of electrospinning. There were no new groups generated, indicating that NC, GAP, and nano-LLM-105 do not react with each other. Figure 5b shows the XRD patterns of samples. There were two main peaks at 28.4 and 33.2 • in pattern of nano-LLM-105 [10]. Also, there were no diffraction peaks in the pattern of NC/GAP. This is because LLM-105 is a type of crystal, and NC/GAP is a type of polymer. The peak positions of the LLM-105 nanoparticles were in line with those of the NC/GAP/nano-LLM-105 nanofibers, which means that the crystal phase of the LLM-105 nanoparticles does not transform by electrospinning. This is a superior feature of the electrospinning compared with recrystallization to prepare energetic materials. For the recrystallization method, if the solvent is not properly selected, the crystal phase is liable to transform. Song prepared 1,3,5,7-Tetranittro-1,3,5,7-tetrazocane (HMX) by solvent/non-solvent method, and the crystal phase of HMX changed from β-HMX to γ-HMX [30]. In this work, acetone is chosen as the solvent, and the LLM-105 nanoparticles are suspended in it, which avoids the recrystallization of the LLM-105. Nanomaterials 2019, 9, x FOR PEER REVIEW 6 of 15 groups generated, indicating that NC, GAP, and nano-LLM-105 do not react with each other. Figure 5b shows the XRD patterns of samples. There were two main peaks at 28.4 and 33.2° in pattern of nano-LLM-105 [10]. Also, there were no diffraction peaks in the pattern of NC/GAP. This is because LLM-105 is a type of crystal, and NC/GAP is a type of polymer. The peak positions of the LLM-105 nanoparticles were in line with those of the NC/GAP/nano-LLM-105 nanofibers, which means that the crystal phase of the LLM-105 nanoparticles does not transform by electrospinning. This is a superior feature of the electrospinning compared with recrystallization to prepare energetic materials. For the recrystallization method, if the solvent is not properly selected, the crystal phase is liable to transform. Song prepared 1,3,5,7-Tetranittro-1,3,5,7-tetrazocane (HMX) by solvent/non-solvent method, and the crystal phase of HMX changed from β-HMX to γ-HMX [30]. Morphology and Structure In this work, acetone is chosen as the solvent, and the LLM-105 nanoparticles are suspended in it, which avoids the recrystallization of the LLM-105. [31]. For the C1s spectrum of NC/GAP/nano-LLM-105, the peak was fitted to six peaks. The peaks located at 284.5 eV, 286.3 eV, and 288.2 eV were assigned to -C-C, C-N 3 , and -C-ONO 2 of GAP and NC. The peaks at 284.7 eV, 286.8 eV, and 287.6 eV corresponded to -C-C, -C-NH 2 , and -C-NO 2 in nano-LLM-105 [32]. The XPS spectrum of N1s consisted of seven peaks at 400.4 eV, 401.1 eV, 404.1 eV, 404.2 eV, 406.9 eV, 407.7 eV, and 408.2 eV, which corresponded respectively to -N=N=N, -NH 2 , -N=N=N, C-N in ring, -NO 2 , -ONO 2 , and N-O in the ring [28]. The -N 3 and -ONO 2 groups were ascribed to NC and GAP, respectively. Finally, the groups of C-N in ring, N-O in ring, -NH 2 and -NO 2 belong to LLM-105. Hence, we infer the existence of LLM-105 nanoparticles on the surface of NC/GAP, and there are no new chemical bonds being produced on the surface of the nanofibers. Figure 5b shows the XRD patterns of samples. There were two main peaks at 28.4 and 33.2° in pattern of nano-LLM-105 [10]. Also, there were no diffraction peaks in the pattern of NC/GAP. This is because LLM-105 is a type of crystal, and NC/GAP is a type of polymer. The peak positions of the LLM-105 nanoparticles were in line with those of the NC/GAP/nano-LLM-105 nanofibers, which means that the crystal phase of the LLM-105 nanoparticles does not transform by electrospinning. This is a superior feature of the electrospinning compared with recrystallization to prepare energetic materials. For the recrystallization method, if the solvent is not properly selected, the crystal phase is liable to transform. Song prepared 1,3,5,7-Tetranittro-1,3,5,7-tetrazocane (HMX) by solvent/non-solvent method, and the crystal phase of HMX changed from β-HMX to γ-HMX [30]. In this work, acetone is chosen as the solvent, and the LLM-105 nanoparticles are suspended in it, which avoids the recrystallization of the LLM-105. [31]. For the C1s spectrum of NC/GAP/nano-LLM-105, the peak was fitted to six peaks. The peaks located at 284.5 eV, 286.3 eV, and 288.2 eV were assigned to -C-C, C-N3, and -C-ONO2 of GAP and NC. The peaks at 284.7 eV, 286.8 eV, and 287.6 eV corresponded to -C-C, -C-NH2, and -C-NO2 in nano-LLM-105 [32]. The XPS spectrum of N1s consisted of seven peaks at 400.4 eV, 401.1 eV, 404.1 eV, 404.2 eV, 406.9 eV, 407.7 eV, and 408.2 eV, which corresponded respectively to -N=N=N, -NH2, -N=N=N, C-N in ring, -NO2, -ONO2, and N-O in the ring [28]. The -N3 and -ONO2 groups were ascribed to NC and GAP, respectively. Finally, the groups of C-N in ring, N-O in ring, -NH2 and -NO2 belong to LLM-105. Hence, we infer the existence of LLM-105 nanoparticles on the surface of NC/GAP, and there are no new chemical bonds being produced on the surface of the nanofibers. The nitrogen adsorption-desorption isotherms of NC/GAP and NC/GAP/nano-LLM-105 are displayed in Figure 7. The specific surface areas, pore volumes, and pore sizes of the samples are shown in Table 2. The isotherms are considered as class IV (H3-type hysteresis loop), indicating that the prepared nanofibers were mesoporous materials. At a low p/p o , there is the first steep portion of the isotherm, as the p/p o increases, adsorption of multiple layers begins. In the multi-layer adsorption process, capillary condensation a common accompaniment (IV isotherms). Capillary condensation and capillary evaporation generally do not occur at identical p/p o , resulting in the generation of hysteresis loops. The specific surface areas of the NC/GAP/nano-LLM-105 and NC/GAP nanofibers were 6.0545 and 4.3573, respectively. The higher surface area value of NC/GAP/nano-LLM-105 is ascribed it having a rough surface. Compared to the energetic composite prepared by other methods, energetic nanofibers have a larger specific surface area. For example, the specific surface area of NC/GAP/CL-20 when prepared by a sol-gel-supercritical method, is 2.7 [33]. The nitrogen adsorption-desorption isotherms of NC/GAP and NC/GAP/nano-LLM-105 are displayed in Figure 7. The specific surface areas, pore volumes, and pore sizes of the samples are shown in Table 2. The isotherms are considered as class IV (H3-type hysteresis loop), indicating that the prepared nanofibers were mesoporous materials. At a low p/p o , there is the first steep portion of the isotherm, as the p/p o increases, adsorption of multiple layers begins. In the multi-layer adsorption process, capillary condensation a common accompaniment (IV isotherms). Capillary condensation and capillary evaporation generally do not occur at identical p/p o , resulting in the generation of hysteresis loops. The specific surface areas of the NC/GAP/nano-LLM-105 and NC/GAP nanofibers were 6.0545 and 4.3573, respectively. The higher surface area value of NC/GAP/nano-LLM-105 is ascribed it having a rough surface. Compared to the energetic composite prepared by other methods, energetic nanofibers have a larger specific surface area. For example, the specific surface area of NC/GAP/CL-20 when prepared by a sol-gel-supercritical method, is 2.7 [33]. generation of hysteresis loops. The specific surface areas of the NC/GAP/nano-LLM-105 and NC/GAP nanofibers were 6.0545 and 4.3573, respectively. The higher surface area value of NC/GAP/nano-LLM-105 is ascribed it having a rough surface. Compared to the energetic composite prepared by other methods, energetic nanofibers have a larger specific surface area. For example, the specific surface area of NC/GAP/CL-20 when prepared by a sol-gel-supercritical method, is 2.7 [33]. Thermal Analysis The DSC thermograms of samples collected at different heating rates are displayed in Figure 8a-c. The kinetic and thermodynamic parameters for thermal decomposition are calculated with the DSC data; the results are displayed in Table 3. For all samples, the exothermic peak temperature increases with the increase of heating rate. For NC/GAP/nano-LLM-105, there was only one exothermic peak, demonstrating that NC, GAP, and nano-LLM-105 decompose synchronously. In addition, the exothermic peak temperature of NC/GAP/nano-LLM-105 is slightly lower than that of NC/GAP, and it is markedly lower than those of the LLM-105 nanoparticles. This manifests that the thermolysis of NC, LLM-105 nanoparticles, and GAP cooperate with each other. The activation energy (E K ), pre-exponential factor (A K ), and rate constant (k), are calculated by the Kissinger equation (Equation (1)) [29] and the Arrhenius equation (Equation (2)) [34]. The E K of NC/GAP/nano-LLM-105 (185.139 kJ·mol −1 ) is lower than that of the LLM-105 nanoparticles and is higher than that of NC/GAP. The k of NC/GAP/nano-LLM-105 is higher than the k of the NC/GAP and LLM-105 nanoparticles, implying that NC/GAP/nano-LLM-105 has a higher decomposition rate. ln β where T p is the peak temperature in the DSC trace, with a heating rate of 15 • C·min −1 ; K B and h are the Boltzmann (K B = 1.381 × 10 −23 J/K) and Planck constants (h = 6.626 × 10 −34 J/s), respectively; β is the heating rate. The thermal decomposition of the energetic molecules originates from the activation and rupture of the weakest bond, which is quite significant for the decomposition process. As the temperature of the explosive increases, the molecular thermal motion is enhanced. When the temperature attains a critical point, the weakest bond will be stretched. Subsequently, a rupture occurs. This activation process could be described by the parameters of activation enthalpy (∆H ), activation free energy (∆G ), and activation entropy (∆S ), as calculated by Equations (3)-(5) [27]. ∆H is the energy that the molecules absorb to transform from a common state to an activated state. Compared with nano-LLM-105, NC/GAP/nano-LLM-105 needs a lower level of energy to be activated. ∆G is the chemical potential of the activation course. For all of these samples, the values of ∆G are positive numbers, indicating that none of the activation courses proceed spontaneously [28]. Figure 8e shows a kinetic compensation effect during the thermolysis of NC/GAP, nano-LLM-105, and NC/GAP/nano-LLM-105. The three points do not present a linear relationship, which means that the three samples have disparate kinetic mechanisms of decomposition. In addition, Wang prepared 1,3,5,7-tetranittro-1,3,5,7-tetrazocane/nitrocellulose (HMX/NC) and NC/GAP/CL-20 by the sol-gel method. Similarly, the samples do not have just one exothermic peak, indicating that thermal decomposition for them is carried out in multiple steps. Moreover, E K of HMX/NC iii and NC/GAP/CL-20 i are 277.68 kJ·mol −1 and 296 kJ·mol −1 , respectively, which are significantly higher than the E K of NC/GAP/nano-LLM-105 [29,33]. This is because, compared with the nanofibers that are prepared by electrospinning, the components of an energetic composite prepared by the sol-gel method cannot be tightly combined with each other, and there is a lower mutual promotion of the components for thermal decomposition. indicating that thermal decomposition for them is carried out in multiple steps. Moreover, EK of HMX/NC iii and NC/GAP/CL-20 i are 277.68 kJ·mol −1 and 296 kJ·mol −1 , respectively, which are significantly higher than the EK of NC/GAP/nano-LLM-105 [29,33]. This is because, compared with the nanofibers that are prepared by electrospinning, the components of an energetic composite prepared by the sol-gel method cannot be tightly combined with each other, and there is a lower mutual promotion of the components for thermal decomposition. The products for the thermal decomposition of NC/GAP/nano-LLM-105 and NC/GAP were investigated by TG-IR. The TG and DTG curves are displayed in Figure 9a,b, and the IR spectra at different temperatures are shown in Figure 9c,d. For NC/GAP, the decomposition began at 177 °C, The products for the thermal decomposition of NC/GAP/nano-LLM-105 and NC/GAP were investigated by TG-IR. The TG and DTG curves are displayed in Figure 9a,b, and the IR spectra at different temperatures are shown in Figure 9c nanofibers, the positions of the main peaks are mainly identical. The only difference is the generation of -C-O-C-fragments for NC/GAP nanofibers. However, there are no -C-O-C-fragments for NC/GAP/nano-LLM-105. In addition, compared with NC/GAP, the peak intensities of the-CH2O fragments decreased significantly. This indicates that for NC/GAP/nano-LLM-105, the fragments of -CH2O and -C-O-C-further reacted to generate CO2, CO, and H2O. This is because the interposition of the LLM-105 nanoparticles in NC/GAP improved the oxygen balance. Energetic Performance and Sensitivities To further explore the energy properties of NC/GAP/nano-LLM-105, their energy performances and impact sensitivities were evaluated. The standard specific impulse (Isp), characteristic velocity (C), combustion chamber temperature (Tc), and average molecular weight (Mc) were calculated, the results are listed in Table 5. The functional relationships between the energy performances of Energetic Performance and Sensitivities To further explore the energy properties of NC/GAP/nano-LLM-105, their energy performances and impact sensitivities were evaluated. The standard specific impulse (I sp ), characteristic velocity (C), combustion chamber temperature (T c ), and average molecular weight (M c ) were calculated, the results are listed in Table 5. The functional relationships between the energy performances of NC/GAP/LLM-105 nanofibers and the weight percentage of LLM-105 are shown in Figure 10. Table 4. Furthermore, the combustion products and their mass molar ratios were calculated, and the results are shown in Figure 11. The feature height (H 50 ) for NC/GAP/nano-LLM-105 was significantly higher than H 50 for NC/GAP, and it was a little lower than nano-LLM-105. This indicates that the impact sensitivity of NC/GAP/nano-LLM-105 was distinctly lower than that of NC/GAP, and slightly higher than that of nano-LLM-105. For energy performance, the standard specific impulse (I sp ) of NC/GAP was 2013.8 N·s·kg −1 . For NC/GAP/nano-LLM-105, the I sp value increased to 2032.4 N·s·kg −1 . This was not attributed to the high energy of LLM-105, but it was rather due to the higher formation enthalpy (∆H f ) and C/H values of LLM-105, in contrast to those of NC/GAP. The formation enthalpy of LLM-105 (−13 kJ·mol −1 ) was observably larger than that of NC/GAP (−294.6 kJ·mol −1 ). Therefore, after LLM-105 was introduced into NC/GAP, the energy performance was dramatically improved. The value of oxygen balance (OB CO2 ) was a crucial element for assessing the energy performance. As the OB CO2 increased, the energy performance is enhanced. The OB CO2 of LLM-105 (−37.03) is higher than that of NC/GAP (−76.1). Hence, the introduction of LLM-105 is in favor of improvement for the energy performance. Additionally, the C/H mass ratios of nano-LLM-105 and NC/GAP are 12 and 8.24, respectively. We infer that the higher OB CO2 and C/H are beneficial to enhancement under combustion temperature (T c ) [33,35]. T c represents the chemical energy storage of an energetic formulation, which is proportional to the explosive heat of propellants. Figure 10c shows that the value of T c increases as the weight percentage of LLM-105 increases. The energy performance is mainly determined by the heat released from combustion, and the energy conversion efficiency. The latter is related to the hydrogen content in the molecules. The decrease of hydrogen content leads to a decrease of H 2 content. Therefore, the average molecular weight (M c ) increases, which is disadvantageous to energy conversion efficiency. From Figure 11, for NC/GAP, H 2 accounts for 28% in combustion products, and the H 2 proportion of LLM-105 is 19%. Hence, the M c of LLM-105 is higher than that of NC/GAP in Table 4. Figure 10d shows that the value of M c increases as the weight percent of LLM-105 increases. In this case, although M c of LLM-105 is higher than that of NC/GAP, the I sp of LLM-105 is still significantly higher than I sp of NC/GAP. This is because the negative effect of low energy conversion efficiency is offset by the high storage of chemical energy. The higher chemical energy storage of LLM-105 is attributed to its higher C/H mass ratio, OB CO2 , and the formation enthalpy. Therefore, the interposition of LLM-105 improves the energy performance of NC/GAP. Table 5. Impact sensitivity and energy performance of the samples. Figure 11. (a-c) Combustion products and their molar ratios for NC/GAP/LLM-105 nanofibers. The results in Figure 11 were calculated by the means of the ProPep 3.0 software under conditions of P c /P e = 70/1 (P e = 0.1 MPa) and T 0 = 298 K. Conclusions The NC/GAP/nano-LLM-105 composite nanofiber with a large specific surface area was prepared by an electrospinning technique. Compared with NC/GAP and LLM-105 nanoparticles, NC/GAP/nano-LLM-105 nanofibers have lower decomposition temperature and distinctly higher decomposition rate. The activation energy of NC/GAP/nano-LLM-105 for thermolysis is lower than that of LLM-105 nanoparticles. These indicate that NC/GAP/nano-LLM-105 decompose relatively easily and violently. If it is used in solid rocket propellant systems, it will decompose first, and then induce the decomposition of other components. The I sp and T c of NC/GAP/nano-LLM-105 are higher than those of NC/GAP, which means that NC/GAP/nano-LLM-105 possesses a distinguished energy performance. In addition to the energy performance, safety is another rather crucial factor for energetic materials. The impact sensitivity of NC/GAP/nano-LLM-105 is signally lower than that of NC/GAP. Hence, it has a good safety performance rating. NC/GAP/nano-LLM-105 possess both energy performance and low impact sensitivity. Therefore, this composite nanofiber has enormous potential in the field of solid rocket-propellant systems. This versatile preparation method may provide a concept for synthesizing energetic nanocomposites.	2019-06-07T20:32:30.250Z	2019-06-01T00:00:00.000	{ "year": 2019, "sha1": "45d30275e3e595d218e6365ae10c5f28d0ff851e", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2079-4991/9/6/854/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "45d30275e3e595d218e6365ae10c5f28d0ff851e", "s2fieldsofstudy": [ "Engineering", "Physics" ], "extfieldsofstudy": [ "Medicine", "Materials Science" ] }
501328	pes2o/s2orc	v3-fos-license	The carboxyl-terminal tail of the stalk of Arabidopsis NACK1/HINKEL kinesin is required for its localization to the cell plate formation site Plant cytokinesis is achieved by formation of cell plates in the phragmoplast, a plant-specific cytokinetic apparatus, which consists of microtubules (MTs) and microfilaments. During cytokinesis, the cell plate is expanded centrifugally outward from the inside of cells in a process that is supported by dynamic turnover of MTs. M-phase-specific kinesin NACK1, which comprises the motor domain at the amino-terminal half to move on MT bundles and the stalk region in the carboxyl-terminal half, is a key player in the process of MT turnover. That is, the specific region in the stalk binds the MAP kinase kinase kinase to activate the whole MAP kinase cascade, which stimulates depolymerization of MTs for the MT turnover. The stalk is also responsible for recruiting the activated kinase cascade to the mid-zone of the phragmoplast, which corresponds to the cell-plate formation site. It should be crucial to uncover roles of the NACK1 kinesin stalk as well as the motor domain in the formation of cell plates in order to understand the mechanisms of cell plate formation. Using dissected Arabidopsis NACK1 (AtNACK1/HINKEL) molecules and AtNACK1-fused GFP, we showed that the C-terminal tail of the stalk in addition to the motor domain is critical for its proper localization to the site of cell plate formation in the phragmoplast, probably by affecting its motility activity. Introduction Cytokinesis is the process whereby one cell physically separates into two at the end of the cell cycle, which is essential for the growth and development of multicellular organisms. If cytokinesis fails to occur in animals, multinucleate cells, heteroploid cells, and aberrantly amplified centrosomes are formed, which are often related to tumorigenicity and apoptosis of cells (Normand and King 2010). In plants, defects in cytokinesis cause the generation of multinucleate cells and accumulation of stubs containing premature cell plates in aberrantly enlarged cells, which results in dwarfism of plant bodies and organs Nishihama et al. 2002;Sazuka et al. 2005;Strompen et al. 2002). Abstract Plant cytokinesis is achieved by formation of cell plates in the phragmoplast, a plant-specific cytokinetic apparatus, which consists of microtubules (MTs) and microfilaments. During cytokinesis, the cell plate is expanded centrifugally outward from the inside of cells in a process that is supported by dynamic turnover of MTs. M-phase-specific kinesin NACK1, which comprises the motor domain at the amino-terminal half to move on MT bundles and the stalk region in the carboxyl-terminal half, is a key player in the process of MT turnover. That Plant cytokinesis is achieved by the formation of a cell plate in the characteristic cytokinetic machinery of plant cells called the phragmoplast, which is mainly composed of aligned microtubules (MTs) and actin filaments (Müller et al. 2009;Otegui et al. 2005). Formation of the phragmoplast starts at a certain time in anaphase of the late M phase between two separating daughter nuclei in the internal region of the cytoplasm Otegui et al. 2005). As cytokinesis proceeds in a cell, bundles of phragmoplast MTs expand centrifugally from the interior to the periphery of the cell, driving the centrifugal development of a cell plate behind the expanding phragmoplast MTs. In the newly synthesized cell plate, a new cell wall is constructed within the lumens of membranous tubules that are formed by the fusion of membrane vesicles transported by membrane trafficking systems with MTs (Jürgens 2005;McMichael and Bednarek 2013;Van Damme and Geelen 2008;Woollard and Moore 2008). Although some vesicle trafficking-related components such as KNOLLE, Rab-A GTPases, and PATL1 have been identified (Chow et al. 2008;Lauber et al. 1997;Peterman et al. 2004;Waizenegger et al. 2000), the molecular mechanisms for the trafficking involving these factors in the formation of cell plates have yet to be determined. Expansion of the phragmoplast in tobacco BY-2 cells is positively regulated by the mitogen-activated protein (MAP) kinase cascade, all the components of which localize to the equatorial zone of the phragmoplast corresponding to the site of cell plate formation Soyano et al. 2003). Activity of the cascade is controlled by cytokinesis-specific kinesins, which are designated NACK1 and NACK2 in tobacco cells. NACK1 protein is located in the fraction that is purified biochemically from phragmoplasts by following the microtubulebased motility activity, suggesting that NACK1 may have a phragmoplast-associated motility activity . Both kinesins localize to the equatorial zone of the phragmoplast during cytokinesis and directly bind to the NPK1 MAP kinase kinase kinase (MAPKKK) to increase NPK1 activity at late M phase during cytokinesis , and this process is inhibited by phosphorylation of NACK1 and NPK1 by cyclin-dependent kinase (CDKs) before metaphase (Sasabe et al. 2011a). These kinesins have the MT-based motor domains at the aminotermini and the stalk regions in the carboxyl-terminal halves that are rich in several predicted coiled-coils structures, one of which provide binding sites for NPK1 MAPKKK, and they also consistently localize to the equatorial region of the phragmoplast during its expansion. Thus, NACK1 kinesin recruits NPK1 MAPKKK, which binds NQK1 MAP kinase kinase (MAPKK) and NRK1 MAP kinase (MAPK) to the phragmoplast equator, in addition it activates NPK1 and the whole kinase cascade . On the basis of the above results, the NACK-PQR MAPK pathway including these components is proposed to positively facilitate expansion of the phragmoplast (Soyano et al. 2003). Activated NRK1 MAPK phosphorylates MAP65 to reduce its activity of bundling MTs, which increases the rate of the MT turnover to allow phragmoplast MT arrays to expand centrifugally (Sasabe et al. 2006). Genetic analyses have shown that a similar pathway is also conserved in Arabidopsis thaliana Sasabe et al. 2011b;Takahashi et al. 2010). AtNACK1/HINKEL (HIK), an Arabidopsis homolog of NACK1, appears to be a key regulator of the progression of cytokinesis, because knockout and knockdown mutants of AtNACK1/HIK and/or AtNACK2/STUD (STD)/TETRASPORE (TES) show defects in cell plate structures and gametophytic lethality in Arabidopsis plants (Tanaka et al. 2004). The myc-tagged AtNACK1/HIK protein is also localized to the phragmoplast equator (Krupnova et al. 2013). In addition to Arabidopsis and rice (Sazuka et al. 2005), the genomes of all multicellular plants so-far examined encode for proteins that are homologous to NACK1 (Miki et al. 2014), suggesting a ubiquitous role of these homologues in cytokinesis. The cellular biological information of AtNACK1/HIK in Arabidopsis, however, is still limited. In the present study, we took advantage of the tobacco cell line BY-2, which is considered to be highly proliferative, to investigate the molecular and cellular bases for the characteristics of AtNACK1/HIK. We have investigated which regions of AtNACK1 are required for its localization to the equatorial zone of the phragmoplast corresponding to the site of cell plate formation. The results in the present study show that the C-terminal region of the stalk, as well as the motor domain, is necessary for the proper localization of AtNACK1/HIK to the phragmoplast equatorial zone. DNA constructs The deletion constructs were made by PCR and overlap PCR. Each fragment of AtNACK1 cDNA was amplified by PCR with specific primers and/or overlap primers (sequences of these primers will be sent on request) and cloned into the pENTR1A vector (Invitrogen, Carlsbad, CA). To express GFP fusions in BY-2 cells, the fragments of AtNACK1 cDNAs in pENTR were subcloned into the binary vector pGWB452 (Nakagawa et al. 2007) by using Gateway ® LR Clonase II (Invitrogen). All of these constructs were N-terminally GFP-fused. To express N-terminal GFP fusion in cells by estradiol, we introduced the sGFP-fused Gateway ® cassette from the pGWB6 vector (Nakagawa et al. 2007) into the pER8 vector that contains the estradiol-inducible promoter sequence (XVE sequence) (Zuo et al. 2000), which yielded the pER8GW6 vector. The full length cDNA fragment of AtNACK1 was also subcloned into the binary vector pER8GW6 by the Gateway system. To express the C-terminal GFP fusion in cells with estradiol, the AtNACK1-GFP fragment was amplified by PCR with the C-terminally GFP-fused AtNACK1 sequence cloned in pGWB451 (Nakagawa et al. 2007) as a template, and cloned into pER8 vector. The 6.1-kb DNA fragment covering the AtNACK1 locus from a position 1.3 kb 5′-upstream of the initiation codon to a position 0.75 kb 3′-downstream of the termination codon (Sasabe et al. 2011a) was used as a template to make a construct which expressed a GFP-fusion protein under the original promoter of AtNACK1. The 5.4-kb DNA fragments, excluding the 3′ UTR of AtNACK1, were amplified by specific primers and cloned into the pCR8 vector (Invitrogen). This fragment was sequentially subcloned into the binary vector pGWB504 (Nakagawa et al. 2007) to fuse sGFP to the C-terminal of AtNACK1. The sequences of all constructs were confirmed by sequencing analysis. Plant materials, transformations, and selection of transgenic lines Tobacco BY-2 cells were maintained in suspension culture at 26 °C in darkness with weekly subculture in modified Linsmaier and Skoog medium. Arabidopsis thaliana plants (Col-0) were grown with a 16-h photoperiod at 22 °C either in soil or on Murashige and Skoog (MS) plates that contained 0.8 or 1.5 % agar. Each construct was transformed into BY-2 cells and/or wild-type Col-0 plants by using the Agrobacterium tumefaciens-mediated method (An 1985;Clough and Bent 1998). More than 10 lines of transformants including each construct of BY-2 were used for observation of subcellular localization, and the offspring of seven independent T1 plants were used for subsequent analysis. The BY-2 cells transformed with transgenes under the control of an estradiol-inducible promoter were cultured in liquid medium containing 0.1 µM 17-β-estradiol (Sigma-Aldrich, St Louis, MO, USA) for 16 h to express GFP fusion. Microscopy BY-2 calli were suspended in liquid medium and used for observation. GFP fluorescence was detected with an AXIO Imager (Carl Zeiss, Oberkochen, Germany) and a confocal microscope, FV1000 (Olympus, Tokyo, Japan). Transgenic plants for microscopic observation were grown on MS plates containing 1.5 % agar. Plates were placed vertically in growth chambers for 7-14 days. To observe the dividing cells of root-tips in the 7-day-old transgenic plants, a 10 µM solution of FM4-64 (Invitrogen) was applied dropwise onto the root-tips, which were then incubated for 10 min. Fluorescence due to GFP and FM4-64 was sequentially recorded by confocal microscopy (LSM780; Carl Zeiss). Phenotypic analysis Leaves from 14-day-old plants grown on the MS plates were cleared, basically, as described by Ishikawa et al. (2003). Leaves were fixed overnight at room temperature in a mixture of ethanol and acetic acid (9:1, v/v). After rehydration in a graded ethanol series (90, 70, 50, and 30 % for 20 min each), they were cleared in a solution of chloral hydrate (trichloroacetaldehyde monohydrate, 80 g; glycerol, 10 ml; distilled water, 20 ml). Cleared leaves were observed with by Nomarski differential interference microscopy (AxioPlan2; Carl Zeiss). Results GFP-AtNACK1 localizes to the site of cell plate formation, expands to the parental cell walls, and disappears in living tobacco BY-2 cells As is the case for tobacco NACK1 , in silico analysis by means of the InterPro (http://www.ebi.ac. uk/interpro/) and COILS programs (Lupas et al. 1991) predicts with high probability the presence of a putative motor domain (amino acids 1-360) and several coiled-coil structures in the carboxyl-terminal (C-terminal) half, which is called the stalk region in NACK1, of AtNACK1/HIK (Fig. 1a, residues 361-974). In addition to the highly probable coiled-coil structures, a single coiled-coil region is predicted with lower probability in the C-terminal tail (Fig. 1a, residues 764-974). We made the XVE::GFP-AtNACK1 construct, in which the coding sequence for green fluorescent protein (GFP) was fused in frame to the amino-terminal (N-terminal) coding sequence of AtNACK1/HIK cDNA (Fig. 1b). This fusion gene was driven by the estrogen-inducible promoter XVE (Zuo et al. 2000) and introduced into cells of the tobacco cultured-cell line BY-2. We cultured the transformed BY-2 cells in liquid medium containing 0.1 µM 17-β-estradiol for 16 h to induce expression of GFP-AtNACK1. We observed subcellular localization of GFP-AtNACK1 in the transformed cells (Fig. 2), where it appeared in the central region corresponding to the expanding cell plate. The GFP signals disappeared as the formation of cell plates ended, suggesting that AtNACK1 disappears upon the maturation of cell plates. These results are consistent with the pattern of a sharp accumulation of tobacco NACK1 protein seen at the M phase of the cell cycle . The amino-terminal motor and the carboxyl-terminal region of the stalk of AtNACK1/HIK are required for its localization to the site of cell plate formation To examine which sequences of AtNACK1/HIK are involved in the localization to the site of cell plate formation, we made a series of deletion constructs of AtNACK1 cDNA (Fig. 1b), which contained the DNA sequence corresponding to the full length cDNA (full), the motor domain (MD), the stalk (ST) sequence, the MD-A sequence covering the MD and A regions in ST, the MD-AB sequence covering MD, the A and B regions in ST, the MD-AC sequence covering MD, the A and C regions in ST, and the MD-BC sequence covering the MD and BC regions in ST. Note that the B region contained the predicted binding sites of ANP1 MAPKKKs and the putative CDK phosphorylation sites Takahashi et al. 2010). These DNA constructs were fused to the C-terminus sequence of the GFP gene, whose expression was under the control of the constitutive active Cauliflower mosaic virus (CaMV) 35S promoter, and the chimeric DNA constructs were introduced into cells of the cultured tobacco cell line BY-2. We observed GFP signals during cytokinesis in transformed BY-2 cells (Fig. 3). In cells expressing GFP-AtNACK1:MD, no GFP signals were observed in either the phragmoplast region or the site of cell plate formation (Fig. 3b). Although we also transformed BY-2 cells with the GFP-AtNACK1:ST and obtained some calli, these calli were unable to grow further (data not shown), which might be due to the mislocalization of the motor-less NACK1:ST, because this observation is consistent with the result that the overexpression of the stalk region of tobacco NACK1 prevents NPK1 MAPKKK from locating to the phragmoplast equator and has a dominant-negative effect . Taken together, the motor domain seems to play a critical role in the AtNACK1 localization to the site of cell plate formation, but only the motor domain is not sufficient for its proper localization. GFP signals were not detected at the site of cell plate formation in the transformed cells with GFP-AtNACK1:MD-A, GFP-AtNACK1:MD-AB, and GFP-AtNACK1:MD-AC ( Fig. 3c-e). Obvious GFP signals were observed, however, at the cell plate formation site in the transformed BY-2 cells expressing GFP-AtNACK1:MD-BC (Fig. 3f). These results indicate that the BC region in ST, in addition to the motor domain of AtNACK1/HIK, is sufficient for the localization of AtNACK1 to the site of cell plate formation, but that only the AB region and the C region when separated from the B region are not sufficient for the appropriate localization. These results also indicate that the A region is not required for its localization to the site of cell plate formation. AtNACK1-GFP fusion does not localize to the site of cell plate formation in the living cells of Arabidopsis roots and causes dominant-negative effects on cytokinesis in Arabidopsis plants We fused the C-terminal coding sequence of the entire AtNACK1/HIK genomic sequence, which covered the Fig. 1 Domain and motif organizations of AtNACK1. a The probability of coiled-coil formation plotted against the amino acid number for AtNACK1 by using the COIL program and a 28-amino acid prediction window (Lupas et al. 1991). b Schematic representation of wild-type and deletion mutant constructs of AtNACK1/HIK. Predicted domain organization and relevant amino-acid positions for the AtNACK1 protein sequence are indicated in the wild-type schematic diagram. The coding region of the stalk (ST) is divided into three parts (ST-A, ST-B, and ST-C), which were fused to the region corresponding to the predicted motor domain (MD) and designated as AtNACK1:MD-A, MD-AB, MD-AC, and MD-BC. The coding sequence corresponding to each construct was fused to the C-terminus of the sequence of G3GFP or sGFP 5′-upstream promoter and coding regions, to the N-terminus of the GFP sequence. The genomic sequence we used has the ability to complement the phenotypes of atnack1 mutants (pAtNACK1::genome AtANCK1; Sasabe et al. 2011a). We used this construct to transform the wild-type Arabidopsis plants (Col-0). We observed the GFP signals by means of confocal laser scanning microscopy. By analyzing time-lapse images of living cells in the transgenic plants (Fig. 4a), we observed that the signals derived from pAtNACK1::genome AtNACK1-GFP did not localize at the site of cell plate formation, as visualized by staining with FM4-64, but were detected within wide areas corresponding to the whole phragmoplast in transformed root cells. The cell plate stained by FM4-64 in the transgenic Arabidopsis expressing AtNACK1-GFP appeared to expand more slowly than that in non-transgenic Arabidopsis (Fendrych et al. 2010;Kosetsu et al. 2010). These results indicate that the C-terminal GFP fusion of AtNACK1 appears to interfere with its localization to the site of cell plate formation, suggesting that the C-terminal region of AtNACK1/HIK plays an important role in its localization to the site of cell plate formation as well as in the cell plate formation itself. We observed the phenotypes of Arabidopsis plants transformed with pAtNACK1::genome AtNACK1-GFP (Fig. 4bd). In T2 plants, 6 out of 7 transgenic lines displayed a dwarf phenotype (Fig. 4b; 6-72 %), although the extent of the phenotype depended on each transgenic line. Epidermal cells in the #1 transgenic line, which displayed the highest proportion (72 %) of the most severe dwarf phenotype, exhibited cytokinesis defects with incomplete cell walls (Fig. 4c). Although the #8 transgenic plant line did not show dwarfism in T2 plants, the T3 plants homozygously expressing AtNACK1-GFP showed 8 % of dwarf phenotypes (Fig. 4d, nested image; 4 of 50 seedlings). The dwarf phenotypes might be due to a growth defect of the transgenic plants owing to the mislocalization of AtNACK1-GFP as described above. We also examined subcellular localization of AtNACK1-GFP in BY-2 cells and compared the pattern of its localization with that of the pattern of localization of AtNACK1-GFP in Arabidopsis. We made the XVE::AtNACK1-GFP construct, in which the GFP sequence was fused to the C-terminal coding sequence of AtNACK1 cDNA, and introduced into BY-2 cells. We cultured the transformed BY-2 cells in liquid medium containing 0.1 µM estradiol for 16 h to induce expression of AtNACK1-GFP. As compared with the GFP signal derived from XVE::GFP-AtNACK1 (Fig. 4e), those derived from XVE::AtNACK1-GFP were detected in the wide regions around the site of cell plate formation (Fig. 4f, g), which might have corresponded to the phragmoplast area. These results indicate that the C-terminal GFP fusion of AtNACK1 appears to disturb its localization to the site of cell plate formation in BY-2 cells as well as in Arabidopsis plants. We have also made the AtNACK1-GFP construct with a constitutive active CaMV 35S promoter and have attempted to analyze the subcellular localization of AtNACK1-GFP in tobacco BY-2 cells. Although we obtained transformed BY-2 cells, we failed to detect GFP signals. Discussion AtNACK1/HIK is localized to the cell plate formation site in the phragmoplast of tobacco BY-2 cells and degraded after cell plate formation The results of the present study with the GFP-AtNACK1/HIK construct have shown that the GFP-AtNACK1 protein is localized to the site of cell plate formation during cytokinesis in tobacco BY-2 cells (Fig. 2). The GFP signals disappeared when the signals reached the parental cell walls (Fig. 2), suggesting that the fusion protein degrades after cell plate formation, which is consistent with the previous results obtained by western blot analysis: tobacco NACK1 protein is accumulated specifically during M phase and disappears at the end of the M phase in synchronized BY-2 cells . The M-phase-specific accumulation partially depends on (Ito et al. 2001;Araki et al. 2004). The specific accumulation should also be due to the disappearance of NACK1 protein around the end of M phase, which is inhibited by the persistent CDK activity (Sasabe et al. 2011a) and the treatment with MG-132, a proteasome inhibitor (our unpublished data), suggesting that NACK1 is degraded by the action of the ubiquitin-proteasome pathway. AtNACK1 of Arabidopsis might be degraded by a similar mechanism at the end of cytokinesis in tobacco BY-2 cells (Fig. 2). The mechanism involved in this phasespecific degradation of NACK proteins should be further investigated experimentally. We also transformed Arabidopsis plants with the same GFP-AtNACK1 construct, and obtained sufficient numbers of transgenic plant lines. We, however, failed to find plants that were GFP-positive after induction with estrogen. The BC region in the stalk of AtNACK1/HIK is active in its binding to phragmoplast microtubules and localizing to the site of cell plate formation The sequences covering the motor domain and the BC region of the stalk of AtNACK1 (Fig. 1) are sufficient for its localization to the site of cell plate formation in the phragmoplast, but the A region is not required (Fig. 3). Characteristic features of the BC region in AtNACK1 are the presence of three predicted coiled-coil structures (4th, 5th, and 6th positions of coiled-coils in Fig. 1a), one of which overlaps with the site of interaction with ANP1 MAPKKK (corresponding to NPK1 in tobacco), a component of the PQR MAPK pathway that positively controls the formation of cell plates Kosetsu et al. 2010;Nishihama et al. 2002;Sasabe et al. 2011a;Takahashi et al. 2010). Considering the general concepts of configurations and dynamics of kinesin motors (Seeger and Rice 2010;Sablin 2000;Verhey and Hammond 2009), these coiled-coil segments together with the motor domain of AtNACK1 possibly play at least two critical roles in their specific localization. First, these coiled-coil structures might be involved in oligomerization of AtNACK1 kinesin, which might be required for their binding to microtubules and motility along the microtubules to move to the site for cell plate formation. The motility property is generated by coordinate stepping of at least two motor domains in the kinesin oligomers. Second, the coiled-coil structures of the BC region might provide binding sites for cargos to be transported along the microtubules, and the binding of the AtNACK1 kinesin motor to microtubules and its motility activity on the microtubule might be stimulated by the interaction with the PQR MAPK pathway similarly as by interactions between motors and cargos of other known kinesins. Kinesins, in general, bind to proper cargos at their non-motor regions, such as stalks and C-terminal tail regions, and transport the cargos along microtubule filaments to genetically determined destinations in specific subcellular compartments where they exert their precise functions (Verhey and Hammond 2009;Verhey et al. 2011). In the absence of cargos, kinesins are inactive, in terms of binding to microtubules and motility activities, due to their conformational repressive states, which have been recently defined as autoinhibition mechanisms (Hammond et al. 2010;Kaan et al. 2011;Verhey and Hammond 2009;Verhey et al. 2011). Since the complex of components in the PQR MAPK pathway binds to the B region of the AtNACK1 stalk Takahashi et al. 2010), the protein complex might release the autoinhibition of AtNACK1 to bind microtubules and activate its motility function. A well-characterized example is the cargo protein of the mammalian Kinesin-1 motor, which is the JIP scaffold protein that assembles the JNK MAPK pathway consisting of MAPKKK, MAPKK, and JNK MAPK and activates the Kinesin-1 motor (Blasius et al. 2007;Verhey et al. 2001). Activation of the JNK pathway seems to negatively regulate the interaction of kinesin motors with their cargos (Horiuchi et al. 2007), suggesting the presence of crosstalk between the MAPK signaling and trafficking pathways. It must be worth testing whether similar crosstalk between PQR MAPK and the vesicle trafficking systems modulated by AtNACK1 kinesin might be involved in the formation of cell plates. The C-terminus of AtNACK1/HIK is involved in the processive motility of AtNACK1/HIK on microtubules It has been reported that Myc-tagged AtNACK1/HINKEL proteins expressed under the original promoter are localized at the midzone of dividing cells in Arabidopsis plants (Krupnova et al. 2013). AtNACK1-GFP proteins that were synthesized in the Arabidopsis plant from the pAtNACK1::genome AtNACK1-GFP construct and in BY-2 cells from the XVE::AtNACK1-GFP construct were not clearly localized to the site of cell plate formation, however, but rather distributed over wide areas, which appear to correspond to phragmoplast microtubules, around the newly formed midzone (Fig. 4a, f, g). Considering the potential microtubule-binding activity and the motor activity of NACK proteins , the observed diffusive localization of AtNACK1-GFP might be due to inhibitory effects of GFP, which had been joined to the C-terminus of AtNACK1, on its motility activity; however, AtNACK1-GFP might still maintain the binding activity to microtubules because the GFP signals appear to localize on the phragmoplast area (Fig. 4a, f, g). Within the population of these transgenic plants expressing pAtNACK1::genome AtNACK1-GFP, we observed plants showing the dwarf phenotype and cytokinetic defects (Fig. 4b-d), suggesting that the AtNACK1-GFP protein has a dominant-inhibitory effect on wild-type AtNACK1 in the transgenic plants. The Arabidopsis mutation of RUNKEL exhibits severe cytokinetic defects and results in seedling lethal (Krupnova et al. 2009). Recently, it has been reported that the transgenic plants expressing truncated RUNKEL proteins cause the mislocalization of formation. Functions of NACK kinesins and their regulatory mechanisms in the formation of cell plates should be further investigated. Fig. 4 Subcellular localization of AtNACK1-GFP during the formation of cell plates in Arabidopsis plants and BY-2 cells, and effects of its expression on the plant phenotypes. a Time-lapse micrographs of AtNACK1-GFP in the living root-tip cell of an Arabidopsis plant that was transformed with genomic AtNACK1-fused GFP. Genomic AtNACK1 constructs were controlled under the original promoter. Confocal images were recorded at indicated times after the detection of cell plate formation. Fluorescence images show the signals due to magenta) and GFP (center, green). The merged images (merged) are shown in the right panels. Arrowheads indicate the site of expanding cell plates. Bar 10 µm. b Comparison of gross morphology of one 14-day-old wild type Col-0 plant and three T2 plants transformed with genomic AtNACK1-fused GFP. T2 plant line 1 (#1) showed the most severe dwarf phenotype. Bar 10 mm. c Cytokinetic defects in epidermal cells of leaves in transgenic plants. Fourteen-day-old Col-0 plants and T2 plants, line 1, were cleared and the epidermal cells in their leaves were observed. Arrows indicate the incomplete cell walls. Bars 10 µm. d Comparison of gross morphology of 46-day-old wild type Col-0 plant and T3 plant transformed with genomic AtNACK1-fused GFP. These plants were grown on soil under the long-day condition. T3 plant line 8 (#8) that homozygously expressed AtNACK1-GFP showed dwarf phenotypes at an efficiency of 8 %. Bars 10 mm. e-g Comparison of the patterns of subcellular localization of AtNACK1-GFP and GFP-AtNACK1 in BY-2 cells. Expression of GFP-AtNACK1 (e) and AtNACK1-GFP (f, g) were induced by 0.1 µM estradiol and fluorescence due to GFP was observed by fluorescence microscopy during cell plate formation. Fluorescence images (GFP), Nomarski images (DIC) and merged images (merged) in the cell exhibiting the representative patterns are shown. Images observed in transformed lines 16 (f, #16) and 43 (g, #43) are shown. Arrowheads indicate the sites of expanding cell plates. Bar 20 µm	2017-08-02T21:23:52.588Z	2014-12-14T00:00:00.000	{ "year": 2014, "sha1": "8b0519a15db50bff350428e05c2b23653183acf9", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/s10265-014-0687-2.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "8b0519a15db50bff350428e05c2b23653183acf9", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
264935299	pes2o/s2orc	v3-fos-license	Multi-Operational Mathematical Derivations in Latent Space This paper investigates the possibility of approximating multiple mathematical operations in latent space for expression derivation. To this end, we introduce different multi-operational representation paradigms, modelling mathematical operations as explicit geometric transformations. By leveraging a symbolic engine, we construct a large-scale dataset comprising 1.7M derivation steps stemming from 61K premises and 6 operators, analysing the properties of each paradigm when instantiated with state-of-the-art neural encoders.Specifically, we investigate how different encoding mechanisms can approximate expression manipulation in latent space, exploring the trade-off between learning different operators and specialising within single operations, as well as the ability to support multi-step derivations and out-of-distribution generalisation. Our empirical analysis reveals that the multi-operational paradigm is crucial for disentangling different operators, while discriminating the conclusions for a single operation is achievable in the original expression encoder. Moreover, we show that architectural choices can heavily affect the training dynamics, structural organisation, and generalisation of the latent space, resulting in significant variations across paradigms and classes of encoders. Introduction To what extent are neural networks capable of mathematical reasoning?This question has led many researchers to propose various methods to train and test neural models on different math-related tasks, such as math word problems, theorem proving, and premise selection (Lu et al., 2023; Meadows and Freitas, 2023; Mishra et al., 2022a; Ferreira et al., 2022; Ferreira and Freitas, 2020; Welleck 1 Code & data available at: https://github.com/neuro-symbolic-ai/latent_mathematical_ reasoning Figure 1: Can neural encoders learn to approximate multiple mathematical operators in latent space?Given a premise x, we investigate the problem of applying a sequence of latent operations (t 1 , . . ., t n ) to derive valid mathematical expressions (y 1 , . . ., y n ).et al., 2021;Valentino et al., 2022;Mishra et al., 2022b;Petersen et al., 2023).These methods aim to investigate how neural architectures learn and generalise mathematical concepts and symbolic rules, and how they cope with characteristic challenges of mathematical inference, such as abstraction, compositionality, and systematicity (Welleck et al., 2022;Mishra et al., 2022a). In general, a key challenge in neural mathematical reasoning is to represent expressions and formulae into a latent space to enable the application of multiple operations in specific orders under contextual constraints.Existing methods, however, typically focus on single-operational inferencei.e., optimising a latent space to approximate a specific mathematical operation (Lee et al., 2019;Lample and Charton, 2019;Welleck et al., 2022).Encoding multiple operations in the same latent space, therefore, remains an unexplored challenge that will likely require the development of novel mechanisms and representational paradigms. To investigate this problem, this paper focuses on equational reasoning, intended as the derivation of expressions from premises via the sequential application of specialised mathematical operations (i.e., addition, subtraction, multiplication, division, integration, differentiation).As derivations represent the workhorse of applied mathematical reasoning (including derivations in physics and engineering), projecting expressions and operators into a well-organised geometric space can unveil a myriad of applications, unlocking the approximation of mathematical solutions that are multiple steps apart within the embedding space via distance metrics and vector operations. Specifically, this paper posits the following overarching research questions: RQ1:"How can different representational paradigms and encoding mechanisms support expression derivation in latent space?";RQ2:"What is the representational tradeoff between generalising across different mathematical operations and specialising within single operations?";RQ3:"To what extent can different encoding mechanisms enable multi-step derivations through the sequential application and functional composition of latent operators?";RQ4:"To what extent can different encoding mechanisms support out-of-distribution generalisation?" To answer these questions, we investigate jointembedding predictive architectures (LeCun, 2022) by introducing different multi-operational representation paradigms (i.e., projection and translation) to model mathematical operations as explicit geometric transformations within the latent space.Moreover, by leveraging a symbolic engine (Meurer et al., 2017), we build a large-scale dataset containing 1.7M derivation steps which span diverse mathematical expressions and operations.To understand the impact of different encoding schemes on equational reasoning, we instantiate the proposed architectures with state-of-the-art neural encoders, including Graph Neural Networks (GNNs) (Hamilton et al., 2017;Kipf and Welling, 2016), Convolutional Neural Networks (CNNs) (Li et al., 2021;Kim, 2014), Recurrent Neural Networks (RNNs) (Yu et al., 2019;Hochreiter and Schmidhuber, 1996), and Transformers (Vaswani et al., 2017), analysing the properties of the latent spaces and the ability to support multi-step derivations and generalisation. Our empirical evaluation reveals that the multioperational paradigm is crucial for disentangling different mathematical operators (i.e., crossoperational inference), while the discrimination of the conclusions for a single operation (i.e., intraoperational inference) is achievable in the original expression encoder.Moreover, we show that architectural choices can heavily affect the training dynamics and the structural organisation of the latent space, resulting in significant variations across paradigms and classes of encoders. Overall, we conclude that the translation paradigm can result in a more fine-grained and smoother optimisation of the latent space, which better supports cross-operational inference and enables a more balanced integration.Regarding the encoders, we found that sequential models achieve more robust performance when tested on multistep derivations, while graph-based encoders, on the contrary, exhibit better generalisation to out-ofdistribution examples. Multi-Operational Derivations Given a premise x -i.e., a mathematical expression including variables and constants, and a set of operations T = {t 1 , t 2 , . . .t n } -e.g., addition, multiplication, differentiation, etc., we investigate the extent to which a neural encoder can approximate a mathematical function f (x, t i ; V ) = Y t i that takes the premise x and any operation t i ∈ T as inputs, and produces the set of valid expressions Y t i = {y 1 , y 2 , . . ., y m } derivable from x via t i given v ∈ V , where V is a predefined set of operands such that y j = t i (x, v j ).In this work, we focus on atomic operations in which V includes symbols representing variables. For example, consider the following premise x: If the set of operands is V = {z, u}, and the operation t i is addition, then the application of t i to x should result in the set of expressions: Notably, the recursive application of f to any of the expressions in Y t i can generate a new set of conclusions derivable from x in multiple steps.Here, the constraint we are interested in, is that Y add and Y dif f should be derived via a single expression encoder that maps expressions into a vector space and, at the same time, enables a multi-step propagation of latent operations. Architectures To model latent mathematical operations, we investigate the use of joint-embedding predictive architectures (LeCun, 2022).In particular, we introduce two multi-operational paradigms based on projection and translation to learn the representation of expressions and mathematical operators and model an atomic derivation step as an explicit geometric transformation.Figure 2 shows a schematic representation of the architectures. In general, both projection and translation employ an expression encoder to map the premise x and a plausible conclusion y into vectors, along with an operation encoder that acts as a latent prompt t to discriminate between operators.The goal is then to predict the embedding of a valid conclusion e y by applying a transformation to the premise embedding e x conditioned on t.Therefore, the two paradigms mainly differ in how expression and operation embeddings are combined to approximate the target results.This setup enables multi-step inference since the predicted embedding e ′ y can be recursively interpreted as a premise representation for the next iteration (Figure 2, right). Projection.The most intuitive solution to model latent mathematical operations is to employ a projection layer (Lee et al., 2019).In this case, the premise x and the operator t are first embedded using the respective encoders, which are then fed to a dense predictive layer π to approximate the target conclusion e y .The overall objective function can then be formalised as follows: Where δ is a distance function, and π represents the dense projection applied to the concatenation ∥ of t and e x .While many options are available, we implement π using a linear layer to better investigate the representation power of the underlying expression encoder. Translation.Inspired by research on multirelational graph embeddings (Bordes et al., 2013;Balazevic et al., 2019;Valentino et al., 2023), we frame mathematical inference as a multi-relational representation learning problem.In particular, it is possible to draw a direct analogy between entities and relations in a knowledge graph and mathematical operations.Within the scope of the task, as defined in Section 2, the application of a general operation can be interpreted as a relational triple < x, t, y >, in which a premise expression x corresponds to the subject entity, a conclusion y corresponds to the object entity, and the specific operation type t represents the semantic relation between entities.Following this intuition, we formalise the learning problem via a translational objective: Where δ is a distance function, e x , e y , t, are the embeddings of premise expression, conclusion and operation, and T is a diagonal operation matrix. Data Generation We generate synthetic data to support the exploration of the above architectures, inspired by a recent approach that relies on a symbolic engine to generate equational reasoning examples (Meadows et al., 2023b).In particular, we use SymPy (Meurer et al., 2017) to construct a dataset containing expressions in both LaTeX and SymPy surface forms. Here, premises and variables are input to 6 operations to generate further expressions, presently focusing on differentiation, integration, addition, subtraction, multiplication, and division.Concrete examples of entries in the dataset are reported in the Appendix. Premises.To generate a premise expression, a set of symbols is first sampled from a vocabulary.Subsequently, an initial operator is applied to symbols to generate an expression via the SymPy engine.To generate more complex expressions, this process is repeated iteratively for a fixed number of steps.This process is formalised in Algorithm 1 (see Appendix).The final dataset includes 61K premises each containing between 2 to 5 variables. Applying Operations to Premises.For a given premise, a set of operand variables (denoted by V in Section 2) are sampled from the vocabulary and added to the set of symbols that comprise the premise.All valid combinations of premise and operands are then input to each operator (via SymPy) to generate conclusions derivable via atomic derivation steps.The resulting dataset contains a total of 1.7M of such atomic steps.This data is used to train and evaluate models on singlestep inference before testing generalisation capabilities to multiple steps. Multi- Step Derivations.To test the models' ability to derive expressions obtained after the sequential application of operations, we randomly sample 5K premises from the single-step dataset described above and iteratively apply up to 6 operations to each premise using a randomly sampled variable operand from the vocabulary for each step.We adopt this methodology to generate a total of 2.7K multi-step examples. Expression Encoders Thanks to their generality, the multi-operational architectures can be instantiated with different classes of expression encoders.In particular, we experiment with both graph-based and sequential models, exploring embeddings with different dimensions (i.e., 300, 512, and 768).The graph-based encoders are trained on operation trees extracted from the SymPy representation, while the sequential models are trained on LaTeX expressions.We adopted the following expression encoders in our experiments: Graph Neural Networks (GNNs).GNNs have been adopted for mathematical inference thanks to their ability to capture explicit structural information (Lee et al., 2019).Here, we consider different classes of GNNs to experiment with models that can derive representations from operation trees.Specifically, we employ a 6-layer Graph-Sage2 (Hamilton et al., 2017) and Graph Convolutional Network (GCN) 3 (Kipf and Welling, 2016) to investigate transductive and non-transductive methods.To build the operation trees, we directly parse the SymPy representation described in Appendix A. Convolutional Neural Networks (CNNs). CNNs represent an effective class of models for mathematical representation learning thanks to their translation invariance property that can help localise recurring symbolic patterns within expressions (Petersen et al., 2023).Here, we employ a 1D CNN architecture typically used for text classification tasks (Kim, 2014), with three filter sizes 3, 4, and 5, each with 100 filters. Recurrent Neural Networks (RNNs).Due to the sequential nature of mathematical expressions, we experiment with RNNs that have been successful in modelling long-range dependencies for sentence representation (Yu et al., 2019;Hochreiter and Schmidhuber, 1996).In particular, we employ a Long-Short Term Memory (LSTM) network with 2 layers. Transformers.Finally, we experiment with a Transformer encoder with 6 and 8 attention heads and 6 layers, using a configuration similar to the one proposed by Vaswani et al. (2017) 4 .Differently from other models, Transformers use the attention mechanism to capture implicit relations between tokens, allowing, at the same time, experiments with a larger number of trainable parameters. Operation Encoders The operation encoders are implemented using a lookup table similar to word embeddings (Mikolov et al., 2013), where each entry corresponds to the vector of a mathematical operator.We experiment with dense5 embeddings for the translation model and instantiate the projection architecture with both dense and one-hot6 embeddings.The translation model requires the operation embeddings to be the same size as the expression embeddings, admitting, therefore, only dense representations. Training Details As the models are trained to predict a target embedding, the main goal during optimisation is to avoid a representational collapse in the expression encoder.To this end, we opted for a Multiple Negatives Ranking (MNR) loss with in-batch negative examples (Henderson et al., 2017).This technique allows us to sidestep the explicit selection of the negative sample, enabling a smoother optimisation of the latent space.We trained the models on a total of 12.800 premise expressions with 24 positive examples each derived from the application of 6 operations (see Section 2.2).This produces over 307.200 training instances composed of premise x, operation t, and conclusion y.The models are then trained for 32 epochs with a batch size of 64 (with in-batch random negatives).We found that the best results are obtained with a learning rate of 1e-5. Empirical Setup We evaluate the performance of different representational paradigms and expression encoders by building held-out dev and test sets.In particular, to assess the structural organisation of the latent space, we frame the task of multi-operational inference as an expression retrieval problem.Given a premise x, an operation t, a sample of positive conclusions P = {p 1 , . . ., p n }, and a sample of negative conclusions N = {n 1 , . . ., n m }, we adopt the models to predict an embedding e ′ y (Section 2.1) and employ a distance function δ to rank all the conclusions in P ∪ N according to their similarity with e ′ y .We implement δ using cosine similarity, and construct two evaluation sets to assess complementary inferential properties, namely: Cross-operational Inference.A model able to perform multi-operational inference should discriminate between the results of different operations applied to the same premise.Therefore, given a premise x and an operation t (e.g., addition), we construct the negative set N by selecting the positive conclusions resulting from the application of different operations (e.g., differentiation, subtraction) to the same premise x.This set includes a total of 4 positive and 20 negative examples (extracted from the remaining 5 operations) for each premise-operation pair (for a total of 3k dev and 6k test instances). Intra-operational Inference.While we want the models to discriminate between different operators, a well-optimised latent space should still preserve the ability to predict the results of a single operation applied to different premises.Therefore, given a premise x and an operation t, we construct the negative set N by selecting the positive conclusions resulting from the application of the same operation t to a different sample of premises.This set includes a total of 4 positive and 20 negative examples (extracted from 5 random premises) for each premise-operation pair (for a total of 3k dev and 6k test instances). Metrics.The models are evaluated using Mean Average Precision (MAP) and Hit@k.Hit@k measures the percentage of test instances in which at least one positive conclusion is ranked within the top k positions.MAP, on the other hand, measures the overall ranking.We use the average MAP between cross-operational and intra-operational sets (dev) as a criterion for model selection. Results Table 1 shows the performance of different encoders and paradigms on the test sets (i.e., evaluating the best models from the dev set, see Table 3).We can derive the following conclusions: The translation mechanism improves crossoperational inference.The models that use the translation method consistently outperform the models that use the projection method on the crossoperational inference task.This indicates that the translation paradigm can better capture the semantic relations between different operations and preserve them in the latent space.This is attested by the significant improvement achieved by different encoders, involving both graph-based and sequential architectures (e.g., +15.13% and +7.64% for Transformers and GCN respectively). Trade-off between cross-operational and intraoperational inference.The models that excel at cross-operational inference tend to achieve lower performance on the intra-operational set.This suggests that there is a tension between generalising across different operations and specialising within each operation.Moreover, the results suggest that intra-operational inference represents an easier problem for neural encoders that can be achieved already with sparse multi-operational methods (i.e., models using one-hot projection can achieve a MAP score above 90%). LSTMs and GraphSAGE achieve the best performance.LSTMs achieve the highest average MAP score, followed by GraphSAGE.These results demonstrate that LSTMs and GraphSAGE can balance between generalisation and specialisation, and leverage both sequential and graphbased information to encode mathematical operations.Moreover, we observe that graph-based models and CNNs tend to exhibit more stable performance across different representational paradigms (e.g., GraphSage achieve an average improvement of 2.3%), while LSTMs and Transformers achieve balanced results only with the translation mechanism (i.e., with an average improvement of 4.37% and 6.91% respectively). Model size alone does not explain inference performances.The Transformer model, which has the largest number of parameters, exhibits a lower average MAP score (with the projection mechanism in particular).This implies that simply increasing the model complexity or capacity does not guarantee better results (see Table 3 for additional details) and may compromise operational control in the latent space.This suggests that model architec- Intra-operation Figure 4: 2D projection of the latent space before and after an operation-specific transformation.The visualization supports the crucial role of the multi-operational paradigm for cross-operational inference, showing, at the same time, that intra-operational inference concerns larger regions and can be achieved in the original expression encoder.ture and the encoding method are more important factors for learning effective representations supporting multiple mathematical operations. Training Dynamics We conduct an additional analysis to investigate the training dynamics of different architectures.The graphs in Figure 3 show the typical trend for the MAP achieved at different epochs on different evaluation sets.Interestingly, we found that the projection and translation mechanisms optimise the latent space in a different way.The projection paradigm, in fact, prioritises performance on intraoperational inference, with a constant gap between the two sets.Conversely, the translation paradigm supports a rapid optimisation of cross-operational inference, followed by a more gradual improvement on the intra-operational set. This behaviour can help explain the difference in performances between the models.Specifically, since cross-operational inference is about disentangling operations applied to the same premise, we hypothesise it to require a more fine-grained optimisation in localised regions of the latent space.This optimisation can be compromised when priority is given to the discrimination of different premises, which, as in the case of intra-operational inference, involves a more coarse-grained optimisation in larger regions of the space. Latent Space Analysis We further investigate this behaviour by measuring and visualising the latent space in the original expression encoder (i.e., computing δ(e x , e y )) and after applying a transformation via the operation encoder (i.e., computing δ(e ′ y , e y )).In particular, Table 2 reports the average difference between the cosine similarity of the premises with positive and negative examples, a measure to estimate the latent space separation, and therefore, assess how dense the resulting vector space is. From the results, we can derive the following main observations: (1) The separation tends to be significantly lower in the cross-operational set, confirming that the latent space requires a more fine-grained optimisation in localised regions (Fig. 4); (2) Cross-operational inference is not achievable without operation-specific transformations, as confirmed by the impossibility to discriminate between positive and negative examples in the original expression encoders (i.e., δ(e x , e y ), Table 2); (3) The projection mechanism achieves intraoperational separation in the original expression encoders.This is not true for the translation mechanism in which the transformation induced by the operation encoder is fundamental for the separation to appear; (4) The latent space resulting from the translation model is more dense, with values for the separation that are generally lower when compared to the projection mechanism.These results, combined with the performance in Table 1, confirm that the translation paradigm can result in a more fine-grained and smoother optimisation which supports performance on crossoperational inference and a more balanced integration between expression and operation encoders. Multi-Step Inference We investigate the behaviour of different encoders and representational paradigms when propagating latent operations for multiple steps.To experiment, we employ the architectures recursively by interpreting the predicted target embedding e ′ y as a premise representation for the next step (see Fig. 2).In this case, we evaluate the performance using Hit@1, selecting 1 positive example and 4 negative examples for each premise and derivation step (2 for cross-operational and 2 for intra-operational). Figure 5 shows the obtained results.We found that the majority of the models exhibit a latent organisation that allows for a non-random propagation of latent mathematical operations.Most of the encoders, in fact, achieve performances that are significantly above random performance after 6 latent derivation steps (with a peak of 30% improvement for LSTM + translation).Moreover, while all the models tend to decrease in performance with an increasing number of inference steps, we observe significant differences between paradigms and classes of encoders.Most notably, we found that the performance of graph-based encoders tends to decrease faster, while the sequential models can obtain more stable results, in particular with the translation paradigm.The best translation model (i.e., LSTM) achieves a Hit@1 score at 6 steps of up to 50%, that is ≈15% above the best projection architecture (i.e., CNN). Length Generalisation Finally, we perform experiments to test the ability of expression encoders to generalise to out-ofdistribution examples.In particular, we focus on length generalisation which constitutes a notoriously hard problem for neural networks (Shen et al., 2021;Hupkes et al., 2020;Geirhos et al., 2020).To this end, we train the models on the subset of the training set containing premises with 2 variables and assess performance on longer premises (i.e., grouping the test set according to the number of variables).Figure 6 shows the results for different encoders using the translation mechanism. Overall, the results show a decrease in performance as expected, demonstrating, at the same time, a notable difference between encoders on cross-operational inference.In particular, the results suggest that graph-based models can generalise significantly better on longer premises, probably due to their ability to capture explicit hierarchical dependencies within the expressions.Among the sequential models, CNNs achieve better generalisation performance.We attribute these results to the convolution operation in CNNs which may help capture structural invariances within the expressions and allow a generalisation that is similar to GCNs. Discussion From the empirical evaluation, we can derive a set of takeaways for both the joint-embedding architectures and the specific expression encoders. Regarding the architectures, our analysis suggests that the translational paradigm can result in a more fine-grained and smoother optimisation of the latent space (Figure 3 and Table 2).This has the effect of improving multi-operational inference enabling a more balanced integration of different expression encoders, with an overall better tradeoff between cross-operational and intra-operational inference (Table 1).Moreover, we found that the translational paradigm can support better generalisation on multi-step inference when instantiated with sequential encoders such as Transformers, CNNs, and LSTMs (Figure 5), even when the encoders are only trained on single-step derivations. Regarding the specific encoders, we conclude that different models have different characteristics that should inform practitioners and future research in the field.Sequential models (i.e., Transformers, CNNs, and LSTMs), possess a better ability to organise the latent space for enabling latent multi-step derivations (Figure 5).Conversely, graph-based models are more efficient (i.e., they achieve better performance using smaller operation encoders, see one-hot in Table 1) and tend to generalise better to longer expressions when trained to simpler ones (see Figure 6). Related Work The quest to understand whether neural architectures can perform mathematical reasoning has led researchers to investigate several tasks and evaluation methods (Lu et al., 2023;Meadows and Freitas, 2023;Mishra et al., 2022a;Ferreira et al., 2022;Ferreira and Freitas, 2020;Welleck et al., 2021;Valentino et al., 2022;Mishra et al., 2022b;Petersen et al., 2023).In this work, we focused on equational reasoning, a particular instance of mathematical reasoning involving the manipulation of expressions through the systematic application of specialised operations (Welleck et al., 2022;Lample and Charton, 2019;Saxton et al., 2018).In particular, our work is inspired by previous attempts to approximate mathematical reasoning entirely in latent space (Lee et al., 2019).Differently from Lee et al. (2019), we investigate the joint approximation of multiple mathematical operations for expression derivation (Lee et al. (2019) explore exclusively the rewriting operation for theorem proving).Moreover, while Lee et al. (2019) focus on the evaluation of Graph Neural Networks (Paliwal et al., 2020)), we analyse the behaviour of a diverse set of representational paradigms and neural encoders.Our data generation methodology is inspired by recent work leveraging symbolic engines and algorithms to build systematic benchmarks for neural models (Meadows et al., 2023b,a;Chen et al., 2022;Saparov et al., 2023).However, to the best of our knowledge, we are the first to construct and release a synthetic dataset to investigate multi-step and multi-operational derivations in latent space. Conclusion This paper focused on equational reasoning for expression derivation to investigate the possibility of approximating and composing multiple mathematical operations in a single latent space.Specifically, we investigated different representational paradigms and encoding mechanisms, analysing the trade-off between encoding different mathematical operators and specialising within single operations, as well as the ability to support multistep derivations and out-of-distribution generalisation.Moreover, we constructed and released a large-scale dataset comprising 1.7M derivation steps stemming from 61K premises and 6 operators, which we hope will encourage researchers to explore future work in the field. The systematic application of mathematical operators requires reasoning at an intentional level, that is, the execution and composition of mathematical functions defined on a potentially infinite set of elements.Neural networks, on the contrary, operate at an extensional level and, by their current nature, can only approximate such functions by learning from a finite set of examples. Due to this characteristic, this work explored architectures that are trained on expressions composed of a predefined number and set of variables (i.e., between 2 and 5) and operators (i.e., addition, subtraction, multiplication, division, integration, differentiation), and, therefore, capable of performing approximation over a finite vocabulary of symbols.Extending the architectures with a new set of operations and out-of-vocabulary symbols, therefore, would require re-training the models from scratch.Future work could investigate this limitation by exploring, for instance, transfer learning techniques and more flexible neural architectures. For the same reason, we restricted our investigation to the encoding of atomic operations, that is, operations in which the second operand is represented by a variable.While this limitation is circumvented by the sequential application of operators in a multi-step fashion, this work did not explore the encoding of single-step operations involving more complex operands (e.g., multiplication between two expressions composed of multiple variables each).In principle, however, the evaluation presented in this work can be extended with the new synthetic data to accommodate and study different cases and setups in the future. A Data Generation Algorithm 1 formalises the general data generation methodology adopted for generating premises with the SymPy 7 engine.The following is an example of an entry in the dataset with both LaTex and Sympy surface form for representing expressions, considering integration and a single variable operand r.The same overall structure is adopted for the remaining operations and a larger vocabulary of variables: • Premise: -SymPy: Add(Symbol( ′ u ′ ), cos(log(Add(M ul(Integer(−1), 7 https://www.sympy.org/en/index.html Figure 2 : Figure 2: Overview of the proposed joint-embedding predictive architectures for latent multi-operational derivation (left).Schematic workflow for multi-step inference and latent propagation of mathematical operations (right). Figure 3 : Figure 3: Typical training dynamics of different multi-operational paradigms (MAP on the dev set). Figure 5 :Figure 6 : Figure 5: Multi-step derivations in latent space with different multi-operational paradigms and neural encoders. Table 1 : Overall performance of different neural encoders and methods for encoding multiple mathematical operations (i.e., integration, differentiation, addition, difference, multiplication, division) in the latent space. Table 3 : Table3reports the complete results on the dev set for different models and architectures with different embedding sizes.Overall performance of different neural encoders and methods (dev set) for jointly encoding multiple mathematical operations (i.e., integration, differentiation, addition, difference, multiplication, division).	2023-11-03T06:41:28.096Z	2023-11-02T00:00:00.000	{ "year": 2023, "sha1": "c22cc4e2eed78b4b31e50d94ea35da0405aabb87", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "ArXiv", "pdf_hash": "206cb9b9ee2d21154b3027aa84f03d66a8922091", "s2fieldsofstudy": [ "Computer Science", "Mathematics" ], "extfieldsofstudy": [ "Computer Science" ] }
212970695	pes2o/s2orc	v3-fos-license	The concept of "justice" in the legislation of Russia and France, implementation features The relevance of the study is determined by the prevailing variety of approaches to understanding the essence of the concept of “justice”, which is characteristic of both Russia and France. Such variability leads to a persistent distortion of this concept in practice of law enforcement, which, in turn, affects the effectiveness of administration of justice. The aim of the authors is to study scientific theoretical views and practical issues associated with the implementation of the principle of justice in two states. In the work we used following methods: dialectics, analysis, synthesis, deduction, as well as the formal legal and comparative legal method.Based on the analysis of the doctrine, legislation and judicial practice of two states, Russia and France, the following conclusions are drawn. At present, “justice” is understood as a universal notion, which is not limited only by coverage of regulatory norms of the law, but is actively applied by judicial practice. The use of this concept allows the court to make fair decisions depending on specific circumstances of the case, thereby achieving a balance of interests of participants in the process by interpreting and clarifying the law, and in some cases by creating a new rule of law that allows the courts to ensure the effective implementation of the principle of justice. However, in order to avoid variability in understanding this notion, it is proposed to fix the concept of “justice” in the civil procedure code, this will reduce the percentage of judicial conflicts and will contribute to formation of a uniform judicial practice. Introduction The 21st century is characterized by rapid scientific and technological progress, contributing to a large-scale change in social relations and public outlook. Advanced ideas come to life, thereby changing the socio-economic order in society. The legislators are actively working on regulation of new relations, carrying out reforms in most branches of law. New trends challenge scientists to comprehend justice as a general regulator of social relations, where justice appears to be an incentive for the effectiveness of human development. Review of research literature Justice is a multidimensional and controversial notion. It is the subject of research in various humanities (philosophy, psychology, sociology, political science, economics, law). The idea of justice was pondered by Plato, Aristotle, F. Bacon, I. Bentham, M. Sandel and others. Particularly noteworthy is the work of the philosopher-political scientist John Rawls, "Theory of Justice," where justice is seen as a socio-political category and expressed in honesty in fulfilling its obligations. Thus, in the author's opinion, a particular practice should be declared fair if it complies with the principles, which promotion and adoption would be reasonable to expect from all participants, if they were in the same circumstances and had to take firm commitments in advance, without knowing what will be the special conditions they will find themselves in the future [1]. However, D. Rawls does not consider justice as a legal category, law in his understanding is only a phenomenon that formalizes justice [1]. In Russia, such scholars dealt with issues of correlation of the category of justice with the main categories of ethics as V.S. Soloviev, O.G. Drobnitsky, E.L. Dubko, A.P. Butenko et al. In the field of law this issue was discussed in the works of G.F. Shershenevich, L.I. Petrazhitsky, S.S. Alekseev, I.V. Chechelnitsky, I.I. Andrianovskaya, V.A. Waipan et al. G.F. Shershenevich analyzed relations of sense of law and justice, L.I. Petrazhitsky studied the ratio of justice and law-making. S.S. Alekseev believed that justice is acquiring significance of a legal principle to the extent that it is embodied in the regulatory method of regulation, in those principles of proportionality, equality, etc., which are inherent in the very construction of a legal doctrine. A. Vaipan understands justice as universal measure and criterion of human behavior and understanding of the world, moral foundation and means of ensuring law and lawmaking. Any scientific discussion, the author believes, addresses issues of justice (distribution of income and property, taxes, punishment, redress, etc.), without understanding and applying the principle of justice, complex general legal problems of legal understanding, legal awareness, law-making and law enforcement are not solved. Among French scholars who studied the category of justice, one can distinguish S.-L. Montesquieu, J.-J. Russo, L. Boltanski, L. Theveno, J. Ranciere, A. Badiou, S.-K. Colm As A. Badiu correctly notes, at present a special place is occupied by philosophical thought, which makes up a single knot with a political condition. He criticizes modern understanding of democracy; people, he believes, do not strive at all to be virtuous, their private interests clash, the principles of political truth are distorted. This creates new conditions for a philosophy whose goal is to achieve justice [2]. Undoubtedly justice is a moral category that is inherent in both politics and law. The idea of justice has always occupied a special place in the system of human values and has become a fundamental category in the legal consciousness of society. The very Latin origin of the term justice (justicia) consists of the words "law and justice", and lawyers (judges) are its representatives. The purpose of the work is to search for meanings and mechanisms for understanding the legal essence of the category of "justice", to actualize the problematic issues that have arisen during implementation of principle of justice in Russia and France. The novelty of the study lies in formulation, justification and solution of problems aimed at determining the essence of the category of "justice", comprehensive study of this category in the legislation and judicial practice of Russia and France. Methods Dialectics, analysis, synthesis, deduction, formal legal method, comparative legal. Dialectics allowed us to push away from general theoretical ideas and determine the place of justice in law. Analysis and synthesis made it possible to identify problems of legislation and judicial practice in implementation of the principle of justice, as well as give it a definition. The use of the formal legal method in the study made it possible to describe and generalize the category of "justice". The comparative legal method made it possible to compare the legislation of the two countries, to identify similarities and differences between them. Results The legal expression of the requirement of justice is contained in the Rome Convention on the Protection of Human Rights and Fundamental Freedoms of 1950 (hereinafter referred to as the Convention), including in relation to the activities of the court. So, Art. 6 establishes that everyone, in case of a dispute over civil rights and obligations or any criminal charge, has the right to a fair and public hearing within a reasonable time by an independent and impartial court established by law, and Art. 41 secures the right to fair compensation. Russia and France are parties to this convention. France actively participated in development of the Convention and signed it among the first. However, France ratified this document quite late, in 1974, France recognized the right to individual appeal and compulsory jurisdiction of the European Court of Human Rights only in 1981 [3]. The Russian Federation ratified the Convention on March 30, 1998. The generally recognized principles and norms of international law and international treaties are an integral part of the legal systems of Russia and France. The right to a fair trial, enshrined at the level of an international convention, obliges the national legislators to perceive it as a valid value, and the judiciary to form practice in development of this institution [4]. The question arises, how the conventional principle of justice is reflected in the norms of branches of law and in the acts of law enforcement in Russia and France. The French Constitution of October 4, 1958 mentions justice in Art. 4. It states that "the law guarantees pluralistic expression of opinions and fair participation of parties and political associations in the democratic life of the Nation." The Constitution of the Russian Federation mentions justice in its preamble "honoring the memory of the ancestors who conveyed to us love and respect for the Fatherland, faith in goodness and justice." In the French Civil Code (Napoleon's Code) of 03/21/1804, the term "justice" is found in Art. 270, p. 545, Art. 565, Art. 815-13, Art. 1135, Art. 1579 -"fair reparation", "natural justice", "rules of justice". The reform of French contract law (ordinance of February 10, 2016) allowed to strengthen flexibility of the contractual mechanism as well as contractual justice. As Remy Cabriac, a professor at the University of Montpellier, notes, the essence of the reform was to adapt the law to new social realities and new contractual relations, and the desire for contractual justice is a red thread through his many short stories, for example, the restoration of contractual balance in favor of an economically weaker (or considered to be such ) side [5]. The Civil Code of the Russian Federation (hereinafter referred to as the Civil Code of the Russian Federation) mentions justice in paragraph 2 of Art. 5 "if it is impossible to use an analogy of the law, the rights and obligations of the parties are determined on the basis of the general principles and meaning of civil law (analogy of law) and requirements of good faith, reasonableness and justice." It follows from this article that the application of these categories to resolve a disputed legal relationship is subsidiary in case that there is no legal norm that directly regulates this legal relationship or a norm regulating similar relations (analogy of the law). However, justice is not only a subsidiary category. As O.I. Tsibulskaya, notes, the principle of justice has a normative and evaluative nature, is embedded in the very content of the law and is embodied in rights and obligations [6]. From the standpoint of civil liability justice is understood as a representation of compliance with social ideals (parameters) of distribution of losses, other adverse consequences between participants in the legal relationship in connection with nonperformance or improper performance of the contract, harm, dishonest behavior, etc. [7]. So, paragraph 5 of Art. 393 of the Civil Code of the Russian Federation establishes the duty of a judge to independently determine the amount of losses when other circumstances are proved. As noted by S.L. Degtyarev, the need to comply with the principle of justice and proportionality of liability by the court in determining the amount of losses is nothing more than a manifestation of the general principle, the principle of full compensation for losses established by paragraph 1 of Art. 15 of the Civil Code of the Russian Federation, which declares that it is necessary to remember that according to this principle the victim has the right to demand full compensation for losses, but does not have the right to demand more than was lost [8]. Thus, despite the absence of a formal expression of justice as a principle, its "invisible" presence in the provisions of the laws of Russia and France is aimed at achieving fair legal regulation that reflects the laws of social development and the needs of society. In the French Criminal Code, the principle of justice is formulated in Art. 132-24 (chapter "Punishment regime"): "the court imposes a punishment and determines the regime for its execution, depending on the circumstances of criminal act and identity of the performer. In imposing a fine, the court determines its size, taking into account the income and expenses of the perpetrator." This principle expresses requirement of individualization and the amount of punishment, is also characteristic of Russian criminal law (principle of justice, art. 6, part 3 of art. 60 of the Criminal Code of the Russian Federation (hereinafter referred to as the Criminal Code of the Russian Federation). Legal consolidation of the principle of justice received in Art. 6 of the Criminal Code, which established that punishment and other measures of a criminal law nature applied to the person who committed the crime must be fair; they must correspond to the nature and degree of public danger of the crime, the circumstances of its commission and the identity of the perpetrator. Also paragraph 3 of Art. 43 of the Criminal Code of the Russian Federation indicates "restoration of social justice", and paragraph 1 of Art. 60 indicates "fair punishment". The concept of "justice" is also found in the norms of a number of articles of the Code of Criminal Procedure of the Russian Federation (hereinafter referred to as the Code of Criminal Procedure), in particular in paragraph 2 of Article 6, paragraph 4 of article 226.9, Art. 297, Art. 332, Art. 389.9, Art. 389.28 -"fair punishment", "fair sentence", etc. The Arbitration Procedure Code of the Russian Federation (hereinafter referred to as the Arbitration Procedure Code of the Russian Federation) includes such phrases as "fair public trial" (paragraph 3 of article 2, paragraph 1. of article 308.10), "subject to the requirements of reasonableness and justice" (paragraph . 2 Article 97), "on the basis of the principles of justice, reasonableness and inadmissibility" (paragraph 4 of Article 174). The Civil Procedural Code of the Russian Federation (hereinafter referred to as the Code of Civil Procedure of the Russian Federation) mentions the term justice in two articles in the context of a "fair trial" (paragraph 3 of article 206 and article 391.11), in particular, "the court, at the request of the plaintiff, is entitled to award a sum of money in its favor to be recovered from the defendant in case of non-execution of a judicial act, in an amount determined by the court on the basis of the principles of fairness, proportionality and inadmissibility of profit from illegal or unfair behavior (Clause 3, Article 206, this clause was introduced by the Federal Law on 03.08.2018 N 34). In the process of working on the concept of the Unified Code of Civil Procedure, it was proposed to consolidate the principle of justice in it. The developers of the project were guided by the recommendations of the Constitutional Court of the Russian Federation and the European Court of Human Rights. This proposal caused heated discussions among jurists. Some lawyers believed that such recommendations to introduce a new principle of the branch of law are clearly not enough. And they noted earlier the need to address a number of important questions about the scope of its action, about the subjects it is designed on, about the content, about guarantees of implementation [9]. Thus, a review of theoretical views on the category of justice and the results of monitoring the legislation of the two countries led to the conclusion that there is an ambiguous understanding of its essence and content in law. The episodic presence of the category of justice in the legislation of Russia and France is noted, which, instead of clarity and unambiguity, generates more legal uncertainty. Such a problem is common to all branches of law of the states under consideration; their relevance is significantly increasing in the light of changing social relations. Let us turn to judicial practice, it reflects the implementation of the principle of justice in civil cases. Fair justice refers to fundamental rights, providing each entity with the effective restoration of its violated rights. In this regard, requirement of fairness should cover the entire trial. In procedural legislation, guidelines for judicial practice, both in Russia and in France, are defined quite clearly: justice is ensured by its accessibility, correct establishment of the actual circumstances of the case, their correct legal assessment, correct interpretation of the applicable norm, which always applies in the system of other norms. This is possible provided that the legal procedures for the consideration of the case, in particular the procedural time limits, as well as the democratic principles of legal proceedings, and above all its implementation by an impartial and independent court on the basis of the adversarial and equal rights of the parties, are observed [10] Discussion However, as Andrei Yegorov noted in his speech on August 5, 2014, the courts referred to the principle of justice in approximately 77,000 judicial acts, and to honesty in 15,000. The author notes that there are still very few scientific works in Russian civil engineering explaining the essence of the principle of justice [11]. The ambiguity of understanding and interpretation of justice is inherent in the Constitutional Court of the Russian Federation. So, according to V.A. Wipman, the analysis of judicial acts confirms the infinite number of understandings of justice used by judges in making decisions, as well as significant subjectivity in assessing certain circumstances of cases through the prism of justice. Judges sometimes use this term only to enhance the significance of their decision, the author believes, and the use of this category does not carry any semantic load and creates even greater legal uncertainty. He notes that such acts as the "constitutional principle of justice", "the requirement of justice", "the principle of justice", "the general principle of justice", and "social justice" are found in acts of the Constitutional Court of the Russian Federation [1]. And as the author rightly notes, the Constitutional Court is responsible for the resulting legal uncertainty in this matter, since the lack of a single, namely constitutional and legal approach entails confusion in legal understanding of the concept of "justice" [12]. Conclusion 1. Despite the absence of a formal expression of justice as a principle, its "invisible" presence in provisions of the laws of Russia and France is obvious. The single result is aimed at achieving fair legal regulation that reflects the laws of social development and the needs of society, which, in fact, is the social essence of legal regulation. 2. Review of theoretical views on the category of "justice" and the results of monitoring the legislation of the two countries led to the conclusion about an ambiguous understanding of its essence and content in law. There is an occasional mention of justice in the legislation of Russia and France, which, instead of clarity and unambiguity, generates more legal uncertainty. Such a issue is common to all branches of law of the states under consideration. 3. An unequivocal understanding of the essence of the principle of justice as a legal category has not developed either in the theory of law or in the law enforcement practice of Russia and France. This leads to a persistent distortion of understanding and a different interpretation of justice, and hinders effective administration of justice and achievement of the goal of legal proceedings namely protection and restoration of violated civil rights. 4. Justice is seen as a universal category, covering the regulatory norms of the law as well as allowing the court to make a fair decision depending on specific circumstances of the case in order to achieve a balance of interests of the participants in the process by interpreting and clarifying the law, and in some cases by creating a new rule of law that allows courts to enforce the principle of justice. In this regard, it seems necessary to consolidate the concept of "justice" in the civil and civil procedural legislation of states; this will allow avoiding contradictions in law enforcement practice.	2019-12-05T09:05:37.069Z	2019-09-01T00:00:00.000	{ "year": 2019, "sha1": "87898438b00848f45fb4650078a40242293df05e", "oa_license": "CCBY", "oa_url": "https://www.e3s-conferences.org/articles/e3sconf/pdf/2019/61/e3sconf_itese18_04006.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "ec2d95b64578ad431fc874a2846abfc94de9d7c4", "s2fieldsofstudy": [ "Law" ], "extfieldsofstudy": [ "Political Science" ] }
267809653	pes2o/s2orc	v3-fos-license	Pedigree genome data of an early-matured Geng/japonica glutinous rice mega variety Longgeng 57 By using PacBio HiFi technology, we produced over 700 Gb of long-read sequencing (LRS) raw data; and by using Illumina paired-end whole-genome shotgun (WGS) sequencing technology, we generated more than 70 Gb of short-read sequencing (SRS) data. With LRS data, we assembled one genome and then generate a set of annotation data for an early-matured Geng/japonica glutinous rice mega variety genome, Longgeng 57 (LG57), which carries multiple elite traits including good grain quality and wide adaptability. Together with the SRS data from three parents of LG57, pedigree genome variations were called for three representative types of genes. These data sets can be used for deep variation mining, aid in the discovery of new insights into genome structure, function, and evolution, and help to provide essential support to biological research in general. Background & Summary In recent years, the planting area for rice (Oryza sativa L.) in Heilongjiang (HLJ) province of China has increased to around 4 million ha 1 .For this global largest planting region for early Geng/japonica rice, which is about 2.6 times larger than the rice planting area of Japan 2 , determining how to transfer its advantages in agriculture to other branches of the economy remains a significant challenge for agriculture researchers. Early-matured Geng/japonica varieties provide the base for food security 3 , and supply critical agro-industrial materials, especially glutinous varieties.Glutinous rice, also called sticky rice, is becoming increasingly popular because of growing public awareness of health issues 4 .Glutinous rice has health benefits in managing diabetes, inhibiting chronic diseases, enhancing digestion, and reducing inflammation 5 .In addition to being an elite cooking material for a low gluten diet and 'good food' 6 , glutinous rice also provides raw materials for environment-friendly industry [7][8][9] .Longgeng 57 (LG57), a glutinous early variety, has favorable quality and stable-yield behavior in the early Geng/japonica planting region; therefore, it is now planted over more than 120,000 ha per year on average. Grain quality traits of rice are largely controlled by major genes, such as Waxy for the amylose content and OsNramp5 for the mineral nutritional quality [10][11][12] .Thus, further improvement of grain quality of glutinous rice, e.g., LG57, also requires more genome information. Currently, joint analysis has become a trend in biotechnology-based rice breeding in HLJ.For example, the Rice Molecular Breeding (RMB) laboratory from the Institute of Crop Science (ICS), Chinese Academy of Agricultural Sciences (CAAS), has set up a genome-based breeding scheme with the aid of both core germplasms of 3K-RG 13 , and the Rice Functional Genomics Breeding (RFGB) information platform 14 .It also widely cooperates with local research institutes from HLJ, including Jiamusi Rice Research Institute (JMS-RRI) and Suihua RRI (SH-RRI) 3 .Herein, we present a dataset from a collaboration between the RMB laboratory and JMS-RRI for early-matured Geng/japonica including LG57.Information based on this dataset for certain target genes, such as Waxy and OsNramp5, were also included as examples for data validation.This dataset comprises more than 770 Gb of pedigree genome data that will be useful for researches in general. Methods Plant material and library construction.The early-matured Geng/japonica variety Longgeng57 (LG57) was developed by our own and licensed to be released in 2017 and is now one mega variety with multiple elite traits and widely planted (more than 120,000 hectare per year) in Heilongjiang province in Northeast of China.High-molecular-weight genomic DNA was extracted from 10-day-old leaves of LG57 pedigree members (multiple seeds) with modified CTAB method followed by 0.5x bead purification for twice.The DNA sample through the qualification processes by both 0.75% agarose gel assay and Nanodrop was quantified with Qubit.Then the sample of LG57 met the standard was submitted to the constructions of PacBio HiFi library for long-read sequencing (LRS).Samples of three parents (Longnuo 2 (LN2), Punian 8 (PN8), and Longgeng 29 (LG29)) were submitted to construct Illumina libraries short-read sequencing (SRS) (Fig. 1). Genomic data were generated for all pedigree members, as listed in Table 1.Among them, PacBio (Menlo Park, CA, USA) protocols were adopted for long-read sequencing of LG57 and Illumina (San Diego, CA, USA) protocols were used for short-read sequencing.The details are as follows. DNA sample testing.DNA extraction from samples was carried out using a routine method that met the quality standard required for sequencing according to a previous study 3 .Sample purity and quantity were detected using a Nano Photometer ® (IMPLEN, Westlake Village, CA, USA) and a Qubit ® 3.0 Fluorometer (Life Technologies, Carlsbad, CA, USA), respectively, in combination with Agarose electrophoresis (concentration 1%, voltage 120 V for 45 min).Library construction and Inventory inspection.Covaris ® g-TUBE 15 was used to break the genomic DNA into suitable large pieces.Magnetic beads were then used for enrichment and purification.SageELF (Sage, Newcastle upon Tyne, UK) was adopted to screen and purify the DNA fragments.An Annoroad ® Universal DNA Library Prep Kit V2.0 (Annoroad Gene Technology, Beijing, China) was used for sample preparation, including end repair and ligation addition. To ensure the quality of the library, a three-step quality check procedure was adopted as follows.After the library was constructed, the Qubit 3.0 was used for preliminary quantification.Then, the library was diluted to 1 ng/μL and the insert size was checked using an Agilent 2100 instrument (Agilent, Santa Clara, CA, USA).The effective concentration of the library was accurately quantified using quantitative real-time reverse transcription PCR (qRT-PCR) in a Bio-Rad CFX96 PCR instrument with a Bio-Rad IQ SYBR GRN Kit (both Bio-Rad, Hercules, CA, USA). Sequencing.The single-molecule real-time (SMRT) method was adopted for the long-read sequencing (LRS) according to standard method (PacBio).Short-read sequencing (SRS) was carried out on the NovaSeq 6000 S4 platform (Illumina) to obtain a 250 bp double-ended sequencing reads.Genome assembly, validation and annotationFor the LRS data obtained by HiFi library sequencing, the raw data (subreads) from the PacBio sequencing was filtered by using SMRT link v9.0.0.92188 (https://www.pacb.com/support/software-downloads/) with default parameters to obtain high-quality circular consensus sequences (CCS) data.For the assembly, hifiasm 16 with default parameters were employed based on the CCS data.Merqury 17 was adopted for the quality check of LG57 assembly.Also, BUSCO (Benchmarking Universal Single-Copy Orthologs) 18 was used for genome assembly quality assessment.BUSCO analysis with default parameters was carried out using a single-copy gene set of several large evolutionary branches based on the OrthoDB (http:// cegg.unige.ch/orthodb).The gene set was compared with the assembled genome using embryophyta_odb10, and the accuracy and completeness were assessed based on the proportions and completeness of the alignment. Based on the LG57 assembly, two strategies were adopted for genome annotation.The first was a homologous strategy.RepeatMasker with default parameters 19 based on RepBase 20 was used to annotate repeats.For gene structures, BLAST 21 with evalue = 1e-5 and GeMoMa 22 with default parameters were used.Prediction of rRNAs, snRNAs, and miRNAs was carried out by aligning the assembly with known non-coding RNA libraries, e.g., Rfam 23 .The second was a de novo strategy.For repeat analysis, RepeatModeler (https://www.repeatmasker.org/RepeatModeler/) with -engine ncbi was adopted.For protein-coding gene prediction, Augustus 24 with-genemodel = partial, SNAP(https://github.com/KorfLab/SNAP), and GeneMark 25 with default parameters were adopted.Based on the above predictions, EVidence Modeler (EVM) 26 with default parameters was used to integrate the gene sets predicted by various strategies into a non-redundant gene set.The resulting predictive gene set was compared with various functional databases using UniProt 27 , NCBI (https://www.ncbi.nlm.nih.gov/nucleotide/), PFAM 28 , eggNOG 29 , GO (gene ontology) 30 , and KEGG (Kyoto Encyclopedia of Genes and Genomes) 31 .For tRNA sequence prediction, we used tRNAscan-SE 32 with parameters of -X 20 and -z 8.The SRS data were aligned to the reference genome and variations were called using a pipeline comprising BWA 33 , SAMtools 34 , and GATK 35 with default parameters, with Nipponbare IRGSP 1.0 36 as the reference genome. Data Records The assembly of LG57 is accessible at NCBI through GenBank 37 or the following accession ID of JAXQPT000000000 37 .Additionally, the raw read data for LG57 in the bam format are also available with accession number of SRR25376496 38 .Other sequencing pedigree genomic data for parents of LG57, including PN8 (SRR24688636) 39 , LN2 (SRR24688637) 40 , and LG29 (SRR24688635) 41 .Annotation data for LG57 are accessible through figshare 42 .All above data except for the bam files are also accessible in RFGB website (https://rfgb.rmbreeding.cn/download/publicDataDownload/download?dataset=3). technical Validation A total 1,671,418 of reads were obtained.The averaged read-length is 16,831.42bp and N50 value is more than 17 Kb.The distribution of these reads was shown in Fig. 2. A rough assembly for LG57 was carried out.A quality checking for the assembly of Longgeng 57 was also carried out by using Merqury and BUSCO.Based on the output of Merqury, the completeness of assembly was 99.5% and the QV was 62.0 ( 36 and 1 represents the first alternative genotype (ALT). in Table 3, N50 of contig has arrived at more than 27 Mb, which is over 10 times of our previous work with SJ18 3 .As shown in Table 4, a total of 1614 groups were searched by BUSCO, the complete groups accounted for about 98.8%.Functional genes predicted in LG57 comparing with those from databases were shown in Table 5. Identified by RepeatMasker, the total length of the repeat sequences is approximately 170MB, accounting for 43.13% of the whole LG57 genome (Table 6).Prediction results of different types of non-coding RNA including miRNA, tRNA, rRNA, and snRNA were listed in Table 7.These RNAs together accounting for 81.3% of the LG57 genome.We also compared the parameters of LG57 to the other assemblies.Averaged gene length of LG57 is longer than those of the others (Table 8). For the SRS data of the three parents (LN2, PN8, and LG29), we firstly aligned them against reference genome IRGSPv1.0 to gain the genome variations.Then we adopted sequences of three representative types of major genes from IRGSPv1.0 as queries and BLAST against LG57 assembly to get target sequences. More details about data validation cases from three key genes for LG57 breeding works based on the pedigree genome data especially the assembly data of LG57 and the alignment data of its three parents were listed in Table 9.The maturing time of Geng/japonica is largely affected by Hd1 gene 43 , which commonly harbors highly-diverse variation panels in rice genome 44 .In this region, LG57 and its three early Geng/japonica parents show extremely high consistency.The grain quality of glutinous rice is mainly controlled by Waxy gene 45 .LG57 possess better grain quality than other glutinous early Geng/japonica varieties, such as PN2 and LN2.There are three differences in the Waxy genes found between PN8 and LN2.Although a common variation in the 5 th exon of Waxy was found in PN8, LN2, and their progeny, LG57, there is a unique 23 bp deletion in the 1 st exon that is shared by LG57 and its non-glutinous parent, LG29.Variations in major gene OsNramp5 affects the mineral concentrations in rice 10 .It's notable that LG57 has variations that are different from all three parents, which is supposed to be caused by spontaneous mutations in breeding process 46,47 .Three types of variations in three representative genes validated the genome data and indicated the possible applications with this dataset.In a word, the quality of the pedigree genome data of LG57 was sufficient for public reuse in the future. Fig. 1 Fig.1Outlines of the workflow used to generate and analyze the pedigree genome data for Longgeng 57 (LG57). Table 3 . Comparison of Longgeng 57 dataset with representative assemblies including mega varieties (MV) or standard references (SR). Table 4 . Assembly quality for Longgeng 57 presented by BUSCO. Table 5 . Functional genes predicted in Longgeng 57 comparing with those from databases. Table 6 . Repeats predicted by different methods in Longgeng 57 assembly. Table 8 . Annotation results of coding region in Longgeng 57 assembly in comparing to the commonly used assemblies. Table 9 . Genome variations in three representative types of genes (Hd1 for maturing time, Waxy for amylose content, and OsNramp5 for mineral concentration, where 0 represents the genotype of the reference genome	2024-02-24T06:17:51.487Z	2024-02-22T00:00:00.000	{ "year": 2024, "sha1": "bfd6d25b210c4aeceddff0239a6a81d03176a796", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/s41597-024-03057-x.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "0810c14a888225e8b478d0583cd621af3256ba72", "s2fieldsofstudy": [ "Agricultural and Food Sciences", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
5953862	pes2o/s2orc	v3-fos-license	How Numbers, Nature, and Immune Status of Foxp3+ Regulatory T-Cells Shape the Early Immunological Events in Tumor Development The influence of CD4+CD25+Foxp3+ regulatory T-cells (Tregs) on cancer progression has been demonstrated in a large number of preclinical models and confirmed in several types of malignancies. Neoplastic processes trigger an increase of Treg numbers in draining lymph nodes, spleen, blood, and tumors, leading to the suppression of anti-tumor responses. Treg-depletion before or early in tumor development may lead to complete tumor eradication and extends survival of mice and humans. However this strategy is ineffective in established tumors, highlighting the critical role of the early Treg-tumor encounters. In this review, after discussing old and new concepts of immunological tumor tolerance, we focus on the nature (thymus-derived vs. peripherally derived) and status (naïve or activated/memory) of the regulatory T-cells at tumor emergence. The recent discoveries in this field suggest that the activation status of Tregs and effector T-cells (Teffs) at the first encounter with the tumor are essential to shape the fate and speed of the immune response across a variety of tumor models. The relative timing of activation/recruitment of anti-tumor cells vs. tolerogenic cells at tumor emergence appears to be crucial in the identification of tumor cells as friend or foe, which has broad implications for the design of cancer immunotherapies. A now receding branch of tumor immunology literature favors the view that antigens expressed by many tumors would be ignored by the immune system due to inadequate antigen presentation (1-3). Hewitt, after examining the immunogenicity of many tumor cell lines, concluded that only virus-induced tumors are likely to induce an immune response against them (4). Beyond this founding observation, and the reports that many tumor cells do not express MHC proteins, several groups have found functional alterations of the proteasome (5, 6) and TAP (7) in tumor cells and APCs (8), reducing tumor visibility to the immune system. In contrast, many studies have proven that the immune system is activated in the presence of spontaneous tumors (9-12), revitalizing the concept of immunosurveillance. This concept, first proposed in 1909 (13), was formally defined in 1957 when -based on the findings that the immune system can specifically recognize and reject tumor cells in a chemically induced murine sarcoma model (14) -Burnet proposed that the immune system may prevent tumor development by recognizing antigens absent in normal tissues (15). According to his theory, the immune reactions against tumor antigens expressed by neoplastic cells generally eliminate them at an early stage before any clinical hint of their existence, and frank tumors can grow only after escaping the immune system by diminishing their immunogenicity. The existence of tumor-specific antigens was indeed confirmed in the 1960s by Klein (16). Later, tumors developing in immunodeficient mice were proven to be more immunogenic than tumors developing in immunoproficient mice (17), suggesting that tumors undergo selection by the immune system. However, in other mouse models, immunodeficiency did not promote tumor development (18)(19)(20), and the ability of tumor cells to diminish the expression of their most immunogenic epitopes by adaptation or selection, also called cancer immunoediting, has been questioned (21). The involvement of innate immunity in tumor surveillance was further explored in studies on natural killer cells (NKs), and has produced similar arguments pro (22-24) and con (20,25,26). The conflicting accounts on the role of ongoing anti-tumor surveillance produced in animal models are mirrored by findings in clinical studies that measured the risk of tumor development in patients with immunodeficiencies. Numerous publications have observed a significant increase of cancer occurrence in immunodeficient patients, but at the same time there is no study reporting an "explosion" of cancer cases in these patients. For example, in a study of 2005 on a very large number of immunosuppressed renal transplantation recipients, Hollenbeak et al. observed that of the 89,786 patients who underwent transplantation, 246 patients developed melanoma, with an ageadjusted incidence rate of 55.9 diagnoses per 100,000 individuals. This represented an increase in age-adjusted, standardized risk that www.frontiersin.org was 3.6 times greater than the general population (27). Thus, while such studies support a real role of immunosuppression in promoting cancer susceptibility, the risk of developing a melanoma in the absence of a functional immune system, if increased and non-negligible, is still only 0.056%. One interpretation is that tumor development in the absence of immune system is still a rare event, another interpretation is that the immunodeficient state in patients mostly increases risk of cancers of viral etiology, and that the impact of immunosurveillance on preventing non-viral human cancer may actually be relatively minor (28-31). On the other hand, in already established tumors, presence of tumor-infiltrating lymphocytes is correlated with improved survival (32), and intra-tumoral CD8 T-cells infiltration is associated with delayed recurrence and extended survival in oncologic patients (33). A consensus is that the immune surveillance may guard against cancer under certain conditions, but the precise nature of these conditions is unclear. The first clue that immune tolerance might be a part of the equation came from the works of Nishizuka and Sakakura. While investigating the role of the thymus in tumor immunity in mice susceptible to mouse mammary tumor virus (MMTV)-induced cancer, they observed that neonatal thymectomy at 3 days of age (day 3 nTx) resulted in reduced frequency of breast cancer in tumor-prone (C3H/HeMs × 129/J)F1 females (34), suggesting that cells produced by the thymus after day 3 may protect the tumor. In the following studies, they also looked at tumor development in extra-mammary tissues. There was no increase in the lung and liver tumors after neonatal thymectomy, but the authors reported increased ovarian, lymphoreticular, and pituitary tumor development (35). A notable point of these studies was the discovery that mammary gland development in day 3 nTx female mice was delayed (34) and that mice became infertile secondary to the development of oophoritis (36). At the time, Sakakura and Nishizuka attributed these features to an endocrine role of the thymus, although it is now known to be the manifestation of T-cell-mediated autoimmunity, which paved the road to the discovery of thymic-derived suppressor T-cells, and active tolerance to the tumors. Treg-MEDIATED TUMOR SURVEILLANCE: EXPAND TO REIGN T-cells capable of suppressing the rejection of implanted tumors were first observed in the late 1970s (37-40). These reports remained underappreciated as were most findings pointing to the existence of suppressor cells, caused, in part, by lack of suppressor-specific cellular markers. The doubts have disappeared only in the 1990s, when Sakaguchi, a former student of Nishizuka demonstrated that CD4 + CD25 + T-cells, baptized "regulatory," were responsible for the induction of dominant immune tolerance to tumors. First, the transplantable tumors grew in immunodeficient hosts transferred with whole splenocytes, but were rejected in hosts that have received splenocytes depleted of CD25 + cells (41). Second, the tumors were rejected following preventive treatments with anti-CD25 antibody (42). In both cases, the presence of CD25 + cells inhibited the anti-tumor immune response and their removal led to the complete elimination of the tumor. In a short time, an impressive number of reports confirmed the association between malignant tumors and the regulatory T-cells (Tregs). Clinical studies have shown that CD4 + CD25 + cells are often present within the tumor mass, and have reported a link between a presence of a tumor and an increase in the proportion and/or the number of CD4 + CD25 + Tregs in the blood (43-47). However some results were more heterogeneous depending of the cancer type, and in some studies, no Treg increase was observed (48). Moreover, sometimes the observed proportion of Tregs seems falsely increased by the reduction of the absolute number of CD4 + CD25 − effector cells (Teffs) (49). Regardless of its causes, an important question was whether the observed increase in Tregs is informative for prognosis. Animal models have argued that Tregs have pro-tumorigenic effects (see above), and tumor volume appears directly correlated to the number of Tregs present in the secondary lymphoid organs in several models (50-52). Starting with the report correlating presence of Tregs within the tumor infiltrate and a poor survival prognosis in patients with ovarian cancer (Curiel,200,456), the majority of studies have agreed that an increase in Tregs/Teffs ratio or in an absolute Treg number confers a poor prognosis in cancer patients [see below, and in these recent reviews (48,53,54)]. Yet there are instances in which Treg increase is actually linked to a good prognosis, for example in lymphomas (55, 56) and in colorectal cancer (57-59). The reasons for this discrepancy appear to depend on the special nature of these cancers, in which inflammation may promote tumor growth if not regulated by Tregs, but may also be related to a difference in the origin of cells with Treg characteristics observed in individual malignancies. Concerning the causes of the tumor-induced increase in Tregs, the literature describes several mechanisms: (i) Preferential recruitment of existing thymic-derived Tregs (tTregs), which may be mediated, in part, by chemokines produced by tumors, such as CCL22, that attracts regulatory T-cells, which predominantly express the cognate agonist receptor (60, 61). However, as effector lymphocytes express chemokine receptors as well, chemokine secretion alone cannot explain the preferential recruitment of Tregs to tumor sites (62, 63). The two alternative explanations are (ii) fate conversion -de novo induction of peripheral Treg (pTregs) out of effector T-cells; and (iii) clonal expansion -cytokine and/or antigen-induced proliferation in the periphery of tTregs. Given the vast variety of tumor systems in which all these scenarios have been explored, it is conceivable that the nature of the transforming event, or the tissue of origin of the tumor may determine the specific biological mechanism leading to an increase in Tregs. PERIPHERALLY AND THYMIC-DERIVED Tregs IN CANCER Discovered in the early 2000s in mice (64) and in humans (65), pTregs quickly became the subject of active investigation in tumor immunology, generating evidence both for and against their role in tumor tolerance. Adoptive transfer of CD4 + CD25 − T-cells in mice challenged with either colon cancer or B cell lymphoma resulted in induction of CD25 expression in a significant proportion of donor Teffs, as well as appearance of Foxp3 transcript (66,67). A major line of research pursued a possible instructive role of TGF-ß, a signaling molecule with pleiotropic functions in both immunity and cancer, and in the conversion of CD4 + CD25 − T-cells to pTreg cells (65). TGF-ß acts by binding to the type II TGF-ß receptor (TGF-ßRII), which is constitutively active as a serine/threonine kinase (68,69). A CD4 + cell-restricted blockade of TGF-ß signaling in mice expressing a dominant negative version of the receptor resulted in eradication of TGF-ß expressing lymphoma or metastatic B16F10 melanoma (70) and has established a firm link between TGF-ß and tumor immune tolerance. In part, such a blockade may impair the pro-tumorigenic conversion to pTregs. Indeed, an in vitro study has implied that TGF-ß expressing kidney or prostate tumor cells can stimulate the pro-tumorigenic conversion to pTregs (71). Accordingly, the anti-TGF-ß treatment of mice injected with these tumor cells resulted in fewer tumor nodules; but the in vivo experiments did not exclude a possibility of a direct effect of TGF-ß-blockade on RENCA and TRAMP-C2 cell growth. Moreover, pancreatic tumor-derived TGF-ß was shown to activate Foxp3 expression in tumor cells themselves (72). The functional significance of this upregulation is unclear, as in the tumor cells the Foxp3 transcription factor remains restricted to the cytoplasm, contrary to nuclear localization in Tregs, but it may result in a lower immunogenicity of the tumor, as siRNA-mediated inhibition of Foxp3 expression in tumor cells may shift their cytokine expression pattern toward IL-6 and IL-9 secretion (72). The effect of TGF-ß on the conversion in vivo in tumorbearing mice was addressed more recently using adoptive transfer of CD4 + 25 − Foxp3 − T-cells into Rag −/− mice. In the presence of a TGF-ß-producing pancreatic Pan02 tumor, the transferred T-cells converted into Foxp3 + pTregs, but few FoxP3 + -converted cells were found when mice were transplanted with a TGF-ß-negative esophageal Eso2 tumors (73). As predicted, the induction of cells with Tregs characteristics in Pan02-bearing mice was blocked by systemic injection of an anti-TGF-ß-antibody. This finding mirrors the clinical situation, when increase of Foxp3 + Tregs is observed in patients with a TGF-ß-producing pancreatic adenocarcinoma but not in those with a TGF-ß-negative esophageal tumor (74). Similarly, in non-small cell lung cancer patients, TGFß plasma concentrations directly correlated with the frequency of circulating Tregs (75). As stated above, the spectrum of biological effects of TGF-ß is wide, and is spread beyond the pTreg induction to regulate other Teff responses. For example, anti-TGF-ß treatment significantly and synergistically improved vaccine efficacy as measured by a reduction in growth of the TC1 lung tumor allografts, but anti-TGF-ß alone without vaccine had no impact (76). Moreover, anti-TGF-ß treatment did not affect Treg numbers in lymph nodes and tumors, or their function (76). The resultant synergistic protection induced by anti-TGF-ß plus vaccine combined treatment was likely mediated by CD8 + T-cells since anti-CD8 treatment completely abrogated this effect (76). These results, of course, do not exclude a role for peripherally derived-CD4 + pTregs, but greatly diminish the chances that CD4 + pTregs are the sole culprit behind the TGF-ß effects on tumor tolerance. Overall, the role of TGF-ß in Treg maintenance is mixed, as it inhibits Teffs and Treg cell proliferation, but is important for tTreg and pTreg survival in the periphery (77). In fact, the nature of TGF-ß/Treg interactions may be more complex than a direct conversion scenario would suggest. For example, a mammary tumor cell line, 4T1, can induce recruitment of TGF-ß-producing Gr-1 + CD11b + monocytes (78), and a mouse melanoma and a rat colon tumor were shown to convert dendritic cells (DCs) into the TGF-ß-producing cells, which then led to Treg proliferation (79), possibly through a GILZ-dependent mechanism (80). A similar hierarchy of APC/Treg exchange has been clearly demonstrated in colitis. There, DC-produced TGF-ß was shown to be critical to avoid colitis due to its Treg inducing power (81). In this paper, the Sheppard team showed that DCs lacking the TGF-β-activating integrin αvβ8 failed to induce Tregs in vitro, and that mice with conditional deletion of αvβ8 in DCs presented reduced proportions of Treg cells in colonic tissue. If It should not be excluded that effector cell expansion may contributes to this observed reduction in the fractional number of Treg cells in the colon, these in vitro and in vivo results reinforce observations that DCs are essential in the maintenance of both pTreg and tTreg cells in the periphery (82)(83)(84)(85). A major complication that weakens the accounts of de novo pTreg induction after adoptive transfer of Teffs in tumor-bearing mice is that the CD4 + CD25 − Teffs subset purified in the majority of the conversion experiments of the pioneer articles, contains around 2% of CD4 + CD25 − Foxp3 + T-cells that exhibit suppressive functions (86) and can gain CD25 expression and expand after stimulation (87,88). The experimental approaches based on the sorting of CD25 − T-cells do not provide supportive evidence for a de novo induction of pTregs, and do not exclude a possibility of tumor-driven activation and expansion of CD4 + CD25 − Foxp3 + thymus-derived tTregs. Accordingly, experiments using Teffs transfer from donor mice expressing a Foxp3reporter indicate that generation of peripherally derived FoxP3 + pTregs out of GFP − Teffs within tumors is inefficient and that tumor-infiltrating GFP + FoxP3 + tTregs are highly stable and do not readily convert back to FoxP3 − T-cells contrary to pTregs (89). Some authors suggest that proliferating Helios + Treg cells are a major population in tumors (90), which may be interpreted against pTregs conversion in tumors, Helios being a tTreg marker (91,92). But Helios may be upregulated in peripherally derived pTregs after activation by DCs (93). Another line of evidence questioning the primary role of pTregs in tumor tolerance comes from a recent paper describing the tolerogenic response against the prostate-associated MJ23 selfantigen expressed by prostate tumors induced by an SV40 TAg transgene. On an immunoproficient background, these tumors are infiltrated with MJ23 tumor-specific Tregs, but no MJ23 tumorspecific Tregs were found in tumors that have developed in Aire −/− mice. As Aire is important for tTreg development but dispensable for pTreg induction, these findings indicate that the tumorinfiltrating Treg cells specific for the highly immunogenic MJ23 are principally of the thymic origin (94). Overall, there is little doubt that pTregs may appear from CD25 − subsets, probably from recent tumor emigrant cells (95), in the presence of tumors under certain experimental conditions. Whether this subset plays a substantial role during spontaneous tumor development, is less clear. The difficulty is best illustrated by recalling the original report of Sakaguchi, showing that the immunodeficient mice reconstituted with CD25 depleted splenocytes acquired efficient anti-tumor responses in various cancer models (41). Stated otherwise, any spontaneous pTreg conversion that may occur in this experimental setup does not prevent clearance of the transplantable tumors. ACTIVATED/MEMORY Tregs IN THE EARLY IMMUNE RESPONSE TO CANCER When we were studying the kinetic of early immune responses in various models of cancer by adoptive transfer of CFSE-labeled T-cells, we were struck by the fact that Treg response was not a late event, secondary to the activation of IL-2-releasing anti-TAA effector T-cells (Tumor-associated antigen-specific Teffs), but was actually a very early event, preceding any Teffs activation (52). Such a rapid tumor-specific response of the immune system was counterintuitive in a model of primary tumor exposure, but it bore well with the earlier reports that tumor growth can activate immune cells very quickly. In 1975, Bhatnagar and colleagues have measured ex vivo thymidine incorporation by splenocytes and detected substantial cellular immune responses as early as 1-2 days after i.p. injection of methylcholanthrene-induced fibrosarcoma cells (96). The intensity and rapidity of the cellular response was dependent of the number of cells injected and was always followed by a gradual loss of cellular reactivity against the tumor cells. The progressive loss of immune recognition for tumor cells correlated with progression of tumor growth (96). These observations were confirmed by Berendt and North, who provided evidence supporting the hypothesis that immunity to tumors declines with time as a result of T-cell-mediated immunosuppression (40). More recently, in several injected tumor models (B16 melanoma, 4T1 carcinoma, AB1 mesothelioma, and more) and in an inducible-oncogenedriven breast tumor model, an increase in T-cell division was detected as soon as 2 days after the emergence of the tumor by measure of CFSE dilution as well as BrdU incorporation (52). But this early response was restricted to CD4 + CD25 + Foxp3 + regulatory T-cells, and it appeared to precede the response of conventional T-cells (Figure 1). The responding Treg cells were specific to the antigens, which, although expressed by tumors, were already present in mice before tumor appearance (Figure 1). In other words, the tumor-derived antigens able to stimulate Tregs were self-antigens. Indeed, no Treg expansion was observed against tumors that were not bearing a cognate self-antigen recognized by the transferred tTregs. These observations confirmed previous observations that the self-specific Tregs suppress anti-tumor responses (97, 98), although it did not exclude a possibility that Tregs specific for tumor neoantigens may also participate to the induction of tolerance to the tumor (99, 100). Recently, it was demonstrated that Aire-mediated expression of peripheral tissue antigens drives thymic development of a subset of organ-specific tTregs, which are likely recruited by tumors developing within the associated organ (94). Concerning the APCs that may be responsible for presentation of the tumor self-Ags to Tregs, the good candidates are tissue DCs, which are known to be especially potent in stimulating and maintaining the actively dividing Treg pool (83). Indeed, DCs from tumor-bearing mice were shown to recruit Tregs and to favor their proliferation in the draining lymph nodes (79) (Figure 1). These DCs may present antigens derived from proteins secreted by the live tumor cells, or those derived from tumor cells that die during transformation-induced apoptosis. Of note, microvesicles that are released by tumors and may be captured by DCs for tumor antigen FIGURE 1 \| Early events during cancer emergence lead to immune tolerance against tumor. Activated memory Tregs (AmTregs or amTr, beige lymphocytes) are the first to be stimulated by the presence of the tumor (gray round-shaped cells) via recognition of self-Ag presented by dendritic cells (DCs, star-shaped cells) coming from the tumor site (t1). AmTreg will then proliferate faster than TAA-specific Teffs (Th, gray lymphocytes) that are naïve (or have already been suppressed at the steady state). AmTreg will then inhibit either Teff activation, proliferation, migration, and function either/or DCs presentation and costimulation (t2). presentation (101) appear to have a role in Treg expansion and activation (102) (Figure 1). Moreover, Treg subset expands after adoptive transfer in MHCII +/+ but not in MHCII −/− tumor-bearing mice, which proves that cytokines released in the tumor-bearing mice are not sufficient by themselves to favor Treg recruitment, and that antigen-driven proliferation is mandatory (83). Isolation of Tregs with activated/memory vs. naïve phenotype from tumor-free mice followed by adoptive transfer to tumorbearing mice showed that the initial proliferation of Tregs in tumor-draining lymph nodes was confined to the pool of activated/memory Tregs (amTregs) present in naive mice, (52). These cells were previously characterized as an activated/memory subtype of Tregs, constantly stimulated by self-antigens at the steady state (103). These amTregs are phenotypically and functionally distinct from naïve Tregs (103,104), and are highly potent at suppressing autoimmune responses (105,106). The intensity of the early anti-tumor Treg response is thus explained by their self-specificity and activated/memory status. The early dividing cells described in tumor-bearing mice since 1975 are thus the tolerogenic amTregs cells, a conclusion that is further confirmed by observing tumor rejection following short-term depletion of proliferating immune cells via early administration of anti-mitotic hydroxyurea (HU) or cyclophosphamide (CY) in mice bearing HU/Cy-resistant tumors (50, 52, 107, 108). The early administration of these drugs has a much stronger effect than the late administration, once again suggesting that the immune cells that divide early in the presence of an emerging tumor favor tolerance. Accordingly, a recent analysis of Treg subsets in Her2/Neu-expressing mammary tumor-bearing mice revealed the existence of a Cy-sensitive CD4 + Foxp3 + CD25 + subset with tumor-seeking migratory phenotype, characteristic of amTregs, and capable of high avidity T-cell suppression (109). In addition, the tumor-infiltrating Foxp3 + T-cells express high levels of memory/tumor-associated CCR8 and CXCR4 receptors, and antigen priming is required for the induction of this trafficking receptor phenotype. Thus only antigen-primed, but not antigeninexperienced naive, FoxP3 + T-cells can efficiently migrate into tumors (89). Of course, the effector T-cells also start to proliferate after an adoptive transfer into tumor-bearing mice, but with a primary kinetics that is much slower (9-12 days) than that observed in Treg subset (2-4 days) (83). This delay appears to be sufficient for the establishment of a stable immunosuppressive environment. To test if tolerance to tumors was due to the Treg/Teffs imbalance induced by the delays between their respective activation/expansion, we adoptively transferred high numbers of HA-specific Teffs in mice bearing HA-expressing tumor cells. We observed complete remission in mice adoptively transferred with antigen-experienced HA-specific Teffs (52). Complete regression was also found (i) in secondary-challenged mice cured from first tumor challenge by temporary Treg-depletion (42, 52) and (ii) in tumor-pre-immunized mice (52, [110][111][112]. The activated/memory Teffs, are able to eradicate very efficiently even poorly immunogenic tumors like B16 melanoma (110,111), regardless of the number of Tregs present in the mice (52). Even highly suppressive adoptively transferred tumor-specific Tregs are not able to reverse the anti-tumor memory response (52). The resistance of activated/memory Teffs (amTeffs) to Treg-mediated suppression demonstrated was also observed in other conditions like allograft rejection (113) and autoimmune inflammation (114). Nishikawa and colleagues also observed that CD45RO + but not CD45RA + tumor-specific CD4 T-cells from cancer patients were resistant to Treg suppression (115). This resistance could be due to the fact that activated Tregs can downregulate expression of costimulation molecules by DC (116), but activation/function of amTeffs is much less dependent on costimulation than that of naive T-cells (117). Together, these observations suggest that antitumor amTeffs could be inherently more resistant to Tregs, and explain why detection of amTeffs correlates with good prognosis in cancer patients (118,119). The memory status of Treg and Teffs in early tolerance induction might be important in other settings than just cancer. Several analogies between pregnancy and cancer [reviewed in (120)] point to similarities between the early Treg responses to embryo implantation and tumor emergence. In a just-released study, we observed that early Treg responses to embryo implantation obey to the same rules as those in cancer setting: Tregs expressing markers of the amTreg subset are rapidly recruited to para-aortic conceptusdraining lymph nodes and are activated in the first days after embryo implantation in both syngeneic and allogeneic matings (121). They are also at least in part self-Ag specific, as seen in tumor emergence. Finally, pre-immunization against paternal tissue Ags results in the increase of aborted fetus frequency, and additional Treg-depletion (by anti-CD25) at the time of pre-immunization against paternal tissue Ags, leads to very high frequencies of fetus loss (121). Thus, thymic-derived amTregs appear as a driving force of tolerance to self-ambiguous tissues in the absence of infectious danger signals or pre-immunization. One can then wonder how an immune system that protects deadly tumor cells may survive evolution. We speculate that the AmTreg tolerant response has been actually positively selected to protect allogeneic fetuses against immune rejection. Indeed, Foxp3-expressing Treg-like cells appeared in the first live-bearing animals like Tetraodon (2400 million years) (122) and zebrafish (123), both histotrophic viviparous species. Tregs were thus probably selected in part to protect allogeneic fetuses against immune rejection (121,124), but the pro-tumorigenic activity of Tregs was not counter-selected because cancers mostly develop late in life (125) without affecting reproductive life span. IMPLICATIONS FOR THE DESIGN OF ANTI-CANCER IMMUNOTHERAPIES Activation kinetics and memory status of different T-cell subsets at tumor emergence are pivotal in the outcome of cancer (Figure 2) and explains why preventive immunization is more effective than therapeutic immunization and suggests (i) that preventive vaccination against cancer should be considered seriously and (ii) that therapeutic vaccination could actually worsen host tolerance to tumor antigens (126,127). Development of vaccination strategies must include treatments aimed at Treg-depletion (128)(129)(130) or at inhibition of their function (131)(132)(133), with mandatory validation of the effect of therapeutic vaccination on the level/function of Tregs. Preventive vaccination with tumor-specific antigens presented in a context that would not stimulate amTregs will improve development of efficient amTeffs, which may mount efficient effector responses when a tumor emerges. Noteworthy, although amTeffs are resistant to Tregs, and can cure mice if provided at the time of tumor implantation, the global www.frontiersin.org Together with vaccination and beyond, ablation of Tregs in cancer patients appears to be a promising direction, especially if performed early in the course of the disease (129,135). Nonetheless, we need to remember that the efficiency of anti-tumor responses after Treg ablation is certainly tumor-and genetic backgrounddependent: Treg ablation results in minimal rejection and delayed growth of B16 tumors in B6 mice, 60% rejection of 4T1 tumors in BALB/c mice (83), and close to a 100% rejection of RL 1, MOPC-70A, and Meth A tumors in BALB/c mice (41, 42). These diverse outcomes may depend upon (a) the percentage of Treg cells in a given strain of mice in the steady state, (b) the natural ability of some mouse backgrounds to favor strong Th1 responses, and (c) the tumor-specific expression of immunodominant antigens able to trigger strong anti-tumor effector responses (136). These observations from tumor-bearing mice must be kept in mind while designing new immunotherapies strategies in cancer patients. Altogether, these recent discoveries on the events taking place during the early tumor immune response highlight the importance of the timing and kinetic of Treg and Teff engagement, which depends on their memory status (Figures 1 and 2). In theory, this may disqualify tumor-induced pTregs from playing a substantial role during the early tumor development as they arise preferentially from naïve recent thymic emigrants (95). This does not exclude their eventual involvement in some later events that may sustain the ongoing tolerance. But the fate of the tumor Frontiers in Immunology \| Immunological Tolerance is being decided early, pTregs are unlikely to have much impact in most cancers. Their late arrival in the battle and the absence of memory status puts pTregs at disadvantage during the early tumor development. In tumor immunology and beyond, the timing of engagement dictates the final outcome of an immune response. ACKNOWLEDGMENTS The authors thank Dan Avi Landau for stimulating discussions and critical reading of the manuscript, and David Klatzmann for his concepts discussed in this review, and notably the role of Tregs in the early immune response to cancer and embryo.	2016-06-17T02:44:45.357Z	2013-07-14T00:00:00.000	{ "year": 2013, "sha1": "48d09bb9a298820dbcc98ece29cd53ab22704353", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fimmu.2013.00292/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "48d09bb9a298820dbcc98ece29cd53ab22704353", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
214353656	pes2o/s2orc	v3-fos-license	Smart Origination and Functional Replication: Thermal Reflow and Innovative 3D Structures This contribution summarizes advanced lithographic methods for polymeric 3D topographies based on the modification of the polymer molecular weight and applying thermal polymer reflow. Initial structures realized with grayscale electron beam and multiphoton lithography were reshaped due to thermal annealing close to the glass transition temperature following high-energy radiation. This allowed for new topographical functionalities such as aspheric and ultra-smooth freeform micro and nano-optics. The covered methods have in common that they exploit a specific contrast in molecular weight that enables post-processing and thus a transformation of the initial pattern by polymer reflow into a new surface topography or shape. Also, simulation methods are quickly summarized. Introduction Nanoimprint lithography (NIL) is a highresolution and high-throughput lithography method that allows for low-cost processing compared to other lithography methods [1,2]. In NIL, always the fabrication of the original -the mold -has been the most significant concern, but it is also its biggest asset. Being able to replicate any freeform surface in a vast variety of materials with ultra-high precision is the unique selling point of NIL. Directly fabricating permanent structures into functional materials, e.g., for micro-optics and integrated photonics bears a considerable application potential. This has all been considered before in the search for the killer-application in this area since the start of nanoimprint in the mid-1990s. Today, the ongoing development of new additive fabrication and 3D surface patterning methods as well as highresolution grayscale electron beam lithography (GEBL) approaches offer a new perspective for nanoimprinting. Direct fabrication is, of course, a competitor in this field, but most processes are unfortunately still serial and slow; so replication is required for high-volume fabrication at the end. There are many new challenges for optics such as augmented/mixed reality and early adopters already employ replication techniques there. However, there is more potential to be used in the field of replication of complex surface topographies [3,4]. This article summarizes some of the latest achievements in grayscale lithography and its combination with post-processing ( Fig. 1) for the fabrication of initial patterns and subsequent replication by nanoimprint lithography. The fabrication of the original patterns involves GEBL as well as multi-photon lithography (MPL). The post-processing exploits either inherent differences in material properties following the initial fabrication ( Fig. 1 -(2) lateral) or it uses additional exposure to high-energy photons to create such a spatial difference in material properties ( Fig. 1 -( 3) vertical) in distinction to classical microlens reflow with homogeneous material [5][6][7][8]. In this work, some special focus is put on the material behavior associated with the material modification (namely molecular weight reduction) and its behavior during post-processing (namely thermal polymer reflow). There is a large collection of published work covering different aspects of this already well know process being recently reviewed in a large depth [10,11]. This article does not want to replicate those works but quickly summarizes the most important aspects along with some new material perspectives in this work. Experimental The typical process flow in NIL builds on the following steps and is supplemented by steps that might be unique for the techniques presented here: (a) pattern origination in resist material, (b) pattern modification (if required), (c) replication into final product material. The article covers the first two steps in the following. Pattern origination with GEBL For origination with GEBL, the high-resolution electron beam resist poly (methyl methacrylate) (PMMA) was used. In electron beam lithography, usually a high molecular weight M w of about 950 kg/mol is used to achieve high contrast and high spatial resolution. However, in NIL, usually the lower the M w is, the better is the polymer displacement at finite temperature and thus the imprint quality. For this reason, PMMA with an M w of about 100 kg/mol and a glass transition temperature T g of about 122 °C (micro resist technology GmbH) was selected to allow for a combination of GEBL and NIL in the same resist material [10]. PMMA was spin-coated on the silicon substrates for the required thickness and soft-baked for 2 min at 140 °C at a hotplate. Subsequently, the resist was exposed to high energy electrons (100 kV, EBPG 5000+, Raith GmbH) with a dose mapped from a respective contrast curve (Fig. 2) using a commercial software package for dose assignment and 3-dimensional pattern calculation involving proximity effect correction (BEAMER, GenISys GmbH). Finally, the pattern was developed in 100% methyl isobutyl ketone (MIBK) or mixtures of MIBK and isopropanol (IPA). Usually, a lower contrast is more suitable for GEBL due to higher resilience against dose variations [12]. However, high-resolution patterns [13] demanded minimal dark erosion for the unexposed patterns, especially in this relative low M w resist, and thus, required IPA to be used as co-solvent. Following the development step, a discrete and stepped profile is obtained (Fig. 3). For laterally high-resolution steps, a quasi-continuous pattern might be obtained due to the non-negligible lateral development and thus an effective merging of neighboring steps [13]. Due to the variation of the GEBL dose, the M w, and thus, the T g varies significantly [10]. This allows for a temperature dependent material behavior and hence the name for Pattern origination with MPL Multi-photon lithography is a full freeform 3D fabrication method, also known as high-resolution 3D printing and is based on the non-linear process of multi-photon absorption [14,15]. In contrast to classical laser direct writing, this method allows for the fabrication of intensively cross-linked polymer patterns without significant height restrictions, thus, it can be used for the origination of original patterns for a later mass replication [16], even with the softmold material poly(dimethylsiloxane) (PDMS) in an industrially appealing thermal NIL process [16,17]. Post-processing The post-processing for GEBL was done on a hotplate at different temperatures ranging from 110 °C (slightly below the T g ) to about 130 °C (slightly above the T g ) for different times ranging from a few minutes up to several hours. In general, the lower the temperature, the slower proceeds the pattern deformation and vice versa. Thus, temperature and time allow for effective control of the final shape of the thermal reflow structure. The reflow process itself can be understood as an energy imbalance or surface-tension-driven creep process: under the constant force due to wetting of the polymer to the substrate, the initial pattern deforms towards a minimal energy shape. This shape has usually a constant curvature and is thus a semispherical ("lens-like") pattern. This shape deformation can be stopped at any stage between the initial and the final stage by lowering the sample temperature significantly below T g . The classical reflow [7,8] uses homogenous material, while in this work particular emphasis is put on heterogeneous material [11]. In GEBL, this material contrast is already present after exposure due to the dose induced variation in the T g (cf. Fig. 1). For the MPL origination, PDMS replication into PMMA as typical working material was used [16]. This is required to enable modification of the PMMA T g with vacuum UV (VUV) light of 172 nm wavelength [9]. Due to the exponential absorption of VUV, the T g follows a comparable increase from a rather low temperature toward the bulk temperature [18]. By controlled heating of the surface close to the T g of the bulk PMMA, a contrast in viscosity in the uppermost region compared to the bulk region (cf. Fig. 1) can be achieved. This, as a consequence, results in an ultra-smooth surface that even surpasses in quality on optimally written MPL structure [19] due to the unmatched self-perfection process of thermal reflow. Because surface energies and surface tensions mainly determine the reflow shape, this process can be simulated with so-called soap-film concepts such as implemented in the freeware Surface Evolver (SE) [20]. With this, the process of reflow can be quite well captured in a minimalistic computational effort [21,22]. The significant reduction of the computational burden is obtained by ascribing the volume properties of the reflowing solid volume to the surface elements and by following pre-defined mathematical routines for surface tension flow well describing the reflow process. For the very early onset of reflow and the associated effects, the computationally relative cumbersome finite element methods (FEM) might be more suitable due to the dominance of volume effects [23]. For analysis, manual crystal oriented cleaving of samples to obtain cross-sectional views for scanning electron microscopy (SEM) was used. Figure 3 depicts a collection of initial GEBL patterns and the reflow shapes obtained for different reflow conditions. As it is visible from the pictures of different step height (Fig. 3a vs. Fig. 3b), the exposed step sidewall angle is mainly determined by the step-height and thus by the different exposure doses being required to achieve the respective step height. The closer the respective step dose comes to the dose-to-clear, the lower is the step height, and the larger is the sidewall inclination deviating from 90° vertical sidewalls [4]. As shown, different step configurations are possible. With further reduced step widths, one would obtain quasi-continuous profiles. All steps of maximum height in Fig. 3 are of non-exposed PMMA and hence have a high M w 100 kg/mol [11]. The exposed step has a much lower M w in the range of 5 to 15 kg/mol. This becomes visible in the mechanical behavior during cleaving in a much rougher and less regular cleaving facet for the high M w -regions. Figure 4 summarizes the effect of exposure and reflow for MPL written patterns. Following the direct laser writing of the micro-lens array with MPL ( Figs. 4a and b), the patterns were replicated via PDMS intermediate stamps into PMMA, then exposed to VUV light and finally underwent selective thermal reflow of the surface only (Figs. 4c and d) [9]. Due to the selective surface modification, only the surface became soft enough to allow a surface tension driven material relocation and equilibration of surface undulations. In principle, the thermal reflow acts as filter that removes high-frequency roughness components [24]. Reflow modeling FEM simulation allows for correct prediction of the onset of reflow (Fig. 5) that is in good agreement with other FEM work [23] and analytical concepts [25]. However, the computational effort is relatively high. For a principle parameter search and fast computation, the soap film method based on the Surface Evolver might be more suitable (Fig. 6). The before mentioned difference on cleaving facet roughness was further analyzed, to understand the reflow of non-symmetric M w -and thus T gdistribution. As it becomes clear in Fig. 7, the lower T g region had a lower viscosity -or in other words, higher mobility of the polymer chains. The initial shape was well comparable to Fig. 3b. Thus, in a given time, the higher mobility part was reshaped to a more significant extent and moved further during wetting along the substrate. In consequence, it pulled over the other section of the pattern in this case to the right-hand side. There is still a clear border between the different regions visible in terms of different facet roughness or granularity due to different M w . Thus, the shape in Fig. 7 arises from Fig. 3b). creep and material contrast. In detail, i) fast wetting along the substrate of the more mobile region deforms the right-hand side part more than the lefthand side part and ii) due to the covalent and physical link between both material regions, this deformation translates into an additional deformation of the left hand side that is more pronounced as it would be due to wetting of this part alone. The left-hand side part is not reflowing, and deformation is almost exclusively due to the pull to the right-hand side. Figure 8 shows the SE simulation for different reflow times at 120 °C using two regions of different mobilities. The exposed part of the step-pattern (higher mobility) wets the substrate and moves along the positive x-axis. The lateral movement is due the softened polymer wetting the surface (creep). This wetting creates a force on the non-exposed part and leads to the before mentioned deformation. The comparison with the experimental results show a good agreement considering the minimal computing effort required with the SE method. Further differences arise from the fact that the SE does not take bulk effects into account. Conclusion The work has shown that thermoplastic reflow combined with a material contrast is a versatile tool for advanced topography fabrication as well as for ultra-smooth surface polishing without affecting the bulk geometry. Essential for this method is the formation of the material contrast with high-energy radiation influencing thermomechanical properties. The reflow close to the glass transition of the thermoplastic polymer involves visco-elastic behavior that can involve some residual elastic contribution, which defines the final reflow shape.	2020-01-23T09:10:47.865Z	2020-01-31T00:00:00.000	{ "year": 2020, "sha1": "8ad9a66b52e79f2ab2a4df31fa9fad821b93eabb", "oa_license": null, "oa_url": "https://www.jstage.jst.go.jp/article/photopolymer/32/6/32_799/_pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "d59706e7cd714b4817a1b90be3f514607ea9d51b", "s2fieldsofstudy": [ "Materials Science" ], "extfieldsofstudy": [ "Materials Science" ] }
246759271	pes2o/s2orc	v3-fos-license	Integration of Augmented Reality into MOOC’s in Vocational Education to Support Education 3.0 This research discoverable (1) the development of a MOOC (Massive Open Course Online) learning innovation that incorporates Augmented Reality; (2). The use of the concept of AI Injected e-Learning (The Future of Online Education) in Vocational High Schools, which supports the idea of Education 3.0. This study employs a quantitative method to educational research and development, with research sample vocational students. The findings of this research lead to the following conclusions: (1) The efficacy and efficiency of MOOC's learning innovation integrated Augmented Reality was evaluated to be 87.3 percent in a usability test; (2). The application of the concept of Education 3.0 proved to have a significant impact on learning comfort, as evidenced by a questionnaire instrument that revealed that around 85.4% of users believed that online and face-to-face learning provided flexibility in learning, 12.2% felt that learning was normal, and 10% felt that learning was not required; (3). The results of the correct hypothesis are acquired from the quantitative data with a significance value of greater than 0.5, indicating that MOCC's integrated learning utilizing Augmented Reality has resulted in a significant increase in the learning outcomes of vocational high school students by 38 percent Background Because there are always new things to learn, the creation of learning innovations is currently considered to have a broad scope. In these circumstances, interactive learning can help build a good, friendly, and comprehensive learning environment [1], [2]. Learning innovation is a benchmark in the Development of comprehensive learning that can increase learning motivation through the development of an online learning system that is integrated with MOOC. It is not new that we know MOOC as an online learning system that is integrated with the Augmented Reality system [3], [4], but what needs to be explored further is the 3.0 education concept embedded in it providing a blend of learning innovations that can accumulate in terms of student-based learning. and monitored directly by the teacher. MOOC is a concept that is predicted to be a good concept in learning, this is because MOOC can be accessed from all over the world and can be directly (free) used by students around the world so that access to information and learning can be spread easily [5]. Education is the Development of a human being to be able to actualize and increase his role in life in social life. It is this role that inspires improvements, strategies or learning approaches. Education also has several improvements to be able to bridge the gap with the Development of technology that is increasingly rapidly nowadays. Students have started to leave conventional media into interactive multimedia in various forms (mobile, augmented reality to artificial intelligence) [6], [7]. As a result, education must adapt as well. One of the modifications presented is the existence of a version for education, with the expectation that each new version would include new features that will make learning easier for students [8]. The update or upgrade in question is the installation of Education 1.0, 2.0, and 3.0, which was inspired by the growth in static web versions 1.0, 2.0, and 3.0, each of which has major improvements in terms of website facilities and usability. The primary principle of Education 3.0 is that education can be altered, customized, and flexibly conditioned to meet the needs of pupils. Educators facilitate basic learning concepts, and all learning activities are centered on educators, in contrast to Education 1.0 and Education 2.0. The essential principle of open learning is carried over into Education 3.0, which is organized in a pattern (Connectors, Creators, Constructivists). These three patterns have a long-term association with learning, particularly in terms of boosting motivation and interest. This concept is the forerunner of National Development in the field of learning innovation throughout the country, by emphasizing the actual growth process in the quality of education and following the trend of technological developments, of course providing accelerated distribution of learning that comes from reliable learning sources. The concept of equal distribution of learning needs to be balanced with the existence of learning content that can be accessed widely and easily, one of which is the distribution of learning through information technology in current learning [9]. MOOC's is a learning resource that is currently needed in online and offline learning, the idea of combining the two (MOOC's and Augmented Reality) comes from the results of research which states that in 2020 there were almost 85.32% of active smartphone users who can access the internet through data. private or free wifi hotspot. This of course proves that digital era learning, in this case online learning, has no problems in its application [10], [5]. In 2020 the Indonesian government also launched a Free Internet Quota for students and teachers, of course this is a breath of fresh air and assistance that can distribute learning evenly through the MOOC's integrated Augmented Reality scheme, besides that youth and accessibility of learning materials are embedded with object 3 Dimensions makes learning easier and interactive for the learning process ( Figure 1). MOOC's Development, which is given Augmented Reality accessibility, not only carries the concept of flexibility in its application [11], this Development is also based on learning innovations that prioritize teacher creativity to provide complete learning content, through 3-dimensional objects that can be directly accessed by students and of course standardized material through material expert validation and media expert validation. Preliminary research also concluded that MOOC's users who were sampled through programming courses (Preliminary Test and Final Test) provided a statement through a questionnaire that the role of MOOC's integrated with Augmented Reality gave a percentage of about 84% as the main supplement of learning, by using this combination students' understanding increased. inline with what is captured by students through the learning process [10], [12], with a mechanism like this students will be student centered and creative [13], [14]. According to [15], [16], also mentions that the use of MOOC's which is integrated with Augmented Reality in the learning process must have 3 criteria so that it can be used in learning, namely user friendly, user usability and effectiveness of use. User Friendly refers to the use of MOOC's integrated Augmented Reality that is easy to use through applications that do not make it difficult for users, so that users will get convenience and do not feel difficult, this usability is emphasized on the level of system quality that is easy to learn and remember and encourages users to use the application as learning. Usability of use is related to the concept of the extent to which the application can be used as a relevant learning resource and can be used as a reference in learning, through a series of tests to see the reliability of MOOC's integrated with Augmented Reality. While the effectiveness of use is testing to test MOOC's later when used can be applied by hardware with low or hard specifications, this is important to do so that the effectiveness of MOOC's can be used by smartphones and devices that may be low end by students and can be generalized in online learning that is integrated. The goal is that MOOC's AR integrated can be accessed by anyone in any part of the world and with any device as long as it is connected to the internet network. Methods The development method in developing the MOOC's integrated Augmented Reality product uses Research and Development or commonly referred to as R&D where this method is carried out by referring to conclusions and procedural steps that are integrated with 10 concrete steps for efficient, flexible and effective product development, as follows: these steps are presented in (Figure 2). Fig. 2. Design stages of R & D borg and gall The 10 procedural steps in R&D are a mechanism for obtaining highly efficient products, through the concept developed borg & gall concrete steps were obtained, including: (1) Information gathering stage for research through preliminary studies and field observations to knowing the extent to which the product can solve the problem; (2) The product planning stage which refers to problem-based design, the problem to be solved is then made a rough design to be continued in the next stage; (3) The pre-product planning stage begins with prototyping the results of product planning; (4) The preliminary test phase in the field is carried out with small and large group trials through prototyping that has been developed; (5) Revision stage of the developed product, in this case MOOC's integrated AR after being tested in small and large groups to get revision results that need to be improved for product perfection; (6) The field testing phase is based on the results of the Development to see the toughness of the product; (7) Operational product revisions that may be obtained in field tests; (9) The final revision stage which is final testing and has been refined in several previous stages and the product is ready to be distributed and used in general; (10). The full socialization and implementation stages in the learning process, in this case MOOC's integrated Augmented Reality can already be used in the learning process and is ready to be used as a supplement that goes through various stages in its development. Results and discussion The results and discussion of the Development of MOOC's integrated Augmented Reality will discuss the Development of learning content in MOOC's, the learning content consists of animation and video which is integrated using 3 Dimensional objects that can be displayed by Augmented Reality. This MOOC's product represents the Development of learning innovations with the concept of flexibility and accessibility of application usage. The results of the Development of MOOC's products include learning content as the main source of students in the teaching and learning process, then learning content which contains learning modules (E-Modules), videos and job sheets which are used in the Computer Graphics course with Basic Competencies of Introduction to 3 Dimensional Objects. Although it was developed in the Graphics course, the generalization of this product can also be used in other courses. MOOC's which were developed are likened to a supermarket that provides all kinds of kitchen needs that students can choose from, students can choose according to their learning needs with a meaning such as supermarkets, students can choose their own learning needs so that it can greatly increase students' learning motivation. MOOC's is also combined with education 3.0 where this concept was developed based on the selection of materials that are in sync with the needs of the industry so that later students are able to compete in the industrial world. This mechanism is indeed embedded in MOOC's, especially in the selection of materials that can be selected according to students' abilities, both with asynchronous and synchronous learning methods. The following is a product development process. Sub instrument development In the first test, it resulted in the Development of MOOC's integrated Augmented Reality by developing an E-Module that will be used in MOOC's learner content, [17] this module was developed based on the needs of the Computer Graphics course, consisting of 9 Main Modules whose Development is based on basic competencies and already validated by a material expert who has been teaching the material for more than 10 years, the selection of the sub-instrument Development is related to the basic competencies that must be achieved by students to be able to pass graduation in the course. The following is a description of the module that was developed into a sub-instrument development which can be seen in Table 1. The Augmented Reality development embedded in MOOC's also uses Four D Development Research whose development mechanism is actually very similar to R&D but is more concise than and faster, the reason for this Four D is so that Augmented Reality developed can be demonstrated and aims to produce new products. through the Development process [18], [19]. Four D uses 4 simple special stages, namely (1) Define the product, which in this case is augmented reality that is ready to be used through a series of tests, then (2) Discover which contains the design of the project being developed, including the framework and wireframes of AR programming; (3) Develop is the process of developing AR products based on the layout of the project framework; and finally the final stage (4). It is a demonstration and dissemination through the Google Playstore and Appstore to see how users respond to the learning concepts that have been developed [12], [20] The 4 steps of Four D in detail can be seen in Figure 3. Testing of augmented reality markers AR marker testing was conducted to find out how strong the readability level of the marker was towards the reader, this was important so that the concept of flexibility and ease factors was found in the application, following Figure 4 is MOOC's integrated Augmented Reality marker. The process carried out is to make a sampling of the markers that have been made, with sampling carried out thoroughly, it is expected to get a marker that has a level of strong readability, a strong level of accuracy and a high level of information delivery. The use of these markers is based on experimental distance, from the lowest to the highest level, here are Figure 5 Levels of readability Marker in MOOC's research integrated Augmented Reality. Table 2 after the readability level test is done, distance test is done, this is done to anticipate the user using more than the specified distance, which is 50 cm, after the marker distance test is done, it turns out the results exceed the standard, namely the marker is able to read distance of 80 cm [21], this proves that the readability test is valid so that it can help the level of distance in the marker distance test [22]. Testing the effect on Education 3.0 in vocational high schools Testing the effect on Education 3.0 applied to Secondary School students in vocational education resulted in 87.3 levels of effectiveness, efficiency and usability tests carried out on 120 respondents or 3 classes with random sampling, along with the description in Table 3. From the results of the test variables, it can be seen in Table 3 that the significance is at 0.678, this is greater than the significance of 0.5 with STD Deviation 2.186 and df 68, this shows very good results, so it can be concluded that MOOC's can increase student motivation in learning, through Testing the effectiveness, efficiency and usability test also got good results, the range value on all aspects is 87.3% which is in the high category with 17 variables studied, indicating that this Development is feasible and ready to be used as a whole. In-depth exposure to these 17 variables has an impact when applied to MOOC's integrated Augmented Reality; this proves that the concepts obtained in preliminary research from observations, field studies, and literacy studies are appropriate and can be scientifically justified. The results of this test are also included in the main conclusions in this study. In the Variable learning comfort also obtained results that fall into the good category with 85.4% results which indicate that the learning comfort obtained can be used as a benchmark in product application. The following in Table 4 is shown in full. Conclusion This integrated Augmented Reality MOOC researcher reveals several issues, including: (1). Development of MOOC's (Massive Open Course Online) learning innovation integrated Augmented Reality that is intended for vocational high school students; (2). The application of the concept of AI Injected e-Learning (The Future of Online Education) which supports the concept of Education 3.0 in Vocational High Schools with a variety of sub-variables that indicate an increase in student learning outcomes; and (3). The results of MOOC's Development are integrated with Augmented Reality that can improve learning styles and increase learning outcomes of vocational high school students. This study uses a model of Development of educational research and Development with a quantitative approach, research sampling was carried out in vocational high schools in Malang and Batu City. The results of this study take the following conclusions: (1) The development of MOOC's (Massive Open Course Online) learning innovation integrated Augmented Reality has a value of effectiveness and efficiency in usability test estimated to be 87.3% which indicates a high level of user comfort in the user friendly category and efficient use of the application; (2). The application of the concept of Education 3.0 proved to have a significant impact on learning comfort, through a questionnaire instrument that proved that around 85.4% of the users considered that online and face-to-face learning provided flexibility in learning, 12.2% of respondents felt normal with learning, while 10% feel not needed; (3). From the quantitative data the results of the correct hypothesis are obtained with a significance value of more than 0.5, thus proving that there is a significant increase in the learning outcomes of vocational high school students by 38% after MOCC's integrated learning using Augmented Reality. Acknowledgment This development research is supported by the Universitas Negeri Malang through licensing of data collection and full support for the use of the laboratory of the Electrical Engineering Department, the Engineering Faculty, Universitas Negeri Malang. Thank you very much to those who have assisted in this development research, especially to vocational students at the Universitas Negeri Malang. Authors Ahmad Mursyidun Nidhom is a member of the International for the engineers and the computer scientist (IAENG), 174192, Hung To Road, Hongkong. He is active in conducting research on foreign collaborations and managing international journals. He also has a network of research collaborations in Southeast Asia. (E-mail: nidhom.ft@ um.ac.id). Andika Bagus Nur Rahma Putra is a productive young lecturer at the Malang State University, Indonesia. He is active in writing scientific papers, researching, and developing learning technology. He also owns more than 100 educational copyrights and often conduct international collaborative research. (E-mail: andika.bagus.ft@um.ac.id). Azhar Ahmad Smaragdina is a lecturer who is involved in international cooperation; he is also one of the innovators in the field of Education, more than 10 copyrights resulting from international research collaborations (E-mail: azhar.ft@um.ac.id). Gres Dyah Kusuma Ningrum is a lecturer who is involved in the Development of national educator certification; She is a lecturer who has service collaborations throughout the nation and is a lecturer in developing learning innovations (E-mail: gres.dyah. ft@um.ac.id).	2022-02-12T16:06:13.481Z	2022-02-10T00:00:00.000	{ "year": 2022, "sha1": "58c819dc100da153870d65d28639f9ece76f14fd", "oa_license": "CCBY", "oa_url": "https://online-journals.org/index.php/i-jim/article/download/28961/10709", "oa_status": "GOLD", "pdf_src": "Adhoc", "pdf_hash": "324225e5f18a866b05f841a768487423a10c58f3", "s2fieldsofstudy": [ "Education" ], "extfieldsofstudy": [ "Computer Science" ] }
5102221	pes2o/s2orc	v3-fos-license	On the number of Enriques quotients of a K3 surface In this paper we discuss the number of Enriques quotients of a fixed K3 surface. We prove the finiteness and unboundedness of the number. We also show an example of Kummer surface of product type where we can successfully classify all the Enriques quotients. Introduction A K3 surface X is a compact complex surface with K X ∼ 0 and H 1 (X, O X ) = 0. An Enriques surface is a compact complex surface with H 1 (Y, O Y ) = H 2 (Y, O Y ) = 0 and 2K Y ∼ 0. The universal covering of an Enriques surface is a K3 surface. Conversely every quotient of a K3 surface by a free involution is an Enriques surface. Here a free involution is an automorphism of order 2 without any fixed points. The moduli space of Enriques surfaces is constructed using the periods of their covering K3 surfaces. Precisely speaking, an Enriques surface determines a latticepolarized K3 surface and vice versa, so that the moduli space of Enriques surfaces can be described by the moduli space of lattice-polarized K3 surfaces. We note that even if we do not fix any polarization on Enriques surfaces, their covering K3 surfaces automatically have a lattice-polarization. Then, what happens if we drop the latticepolarization of the covering K3 surface? We will call two Enriques quotients of a K3 surface distinct if they are not isomorphic to each other as varieties. In his paper [3], Kondo discovered a K3 surface with two distinct Enriques quotients. He computed the automorphism groups of the two quotients. Since then, as far as the author knows, no other examples have been found. In this paper we investigate this phenomenon. We show that K3 surfaces with more than one distinct Enriques quotients have 9-dimensional components (neither irreducible nor closed) in the period domain. Moreover we compute the exact number of distinct Enriques quotients at a very general point of each component (Proposition 3.6). This generalizes Kondo's example in an arithmetic way and results in the following unboundedness theorem. Theorem 0.1. For any nonnegative integer l, there exists a K3 surface X with exactly 2 l+10 distinct Enriques quotients. In particular, there does not exist a universal bound for the number of distinct Enriques quotients of a K3 surface. We also generalize Kondo's example in a geometric way. Its construction is due to Mukai [4]. We introduce his construction and show that a generic Kummer surface X of product type (see Section 4) has exactly 15 distinct Enriques quotients, which can be constructed from classical Lieberman's involution and Kondo-Mukai's involution. Theorem 0.2. X has exactly 15 distinct Enriques quotients which are naturally in one-to-one correspondence with nonzero elements of the discriminant group of NS(X). From the theoretical point of view, we first show the following finiteness theorem on the automorphism group of a K3 surface X using a theorem of Borel. Theorem 0.3. In Aut(X) there are only finitely many conjugacy classes of finite subgroups. This theorem concerns us because it bounds the number of distinct Enriques quotients for any K3 surfaces. Corollary 0.4. Every K3 surface X has only finitely many distinct Enriques quotients. The usage of the theorem of Borel is suggested by an anonimous referee. We remark that Corollary 0.4 follows also from our counting method described in Section 2. There, to count the exact number of distinct Enriques quotients, we consider more directly the embeddings of the Enriques lattice U(2) ⊕ E 8 (2) into Néron-Severi lattices. Notations and Convention. Our main tool is the theory of lattices and their discriminant forms. Here we collect some basic definitions about them. See [7] for the detailed exposition. A lattice L is a free Z-module of finite rank equipped with a Z-valued symmetric bilinear form. L is said to be even if for all l ∈ L, l 2 ∈ 2Z. In this paper we treat only even lattices, so that we sometimes omit mentioning the evenness. For a lattice L, there is a natural homomorphism c : L → L * = Hom(L, Z) defined by l → (l, ·). L is said to be nondegenerate if c is injective, and unimodular if c is bijective. For m ∈ Q, L(m) denotes the same underlying group equipped with the form multiplied by m, assuming that it is Z-valued. U, E 8 and D 4 denote the lattices given by the matrix ( 0 1 1 0 ), the Dynkin diagrams of type E 8 and D 4 respectively. We understand the latter two to be negative definite. A finite quadratic form is a triple (A, q, b) where A is a finite abelian group, q is a map A → Q/2Z and b is a bilinear map A × A → Q/Z which is symmetric and satisfies In the following we abbreviate b(x, y) (resp. q(x)) to xy (resp. x 2 ) and sometimes (A, q, b) to (A, q). We call x 2 the norm of x. As in the lattice case, we have a natural homomorphism c : A → A * = Hom(A, Q/Z) defined by using b. (A, q) is said to be nondegenerate if c is bijective. For an even nondegenerate lattice L, we can canonically associate a finite quadratic form (A L , q L ), called the discriminant quadratic form of L, by putting A L := L * /L and q L is the one naturally induced from the linear extension of the form on L to L * ⊂ L ⊗ Q. The discriminant group of U(2) (resp. D 4 ) is denoted by u(2) (resp. v(2)). For a lattice L, O(L) (resp. O(q L )) denotes the integral orthogonal group of L (resp. of (A L , q L )). We note that there is a natural homomorphism σ L : O(L) → O(q L ). L R (resp. L C ) is the scalar extension of L to R (resp. C). The author is grateful to Professor Shigeru Mukai for many helpful discussions and suggestions. He indicated the example in Section 4. He is also grateful to Professors Shigefumi Mori and Noboru Nakayama for many valuable comments throughout the seminars. Finiteness of conjugacy classes of finite subgroups First we collect some basic definitions about K3 surfaces. Let X be a K3 surface. It is known that all K3 surfaces are diffeomorphic. A K3 lattice is a lattice isomorphic to is the Néron-Severi lattice of X. T X = (NS(X)) ⊥ is the transcendental lattice of X. We recall the structure of the integral automorphism group O(NS) of NS(X). 2. The ample cone A X is the subcone of C X generated as a semigroup by ample divisors multiplied by positive real numbers. 1. The Weyl group W X of X is the subgroup of O(NS) generated by automorphisms of the form s l : x → x + (xl)l for all elements l ∈ NS(X) with l 2 = −2. Further we use the abbreviations The following relation between these subgroups are important. Since the generator s l ∈ W X acts trivially on the hyperplane H l orthogonal to l which intersects with the positive cone, W X preserves the positive cone. If x ∈ NS * , then xl ∈ Z and s l (x) = x modulo Zl. This proves (1). We have the semidirect product decomposition as in (2) because W X and O ↑ (NS) are discrete subgroups of the isometry group of the Lobacevskiȋ space modeled in C X and W X is a reflection group with ample cone as its fundamental domain. See [12]. The proof of (3) is the same. q.e.d. We proceed to the proof of Theorem 0.3. For brevity, we say that a group G has property (F P ) if G has only finitely many conjugacy classes of finite subgroups. For example, let G be an algebraic group defined over Q. Then G Z has property (F P ) by [11,Theorem 4.3] which we call the theorem of Borel. (1) Let α : G → K be a homomorphism of groups. If im α has property (FP) and ker α is finite, then G has property (FP). (2) Let G = W ⋊ K be a semidirect decomposition of a group G. If two finite subgroups F 1 , F 2 ⊂ K are G-conjugate, then they are also K-conjugate. In particular if G has property (FP), then so does K. (3) Let H ⊂ G be a subgroup of finite index. If G has property (FP), then so does H. P roof. (1) Let P 1 , · · · , P n be the complete representatives of conjugacy classes of finite subgroups of im α. Then any conjugacy class of finite subgroups of G has a representative included in at least one of α −1 (P j ), (j = 1, · · · , n). (2) Assume F 2 = gF 1 g −1 . g can be written as Thus F 1 and F 2 are conjugate by k ∈ K. (3) Again let P 1 , · · · , P n be the complete representatives of conjugacy classes of finite subgroups of G. We put G/H = {a 1 H, · · · , a r H}. Then the conjugacy classes of finite subgroups of H are represented by {a −1 i P j a i \|i = 1, · · · , r and j = 1, · · · , n}. q.e.d. Now we show Theorem 0.3. In our words, Theorem 1.5. Let X be a K3 surface. Then Aut(X) has property (F P ). P roof. First we assume that X is projective. Consider the representation r : Aut(X) → O(NS). Since every automorphism in ker r fixes an ample divisor, ker r is finite. Thus it is enough to show that im r has the property (F P ) by Lemma 1.4(1). By the theorem of Borel above, O(NS) has property (F P ). Then, by Lemma 1.4 and Proposition 1.3, the property (F P ) goes down to O ↑ (NS) and O + (NS). Now by the global Torelli theorem [10], im r contains O + 0 (NS), since ϕ ∈ O + 0 (NS) preserves the ample cone A X and can be extended to an isometry of H 2 (X, Z) which acts trivially on T X . Thus we obtain and these inclusions are of finite index. We see that im r has the property (F P ) by Lemma 1.4(3). Next we assume that X is not projective. Nikulin [8] shows that any automorphism of X of finite order acts on T X trivially. Therefore it is enough to show that G = ker(Aut(X) → O(T X )) has property (F P ). We consider the representation r : If alg. dim(X) = 1, then NS(X) has one-dimensional kernel Ze and Q := NS(X)/Ze is negative definite. Every element of G fixes e since e 2 = 0 and exactly one of e and −e is represented by an effective cycle. Thus r induces s : G → O(Q). Let g ∈ G be an element of finite order. Since the fixed part H 2 (X, Z) g is nondegenerate by the lemma below, it follows that if s(g) = id Q then g = id X . On the other hand, O(Q) is a finite group. Thus Aut(X) has only finitely many elements of finite order and Aut(X) has property (F P ). q.e.d. Lemma 1.6. Let L be a nondegenerate lattice and let g be an isometry of L of finite order n. Let M = L g = {x ∈ L\|gx = x} be the fixed lattice. Then M is nondegenerate. Since L is nondegenerate, there exists y ∈ L with xy = 0. Put z = y + g(y) + · · · + g n−1 (y). Obviously z ∈ M and we have As a corollary, Corollary 0.4 follows. In fact, if two free involutions i 1 and i 2 are conjugate by an automorphism g, then g induces an isomorphism between X/i 1 and X/i 2 . Number of distinct Enriques quotients The isomorphism classes of Enriques quotients are exactly the conjugacy classes of free involutions by the next proposition. Proposition 2.1. Let X be a K3 surface and let i 1 and i 2 be free involutions on X. Then, X/i 1 and X/i 2 are isomorphic if and only if there exists an automorphism g of X such that gi 1 g −1 = i 2 . P roof. The "if" part is trivial; see the sentence after Lemma 1.6. Conversely, let h be an isomorphism from where π denotes the covering map. It is clear that g is the desired automorphism of X. q.e.d. We put Recall that U ⊕ E 8 is the Enriques lattice (modulo torsion) and U(2) ⊕ E 8 (2) is the pullback in the covering K3 lattice. For each M ∈ M, we define an isometry Proposition 2.2. On a K3 surface X, there is a one-to-one correspondence between free involutions on X and primitive sublattices M of NS(X) which satisfy (A) and (B) above and the following (C) M contains an ample divisor. In other words, i M defined above is a free involution if and only if M contains an ample divisor. Also any free involution can be written in the form i M . P roof. We associate a free involution with its invariant sublattice in H 2 (X, Z). The statement follows from [6, Corollary 2.5], [9, Theorem 4.2.2, p1426] and the strong Torelli theorem for K3 surfaces [10]. In [9], the assumption is slightly different from ours, but the same proof goes. q.e.d. To count the number of distinct Enriques quotients, we consider the natural action Corresponding lattice automorphisms satisfy i ϕ(M ) = ϕi M ϕ −1 . In the following, Aut(T X , ω X ) is the subgroup of O(T X ) consisting of the integral orthogonal transformations whose scalar extention to C preserves the period Cω X ⊂ T X ⊗ C. Theorem 2.3. Let M 1 , · · · , M k ∈ M be a (finite) complete set of representatives for the action of O(NS) on M. For each j = 1, · · · , k, let be the stabilizer subgroup of M j and σ(K (j) ) its canonical image in O(q N S ). We put (1) The number of distinct Enriques quotients of X does not exceed B 0 . P roof. First we remark that by Proposition 1.15.1 in [7], the set of representatives is always finite. In view of Proposition 2.1, we can count the number of distinct Enriques quotients separately for each orbit O(NS) · M j . Hence, for simplicity, we fix an orbit and omit the index j so that we use the symbols M := M j , O := O(NS) · M and K = K (j) . Step 1. O contains an element which corresponds to a free involution. P roof. The following is a standard argument used in [10]. Our proof is taken from [6]. By the condition (A) in Proposition 2.2, M ∩ C X = ∅. Consider in C X countably many hyperplanes H d = {x ∈ NS R \|xd = 0}, where d runs over (−2) vectors in NS. The union ∪H d is a locally finite closed subset in C X and does not contain M by the condition (B). The complement C X − ∪H d is a collection of (at most) countably many connected open sets, namely chambers, which corresponds to the elements of W X in one-to-one way. The ample cone A X equals one of the chambers. Thus if we choose Thus we can assume that i M is already a free involution of X. Next we set Step 2. Step 3. Let ψ 1 , ψ 2 ∈ O + (NS). If σ(ψ j ) have the same class in O(q N S )/σ(K), then i ψ 1 (M ) and i ψ 2 (M ) are conjugate in Aut(X). P roof. By the assumption σ(ψ −1 1 ψ 2 ) ∈ σ(K), so there exists ϕ ∈ K such that σ(ϕ) = σ(ψ −1 1 ψ 2 ). It follows that σ(ψ 1 ϕψ −1 2 ) = id, so that ψ 1 ϕψ −1 2 \| N S together with id T X gives an automorphism a of X, by the Torelli theorem. It follows that By now, we have proved that O contains at most #O(q N S )/σ(K) distinct Enriques quotients. Assertion (1) follows. Lastly we mention a useful theorem of Nikulin in [7] which saves us from checking one of the conditions in Theorem 2.3 (2). Theorem 2.4 (Nikulin). Let T be an even indefinite nondegenerate lattice satisfying the following two conditions: (1) rank(T ) ≥ l(A Tp ) + 2 for all prime numbers p except for 2. (2) if rank(T ) = l(A T 2 ), then q T 2 contains a component u (2) or v (2). Then the genus of T contains only one class, and the homomorphism O(T ) → O(q T ) is surjective. Here A Tp denotes the p-component of the finite abelian group A T and l denotes the number of minimal generators. Enriques quotients of Ksurfaces in the Heegner divisors In this section we prove Theorem 0.1. We deal with certain divisors of the period domain D of U(2) ⊕ E 8 (2)-polarized marked K3 surfaces. Fix the unique primitive embedding of U(2) ⊕ E 8 (2) in the K3 lattice Λ. Then D is by definition Here P(V ) means the associated projective space of a complex vector space V , which consists of all lines through the origin. It follows from the surjectivity of the period map that every point of D corresponds to a K3 surface X with a marking H 2 (X, Z) ∼ = Λ. Let S ⊂ Λ be a primitive sublattice of rank 11 containing the lattice U(2) ⊕ E 8 (2) fixed above. Then the subset is called the Heegner divisor of type S in D. Let X be a marked K3 surface whose period is in D(S). Since NS(X) is written as Λ ∩ ω ⊥ X , NS(X) contains the primitive sublattice S. P roof. This is a well-known fact. For the latter, the same proof as in [1, Lemma 2.9] works. q.e.d. We consider the case when where −2N is the rank 1 lattice whose generator g has g 2 = −2N. It is easy to see that the K3 lattice Λ contains S as a primitive sublattice. We fix it once and for all. Let the integer N be N = 4p 1 · · · p l , where p 1 , · · · , p l are distinct odd prime numbers. In the next we compute the order of O(q). Then we can use the structure theorem of the unit group (Z/4NZ) × . Similarly the left-hand-side of the latter is Using the commutative diagram where α, β are both 2:1 maps, we can count the number of elements which have order 2 in the bottom row but do not in the top row. q.e.d. P roof. Such element x generates a subgroup x isomorphic to c(−2N). Since it is nondegenerate, x is a direct summand in q. This implies the transitivity. If we put the number of elements in u(2) ⊕5 with norm 0 to be A and norm 1 2 10 − A, we can compute the length of the orbit as 2 l+1 · A + 2 l+1 · (2 10 − A) = 2 11+l . q.e.d. Under these computations we can prove Theorem 3.4. For any nonnegative integer l, there exists a K3 surface X with exactly 2 l+10 distinct Enriques quotients. We see that any primitive embedding as above is a direct summand. This clearly implies the uniqueness. Obviously NS(X) has a primitive sublattice M isomorphic to U(2) ⊕ E 8 (2) and M ⊥ = −2N . Let K be the stabilizer group of M and σ(K) its canonical image in q.e.d. In fact we can classify all the possible Néron-Severi lattices of a K3 surface with Picard number 11 having an Enriques quotient. In each case, we can compute the number of Enriques quotients as follows by an explicit calculation. Details are omitted. The result is as follows. Proposition 3.5. Let X be a K3 surface with Picard number 11 having an Enriques quotient. Then the Néron-Severi lattice of X is one of the followings. If we put 2N = 2 e p e 1 1 · · · p e l l in type I, or 4M = 2 e p e 1 1 · · · p e l l in type II, the bound B 0 in Theorem 2.3 is as follows. in Type I and e = 1 (2 5 + 1) · 2 l+4 in Type I and e = 2 2 l+10 in Type I and e ≥ 3 1 in Type II and e = 2, l = 0 2 l−1 in Type II and e = 2, l > 0 2 2l+5 in Type II and e ≥ 3 The lattice S we used fits in the third case. Enriques quotients of generic Kummer surfaces of product type Kondo found the first example of a K3 surface which has two distinct Enriques quotients in [3, Remark 3.5.3], where he computed the automorphism groups of the two quotients. Recently Mukai generalized Kondo's example which we now describe. Kummer surfaces of product type. Let C 1 and C 2 be elliptic curves and construct the Kummer surface as X = Km(C 1 × C 2 ). We put the 2-torsion points of C 1 (resp. C 2 ) as {b 1 = 0, b 2 , b 3 , b 4 } (resp. {c 1 = 0, c 2 , c 3 , c 4 }) and denote by δ the natural rational map of C 1 × C 2 to X. Let E k (resp. F k ) be the image of C 1 × {c k } (resp. {b k } × C 2 ) by δ. Then X has the configuration of 24 smooth rational curves as in Figure 1, where G ij is the exceptional curve corresponding to (b i , c j ) ∈ C 1 × C 2 . Sometimes it is called the double Kummer configuration. In the following we introduce two kinds of free involutions on X with parameters. Involutions of Lieberman type. Let a = (b i , c j ) ∈ C 1 × C 2 be a 2-torsion point Figure 1: the double Kummer configuration not lying on C 1 × {0} or {0} × C 2 . Let τ (resp. ρ) be the involution of X induced by the automorphism (− id C 1 , id C 2 ) (resp. the translation by a) of C 1 × C 2 . Then σ = τ ρ is a free involution. We remark that a has the parameter i, j with 2 ≤ i, j ≤ 4. Involutions of Kondo-Mukai type. Let τ be as above. X/τ is isomorphic to P 1 ×P 1 with sixteen points blown up, which correspond to G ij 's in Figure 1. We regard P 1 ×P 1 as a smooth quadric in P 3 so that there are morphisms Choose two permutations I = {i 1 , i 2 , i 3 , i 4 } and J = {j 1 , j 2 , j 3 , j 4 } of {1, 2, 3, 4} and we put g k = ε(G i k j k ) ∈ P 1 × P 1 . We project P 1 × P 1 onto a hyperplane ∼ = P 2 of P 3 from g 4 . This birational map ε ′ contracts two curves whose images we denote by P and Q ∈ P 2 . Let g k = ε ′ (g k ), k = 1, 2, 3. Then we obtain two involutions on X: One is the covering transformation ν of the degree 2 map X → P 2 . The other involution µ is induced by the unique Cremona transformation of P 2 centered at g 1 , g 2 , g 3 interchanging P and Q. Then σ = νµ is a free involution. It can be shown that σ depends only on the set {G i k j k } 1≤k≤4 . See [4] for such canonical description of this involution. Therefore the parameter is the set {G i k j k } 1≤k≤4 . There are twenty-four choices of parameters in appearance, but we will reveal that there are essentially six. If C 1 and C 2 are chosen to be general enough, we see that We will call such X a generic Kummer surface of product type. In the following main theorem of this section we classify the all free involutions on X. Theorem 4.1. Let X be a generic Kummer surface of product type. Then X has exactly fifteen distinct Enriques quotients which are naturally in one-to-one correspondence with nonzero elements of A N S(X) . Moreover all of them can be geometrically constructed from the preceding examples by choosing appropriate parameters. We remark that the Lieberman involutions correspond to nine elements of norm 0 of A N S(X) and Kondo-Mukai involutions to six elements of norm 1. In the rest of this section we prove Theorem 4.1. First we determine the isomorphism classes of primitive embeddings of M : 4 . This step is purely lattice-theoretic. We use the following theorem of Nikulin [ (2) There are subgroups Γ M ⊂ A M , Γ S ⊂ A S and a sign-reversing isometry γ : Γ M → Γ S such that if Γ is the pushout of γ, then Assume M is a primitive sublattice of S which is orthogonal to no (−2) vectors of S. (2) There are exactly two such primitive sublattices up to the action of O(S). P roof. By Theorem 4.2 (2), A N is a 2-elementary abelian group and q N takes only integral values on A N . On the other hand, N is a negative definite lattice of rank 8. This implies by the decomposition theorem of 2-elementary finite quadratic forms (see [7, Proposition 1.8.1]). Checking the signature of (A N , q N ), we see that the latter does not occur. In all other cases, we find that N has a unimodular overlattice of rank 8, i.e., E 8 . The index [E 8 : N] is given by 2 a . Claim. Let N ⊂ E 8 be an overlattice and assume N contains no (−2) vectors. Then [E 8 : N] ≥ 9. P roof. We take a basis of E 8 as in Figure 2. Consider the elements f 0 := 0, f j := e 1 + · · · + e j , 1 ≤ j ≤ 7 and f 8 := 2e 1 + 3e 2 + 4e 3 + 5e 4 + 6e 5 + 4e 6 + 2e 7 + 3e 8 . It is easy to see that any difference f j − f i , 0 ≤ i < j ≤ 8 has norm −2. This means that every f i is a distinct element of the reminder class group E 8 /N. Thus we obtain A N = u(2) ⊕4 . Therefore N has the maximal number of minimal generators of the discriminant group. This implies that for any n ∈ N, n/2 ∈ N * . In particular N(1/2) is a unimodular lattice. Thus we see that N ∼ = E 8 (2) and (1) Claim. Let M 1 and M 2 be two primitive sublattices of S satisfying the assumption. For each k = 1, 2, we use the same notation as above, indexed by k. If q N 1 (z N 1 ) = q N 2 (z N 2 ), then there exists ϕ ∈ O(S) which transforms M 1 onto M 2 . P roof. By Witt's theorem on the finite quadratic forms, there exist ψ M and ψ N fitting in the square inside. Again by Witt's theorem we can extend ψ M (resp. ψ N ) to ϕ M (resp. ϕ N ) in the diagram. By the surjectivity of σ M and σ N mentioned above, these isomorphisms lift to an isomorphism between M 1 ⊕ N 1 and M 2 ⊕ N 2 which preserves the overlattice S. This was the assertion. q.e.d. Conversely assume the patching elements v M ∈ A N S concide. By Proposition 4.3, the primitive embeddings M 1 and M 2 are isomorphic and there exists ϕ ∈ O(NS) such that ϕ(M 1 ) = M 2 , namely ϕi M 1 ϕ −1 = i M 2 . By assumption, σ N S (ϕ)(v M ) = v M . We can assume ϕ ∈ O ↑ (NS) by replacing ϕ by −ϕ if necessary. According to Proposition 1.3 (2), ϕ = wψ where w ∈ W X , ψ ∈ O + (NS). Then Lemma 1.4 (2) implies ψ(M 1 ) = (M 2 ). On the other hand, since w acts on A N S trivially, σ N S (ψ) = σ N S (ϕ) and this element fixes v M . We apply Theorem 2.3, Step 3 to M = M 1 , ψ 1 = ψ, ψ 2 = id N S . Both ψ 1 and id stabilize v M , therefore their images by σ N S have the same class in O(q N S )/σ N S (K) by the previous lemma. Thus the conclusion holds. q.e.d. Next we compute the patching elements of involutions of Lieberman and Kondo-Mukai. They involve parameters as mentioned in the beginning of this section and we have to consider the dependence of patching elements on the parameters. This is directly done. We take the following basis of A N S . e 1 = (G 11 + G 13 + G 31 + G 33 )/2, f 1 = (G 22 + G 23 + G 32 + G 33 )/2, e 2 = (G 21 + G 23 + G 31 + G 33 )/2, f 2 = (G 12 + G 13 + G 32 + G 33 )/2. Then the result is as in Figures 3,4. (1) The involution τ acts trivially on NS(X). So it induces a numerically trivial involution on the fifteen Enriques quotients. The Kondo-Mukai case of this is the last and missing result of [5], first found in [3]. When NS(X) ∼ = U(2) ⊕ E ⊕2 8 , the Barth-Peters case, the number B 0 is equal to 1. When NS(X) ∼ = U(2) ⊕ E 8 (2), then X has only one Enriques quotient. Finally using the result of [5], we see that in other 2-elementary cases X has no Enriques quotients. (3) The generators of the whole automorphism group Aut(X) are found in [2].	2009-09-29T14:56:58.000Z	2007-03-01T00:00:00.000	{ "year": 2009, "sha1": "e96c3639083831aafe34fd711b5dc5c4ce79a3fb", "oa_license": null, "oa_url": "http://www.ems-ph.org/journals/show_pdf.php?iss=1&issn=0034-5318&rank=10&vol=43", "oa_status": "BRONZE", "pdf_src": "Arxiv", "pdf_hash": "76556cafdab66b9e3f5925095707eed9987e66f2", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [ "Mathematics" ] }
118990440	pes2o/s2orc	v3-fos-license	The Carbon-Rich Gas in the Beta Pictoris Circumstellar Disk The edge-on disk surrounding the nearby young star Beta Pictoris is the archetype of the"debris disks", which are composed of dust and gas produced by collisions and evaporation of planetesimals, analogues of Solar System comets and asteroids. These disks provide a window on the formation and early evolution of terrestrial planets. Previous observations of Beta Pic concluded that the disk gas has roughly solar abundances of elements [1], but this poses a problem because such gas should be rapidly blown away from the star, contrary to observations of a stable gas disk in Keplerian rotation [1, 2]. Here we report the detection of singly and doubly ionized carbon (CII, CIII) and neutral atomic oxygen (OI) gas in the Beta Pic disk; measurement of these abundant volatile species permits a much more complete gas inventory. Carbon is extremely overabundant relative to every other measured element. This appears to solve the problem of the stable gas disk, since the carbon overabundance should keep the gas disk in Keplerian rotation [3]. New questions arise, however, since the overabundance may indicate the gas is produced from material more carbon-rich than the expected Solar System analogues. 2 CIII) and neutral atomic oxygen (OI) gas in the β Pic disk; measurement of these abundant volatile species permits a much more complete gas inventory. Carbon is extremely overabundant relative to every other measured element. This appears to solve the problem of the stable gas disk, since the carbon overabundance should keep the gas disk in Keplerian rotation 3 . New questions arise, however, since the overabundance may indicate the gas is produced from material more carbon-rich than the expected Solar System analogues. β Pic is an A5 V star, indicating that it is approximately 1.8 times more massive than the Sun. It appears to have solar elemental abundances 4 and is 8 -20 million years old 5 . Narrow atomic absorption lines were studied in the spectra of β Pic even before the star was known to have a dust disk 6 . The absorption features fall into one of two categories. Almost every line observed shows an unvarying narrow absorption at the radial velocity of the star. This component contains the bulk of the circumstellar gas and is called the stable component 1 . On the wings of most absorption lines are broad, variable absorption features, which are typically redshifted with respect to the star. These features arise from gas falling toward the star at high velocity, likely produced by vaporization of star-grazing planetesimals 7 . Various studies of the dynamics of the β Pic stable gas have shown that there is a serious gap in our understanding. The force of stellar radiation pressure on many of the observed atomic and ionic species should rapidly blow them away from the star. The very existence of gas at the velocity of the star (the bulk of the β Pic circumstellar gas) is puzzling. Workers were forced to postulate the existence of an undetected, dense hydrogen torus to brake some species through gas drag 1 . The problem got worse after the detection of resonantly scattered emission from gaseous species in the β Pic disk 8, 2 . These spatially 3 resolved spectra show a gas disk in Keplerian rotation between 13 AU and a few hundred AU from the star. Analysis of the gas dynamics demonstrated that the some of the observed species should not be in Keplerian rotation, unless the total gas mass is much greater than the observed gas mass 2 . We used spectra from the Far Ultraviolet Spectroscopic Explorer (FUSE) to measure the volatile species CII, CIII, and most importantly, OI, in the circumstellar disk. These species are seen in absorption against broad stellar emission lines ( Figure 1). Models of the absorption lines were compared to the data and the best absorption parameters ( Table 1) determined through χ 2 minimization (details of the observations, models, and analysis appear in the online Supplementary Information). Combined with a previous measurement of CI gas 9 , the CII measurement completes the inventory of stable carbon gas in the disk. OI is the only ionization state of oxygen observed to date, and the only one likely to be abundant, given the high first ionization energy of oxygen. We also used archival Hubble Space Telescope Space Telescope Imaging Spectrograph (STIS) high-resolution far-UV spectra to search for absorption lines from a few species that have not yet been observed (CrI, NiI, PI, and PII), but none were detected. Previous observations of the β Pic gas were primarily of metallic species 1 . The addition of our new measurements provides the most complete inventory of the gas in any debris disk. The β Pic elemental abundances, found by summing the abundances of the various ionization stages of each element, are shown in Table 2, along with the elemental abundances for the Sun 10 , a carbonaceous chondrite meteorite 10 , and comet Halley dust 11 . This table utilizes results from studies of the β Pic gas stretching over more than ten years 1,2,9,12,13,14 (see the online Supplementary Information for details). A plot of the elemental abundances from Table 2 appears in Figure 2. It shows that the composition of the β Pic gas 4 is dissimilar to the composition of all three comparison bodies. In particular, carbon is extremely overabundant relative to every other measured element. The β Pic C/O and C/Fe ratios are 18 and 16 times the solar value, respectively. The lithophile elements (e.g. Mg, Al) have roughly solar abundance relative to each other, as do the siderophile elements (e.g. Fe, Ni). However, the lithophile elements as a group seem slightly underabundant relative to the siderophile elements. Neither oxygen nor carbon feels strong radiation pressure in the β Pic disk, since the star is relatively faint in the far-UV where the strong absorption lines of these species lie. By contrast, metals such as Na and Fe feel extremely strong radiation pressure and could be blown out of the system, producing apparent C and O overabundances relative to these elements. But carbon is overabundant relative to every other element, including ones that feel similarly weak radiation pressure, like oxygen. This leads us to suspect that the overabundance reflects the composition of the parent material, which would have to be more carbon-rich than the obvious Solar System analogue materials. The bulk composition of the Beta Pic planetesimals might be carbon-rich. There was a large increase in organics relative to water in the material excavated from comet Tempel 1 during its collision with Deep Impact 15 . Or the planetesimals might be selectively outgassing volatile carbonaceous compounds, trace amounts of which are still found in carbonaceous chondrite meteorites (J. Nuth, personal communication). There are probably many other possible explanations. A new analysis of the gas dynamics in β Pic which took into account Coulomb forces between gaseous ions, neutral atoms, and charged dust showed that these interactions increase the effective collision cross-sections, dramatically reducing the need for additional unobserved braking gas 3 . But unless the dust grains are very highly charged, additional braking gas is still required. However, that analysis assumed solar abundances in the gas. If 5 one considers our measured midplane abundances in conjunction with Coulomb forces, the problem may be solved. The ratio of the radiation and gravitational forces felt by an atom is the radiation pressure coefficient (β). The β-value for a particular species depends on the brightness of the central star at the wavelengths of the absorption lines of the species. Neutral elements that feel strong radiation pressure are quickly ionized before they can be blown away from the star; therefore, the species that must be braked are ionized. Coulomb interactions between the ions couple them together into a single fluid 3 . If the effective radiation pressure coefficient of the fluid (β eff ) is < 0.5, it will be bound to the star and the gas will be self-braking. Since carbon is abundant and moderately ionized in the β Pic disk, it is an important constituent of the ionic fluid. β Pic is faint in the far-UV, where the strong CI and CII absorption lines lie, so carbon feels negligible radiation pressure in the β Pic disk. Increasing the carbon abundance lowers β eff . The ionic fluid will be self-braking (β eff < 0.5) if the carbon abundance is enhanced by a factor of 10 or more over the solar abundance 3 . Carbon in the β Pic stable gas is enhanced relative to oxygen by a factor of 18 over the solar C/O abundance, so additional unseen gas or highly charged grains are not needed to keep the gas disk in Keplerian rotation. The reason why the gas composition is just what is required for Keplerian rotation is unknown. Answering this question will require similar observations and analyses to be done for many debris disks. However, FUSE spectra of another debris disk may provide a hint 16 . Circumstellar gas and dust have been detected around the B9 primary star of the σ Herculis binary system. In this case, the gas is moving away from the star. It has been suggested that the gas and dust are produced by collisions among planetesimals located about 20 AU from the star, where the orbits of small bodies become unstable in the binary system. A radiatively driven wind is generated as the high radiation pressure from this UVbright star blows away the circumstellar gas. This might be the more common fate of debris 6 gas around a high mass star, and perhaps would have occurred in the β Pic disk if not for the overabundance of carbon in its circumstellar gas. Of course, this does not explain how the carbon overabundance occurred in the first place. A great deal more observational and theoretical work is required to determine if the gas composition reflects the bulk composition of the parent bodies and then determine how such carbon-rich planetesimals could have formed in the β Pic protoplanetary disk. Table 2. Elemental Abundances in the β Pic Stable Gas The total abundance for each element in the β Pic stable gas (column 2) was found by summing the column densities of the various ionization stages of the element, which appear in Table 2	2019-04-14T01:45:37.806Z	2006-04-19T00:00:00.000	{ "year": 2006, "sha1": "9e6c780cd6f5c53d3bb4f2a973d5499146073d4b", "oa_license": null, "oa_url": "http://arxiv.org/pdf/astro-ph/0604412", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "9e6c780cd6f5c53d3bb4f2a973d5499146073d4b", "s2fieldsofstudy": [ "Physics", "Environmental Science" ], "extfieldsofstudy": [ "Physics" ] }
238420203	pes2o/s2orc	v3-fos-license	A Prediction Model of Extubation Failure Risk in Preterm Infants Objectives: This study aimed to identify variables and develop a prediction model that could estimate extubation failure (EF) in preterm infants. Study Design: We enrolled 128 neonates as a training cohort and 58 neonates as a validation cohort. They were born between 2015 and 2020, had a gestational age between 250/7 and 296/7 weeks, and had been treated with mechanical ventilation through endotracheal intubation (MVEI) because of acute respiratory distress syndrome. In the training cohort, we performed univariate logistic regression analysis along with stepwise discriminant analysis to identify EF predictors. A monogram based on five predictors was built. The concordance index and calibration plot were used to assess the efficiency of the nomogram in the training and validation cohorts. Results: The results of this study identified a 5-min Apgar score, early-onset sepsis, hemoglobin before extubation, pH before extubation, and caffeine administration as independent risk factors that could be combined for accurate prediction of EF. The EF nomogram was created using these five predictors. The area under the receiver operator characteristic curve was 0.824 (95% confidence interval 0.748–0.900). The concordance index in the training and validation cohorts was 0.824 and 0.797, respectively. The calibration plots showed high coherence between the predicted probability of EF and actual observation. Conclusions: This EF nomogram was a useful model for the precise prediction of EF risk in preterm infants who were between 250/7 and 296/7 weeks' gestational age and treated with MVEI because of acute respiratory distress syndrome. INTRODUCTION With the development of perinatology, there has been a rapid increase in the survival rate of premature neonates annually, followed by the increasing incidence of neonatal respiratory distress syndrome (RDS). Although early use of pulmonary surfactant (PS) and non-invasive ventilation [such as continuous positive airway pressure (CPAP), synchronized nasal intermittent positive pressure (SNIPP), bilevel continuous positive distending pressure (BiPAP), etc.] have been proven effective in the prevention and treatment of RDS, some premature neonates with acute RDS still need mechanical ventilation through endotracheal intubation (MVEI). In addition, MVEI is a vital way to save premature neonates after PS and non-invasive ventilation failed. Fifty percent of neonates born at <28 weeks' gestation who received CPAP failed, and 37% of those who received PS failed (1). More than half of neonates born at <29 weeks' gestation required mechanical ventilation (2). Study Participants Neonates who were born between June 2015 and December 2020 and met the inclusion criteria were recruited from three medical centers, including the First hospital of Anhui Medical University, Second Xiangya Hospital of Central South University, and First Hospital of Jilin University. The neonates recruited from the First Affiliated Hospital of Anhui Medical University born between June 2015 and December 2019 served as the training cohort, and the neonates recruited from all three medical centers born between January 2020 and December 2020 served as the validation cohort. The inclusion criteria were as follows: (1) 25 0/7 to 29 6/7 weeks' gestational age; (2) born in the obstetrics department at the same hospital with NICU, transferred to NICU immediately after birth and treated with MVEI within the first 24 h of postnatal age (to reduce the heterogeneity of the study population). The exclusion criteria were as follows: (1) died before extubation; (2) accidental extubation; (3) received surgical treatment within 5 days after extubation; (4) congenital malformation of vital organs or genetic/metabolic diseases; (5) the lack of important clinical data. Data Collection We collected clinical characteristics (demographics, perinatal characteristics, and peri-extubation characteristics) and clinical outcomes of the infants from the medical records. Demographics were collected, including gender, gestational age (GA), birth weight (BW), small for gestational age (SGA), assisted reproduction, cesarean delivery, multiple births, and maternal age at pregnancy. Perinatal characteristics were collected, including hypertensive disorder complicating pregnancy of the mother, gestational diabetes mellitus of mother, antenatal steroids, prolonged rupture of the membrane [>18 h], 1-min Apgar score, 5-min Apgar score, intubated in the delivery room, epinephrine in the delivery room, PS administration, hypoglycemia, arterial blood gases [pH, arterial partial pressure of carbon dioxide (PaCO 2 )] within 0.5 h before intubation, non-invasive ventilation before intubation, age of starting enteral feeding, and early-onset sepsis (EOS). Peri-extubation characteristics were collected, including ventilator mode at extubation [high-frequency ventilation (HFV), conventional mechanical ventilation (CMV)], hematology examination [hemoglobin concentration (HB), serum albumin (ALB), serum globulin (GLB)] within 48 h before extubation, arterial blood gases within 0.5 h before extubation (pH, PaCO 2 ), highest fraction of inspired oxygen (FiO 2 ) after intubation, FiO 2 at extubation, first MVEI duration, weight at extubation, postmenstrual age (PMA) at extubation, and caffeine treatment. Clinical outcomes were collected, including death, BPD, neonatal necrotizing enterocolitis (NEC), late-onset sepsis (LOS), VAP, pulmonary hemorrhage, pneumothorax, retinopathy of prematurity (ROP), age at total enteral feeding, length of time on MVEI, length of time on hospital stay, length of time on oxygen, antibiotic utilization rate, daily weight gain, and daily hospitalization cost. Definitions EF was defined as the need for reintubation within 5 days after extubation. ES was defined as survival for more than 5 days without MVEI. Intubation/reintubation and extubation criteria were based on the "The Neonatal Mechanical Ventilation Routine" (15) that was published in May 2015. The decisions regarding initial endotracheal intubation, the timing of extubation, post-extubation support, and the need for reintubation were at the discretion of the chief physician. The neonate could be intubated if they met any of the following criteria: (1) frequent apneas, medication or noninvasive ventilator intervention was ineffective; (2) neonates with RDS required PS treatment; (3) FiO 2 > 0.6-0.7, arterial partial pressure of oxygen < 50-60 mmHg or transdermal oxygen saturation < 85%; (4) PaCO 2 > 60-65 mmHg, with persistent acidosis (pH < 7.20). The reintubation criteria were the same as those for initial intubation. If an infant using CMV met all of the following criteria, then extubation was done: (1) the primary disease of the neonate improved, the infection was controlled, and the general condition (including stable breathing, stable oxygen saturation, and active response) was good; (2) Peak inspiratory pressure ≤18 cm H 2 O, positive end-expiratory pressure at 3-6 cm H 2 O, breathing rate at 12-20 breaths/min, FiO 2 ≤ 0.4. There was no uniform weaning standard for HFV. Before weaning from HFV, FiO 2 was lowered, then MAP was lowered, and finally, amplitude was adjusted according to the PaCO 2 . If HFV was being used, it was usually transferred to CMV and then weaned. In this study, non-invasive ventilator was either continued after extubation, or withdrawn directly. Hypoglycemia was defined as peripheral blood glucose <40 mg/dL after birth. Sepsis was defined as systemic inflammatory reaction syndrome caused by various pathogens, including culture-positive and culture-negative septicemia. Definition of culture-negative sepsis refers to culture-negative, a clinical manifestation of a systemic infection, antibiotics ≥3 days, laboratory findings (WBC < 5 × 10 9 /mL or ≥ 30 × 10 9 /L, CRP> 10 mg/L). Early-onset sepsis refers to sepsis that occurs within 3 days after birth; otherwise, it was known as late-onset sepsis. In this study, caffeine treatment implied the use of caffeine citrate before the first extubation and was maintained for over 2 more days after extubation. The method of administration was the first loading dose of 20 mg/kg, and the maintenance dose of 5 mg/kg·d. BPD was defined for infants treated with > 21% oxygen for ≥28 days and/or at 36 weeks postmenstrual age. The diagnosis of pneumothorax is confirmed by a bedside chest radiograph. VAP refers to pneumonia that occurs when infants receive mechanical ventilation for more than 48 h. NEC is defined as stage II and III of correction Bell's staging system (16). ROP was defined as any Stage 3 ROP with plus disease, and Stages 4 or 5 of ROP (17). Statistical Analysis The preterm infants in the training cohort were divided into EF groups and ES groups according to whether to reintubate within 5 days after extubation. For the infants who failed extubation, Kaplan-Meier survival curves were plotted to explore the timing of EF. Univariate logistic regression analysis was performed to identify the risk factors for EF. All P-values were two-sided and P < 0.05 were considered significant. Among all factors that were confirmed to be statistically significant, such as antenatal steroids, 5-min Apgar score, EOS, hemoglobin concentration before extubation, pH before extubation, PCO 2 before extubation, and caffeine treatment before and after extubation, stepwise discriminant analysis [method: Wilk's lambda; criteria: use probability of F (entry of 0.1 and removal of 0.15)] was performed to select the useful combination of factors that could precisely predict EF. A nomogram for EF was created based on the multivariate logistic regression model using the selected factors. Receiver operator characteristic (ROC) curve was performed to evaluate the predictive value of the model. Next, we carried out Harrell's concordance index (C-index) and calibration plot in both training and validation cohorts to assess the performance of the model. A concordance index is a numerical measure of discriminative ability and calibration plots are graphic evaluations of predictive ability that compare observed probabilities with nomogram-predicted probabilities (18). Finally, we compared the clinical outcomes of the preterm infants in both the groups. We used the χ 2 test or Fisher's exact test to compare the categorical variables, and Mann-Whitney Utest to compare the continuous variables. Statistical analysis was conducted using the IBM SPSS Statistics (version 25.0; SPSS Inc., Chicago, IL) and the R software (Version 4.0.3; http://www.Rproject.org). Timing and Cause of EF In the training cohort, 55/128 (43.0%) of the infants who received MVEI treatment were reintubated during hospitalization, and 35/55 (63.6%) were reintubated within 5 days after extubation ( Figure 1A), the median reintubation time was 2.98 days with an IQR of 1.34-9.17 days. Based on whether reintubation was required within 5 days after extubation, all infants of the training cohort were divided into the EF group (n = 35) and the ES group (n = 93). In the EF group, 28/35 (80%) infants were reintubated within 3 days (Figure 1B) Factors Associated With EF In clinical practice, ventilator mode at extubation and the administration of PS might affect the occurrence of EF. In this study, 96.09% of infants were weaned with CMV, 3.91% were weaned with HFV. Among those who were weaned Table 1 presents the clinical characteristics of infants, including perinatal and peri-extubation characteristics in the two groups. In univariate logistic regression analysis, antenatal steroids (95% CI: 0.16-0.90), 5-min Apgar score (95% CI: 0.54-0.90), EOS (95% CI: 1.20-6.17), hemoglobin concentration before extubation (95% CI: 0.96-1.00), pH before extubation (95% CI: 0.00-0.15), PCO 2 before extubation (95%CI: 1.01-1.07), and caffeine treatment (95% CI: 0.11-0.55) were identified as the risk factors of EF. We selected 5-min Apgar score, EOS, pH before extubation, hemoglobin concentration before extubation, and caffeine treatment as the best combination of factors to predict EF through stepwise discriminant method. A nomogram for EF was constructed based on the multivariate logistic regression model using these selected factors (Table 2; Figure 2). Clinical Outcomes of Two Groups Compared with the outcomes of the two groups, infants in the EF group had a higher rate of mortality ( Table 3). The above indicators were statistically significant (P < 0.05). Prediction Model for EF The nomogram which could predict the probability of EF was created using the factors of 5-min Apgar score, EOS, pH before extubation, hemoglobin concentration before extubation, and caffeine treatment (Figure 2). The area under the ROC curve was 0.824 (95% CI: 0.748-0.900; Figure 3). The model with highest Youden index defined EF risk as 0.207, the sensitivity is 0.829, the specificity is 0.667, the positive predictive value is 0.483, and the negative predictive value is 0.912. The usage of the predictive model was drawn with an assumptive preterm infant with a 5-min Apgar score of 6, EOS, HB before extubation of 160 g/L, pH before extubation of 7.35, and no caffeine treatment. The points for 5-min Apgar score, EOS, HB before extubation, pH before extubation, and caffeine treatment were 50, 53, 25, 0, and 60, respectively. The total point added up to 188 for this infant, which represented ∼0.56 of the probability of EF. The performance of this nomogram was assessed by C-index and calibration plots. Next, we collected 58 preterm infants who met the inclusion and exclusion criteria as a validation cohort to validate our model. The C-index in the training and validation cohorts was 0.824 and 0.797, indicating a high discrimination of the nomogram. The calibration plots also showed high coherence between the predicted probability of EF and actual observation (Figure 4), which indicated good calibration of the model. DISCUSSION Infants with EF are known to be more likely to have a higher rate of mortality and morbidity, including VAP, pneumothorax, BPD, neurological damage, etc. Currently, the decision to extubate is based on clinicians' discretion and is influenced by clinical evaluation, blood gas variables, oxygen requirement, and level of ventilator support. There is a lack of strong evidence to support the use of any predictor of extubation readiness in preterm infants over clinical judgment alone. Although current studies on spontaneous breathing trials are highly sensitive to ES, they provide little benefit in the identification of EF (13,14). Notably, studies concentrating on predicting the EF in preterm infants account for less of the studies related to ventilators. The existing studies have shown a high rate of EF in preterm infants. Chawla (2) showed the reasons for reintubation within 7 days after extubation included significant apnea and bradycardia, pulmonary hemorrhage, respiratory acidosis, pneumothorax, lung collapse, and increased work of breathing; cases who were reintubated >7 days after extubation were due to infectionassociated episodes (culture-proved sepsis, pneumonia, NEC, etc.). In this study, we defined EF as needing reintubation within 5 days after extubation to include more neonates with EF and to avoid new disease becoming the cause of reintubation. Similar to the previous studies (1, 12), a lower 5-min Apgar score, lower pH before extubation were associated with a higher likelihood of EF, which implied that infants were more likely to need reintubation. The lower 5-min Apgar score indicated that the hypoxia period might be longer (might last for several min or longer, worse than the lower 1-min Apgar score) in a neonate's early life. Also, the neonate's lung and brain might suffer from serious hypoxia injury, which may affect the recovery of respiratory function. Blood gas pH before extubation is an important marker for extubation readiness. The lower pH indicates that the current oxygen exchange capacity of the lung cannot meet the body's demand for oxygen supply. Our study found that treatment with caffeine prior to extubation and maintained for more than 2 days after extubation reduced the risk of EF compared with non-use of caffeine. Caffeine administration can quickly and effectively increase the sensitivity of the respiratory tract to carbon dioxide, reduce the frequency of apneas, increase the stroke volume of the left ventricle, increase the minute ventilation and tidal volume, and improve lung function, thereby improving the blood oxygen partial pressure in the arteries and reducing the failure rate of weaning (20). The use of caffeine before and after extubation can reduce the occurrence of recurrent respiration and hypoxemia, improve the respiratory function of premature neonates, increase the pH value and decrease the PaCO 2 , and reduce the incidence of apnea and reintubation (21). Amaro et al. (22) noted that caffeine might increase the rate of mortality in neonates, but the reliability of the results was challenged due to the small sample size and the absence of statistical significance. A meta-analysis (23) of six randomized controlled trials, including 620 neonates, suggested that high-dose caffeine could reduce the risk of death. Although studies have shown that the use of caffeine might be beneficial for the long-term prognosis of preterm infants, the timing of caffeine administration is still controversial (22,24). Our analysis results proved that the rate of EF was lower if the caffeine is maintained for over 2 more days after extubation. However, it has failed to give an optimal maintenance time to minimize the rate of EF, and further research is needed to explore it in the future. Our study showed that EOS was associated with EF. Similar to our research, Capasso et al. (25) found that LOS was related to failure of non-invasive support. EOS and LOS have differences in high-risk factors and pathogenic bacteria. In our study, EF occurred mostly after the diagnosis of EOS. Therefore, our research suggests that EOS may play a role in facilitating EF. However, the study of Capasso et al. did not specify the time sequence of LOS and failure of non-invasive support. It is possible that EF increases LOS, or that LOS causes EF, or the two influence each other. We speculated that the relationship between EOS and EF might have the following mechanism. First, sepsis might manifest as dyspnea, apnea, cyanosis, etc., among which apnea or respiratory distress could be the starting manifestation of EOS (26). For neonates who were extubated within 3 days of birth, they might be reintubated due to sepsis. Second, inflammatory factors can attack immature lung tissues during inflammatory storms (27,28). Once the alveolar cell and lung interstitial tissue were damaged by inflammation, the ventilation function and pulmonary vascular hemodynamics were affected, and this damage might be irreversible. Third, sepsis might be complicated by encephalitis, leading to dysfunction of the respiratory center. EOS in our study included culture-positive and culturenegative septicemia. Many neonates are diagnosed with a "probable or possible" sepsis or a "presumed symptomatic infection but no bacterial cause identified"; conditions often referred to as "culture-negative sepsis" (29). Robust epidemiological data on culture-negative sepsis are limited, especially in younger gestational age preterm infants. The incidence of EOS is substantially higher in preterm infants (30). A study in Norway showed that the incidence of negative blood culture and culture-proven EOS in term newborns was 15.12% (31). The incidence of EOS in our study was 29.7% (38/128). The infants in our study have two high-risk factors: very preterm or extremely preterm (gestational age of 25 0/7 ∼29 6/7 weeks), invasive procedures (100% with endotracheal intubation; most with an indwelling nasogastric tube, umbilical vein catheter, or peripheral venous catheter). Thus, the higher incidence of EOS seems reasonable. This study provides data on the incidence of EOS, including culture-positive and culture-negative septicemia, on the condition that epidemiological data are limited. Our study showed that the lower HB within 48 h before extubation was associated with EF, which was not mentioned in the published literature that we can retrieve. The result suggests that if infants with invasive ventilation have anemia, indications for red blood cells transfusion might need to be appropriately broadened. As we know, hemoglobin is responsible for transporting oxygen and carbon dioxide as the important component of red blood cells. Previous studies have reported a significant association between anemia and respiratory disease. Watanabe et al. (32) showed that hemoglobin concentration was probably related to brain development. Duan et al. (33) demonstrated that early anemia (≤14 days of postnatal life) was associated with an increased risk of BPD in preterm infants. Hussain et al. (34) reported that low HB levels were a risk factor for acute lower respiratory tract infections. In traditional physiology, anemia is associated with an increase in heart rate and cardiac output. We speculate that low HB before extubation was associated with EF because of its influence on the respiratory center, lung, and hemodynamics. Lower HB might decrease oxygen delivery to the respiratory center of the brain and cause respiratory symptoms, including tachypnea, dyspnea, and apnea, which may increase days of mechanical ventilation. Low HB levels reduce the ability of blood to transport oxygen from the lungs to the tissues, leading to increased anaerobic glycolysis and increased production of lactic acid, resulting in a compensatory increase in respiratory rate and heart rate, and an increased burden on the heart and lungs. Our results were not consistent with the study by Manley et al. (11) on extremely premature neonates (n = 174, <28 weeks GA), which found that higher GA was associated with ES. In univariate regression analysis, the 95% CI of gestational age was 0.60-1.13. This was probably due to the composition of the research population. In China, the termination of pregnancy with a fetal gestational age of fewer than 28 weeks is defined as abortion in obstetrics. It often happens that fetuses with lower gestational age or weaker viability after birth are more likely to be aborted or be abandoned treatment by their parents. Therefore, the smaller the gestational age of a fetus was, the less likelihood of the fetus being transferred to our NICU and received MVEI. The relationship between GA and EF needs to be further confirmed. The primary limitation of this study is that the included predictors were not comprehensive. Researches by Ding et al. (35) and Capasso et al. (25) showed that different noninvasive ventilation modes (such as SNIPP, CPAP, BiPAP, etc.) after extubation had different reintubation rates. However, due to the limited study conditions, the non-invasive ventilation modes were not included in our analysis. Future studies can conduct randomized controlled trials to further explore the relationship between different non-invasive ventilation modes and EF. Intraventricular hemorrhage (IVH) and patent ductus arteriosus (PDA) are known to occur more frequently in preterm infants (36,37). Thus, the damage of IVH to nerve function and the influence of PDA on hemodynamics might also be risk factors for EF. Our NICU did not routinely perform bedside echocardiographic and intracranial ultrasound until October 2017. Only infants with high suspicion of heart or brain problems would qualify for bedside ultrasound by the time. Thus, there was lack of data on cardiac and cranial imaging in the medical records of some infants before the first extubation. Considering the accuracy of the study, we did not include PDA and IVH in the analysis. Future studies could focus on the relationship between IVH, PDA and EF. The secondary limitation of this study is the small sample size due to the following reasons: (1) The number of preterm infants is small, especially lower gestational age preterm infants. And extremely premature infants were usually abandoned treatment; (2) The INSURE approach is the best choice for most premature infants with mild RDS and reduces the need of MVEI in some severe RDS infants; (3) Strict inclusion criteria were set to reduce the heterogeneity of the study population. Multi-center prospective and larger sample size of studies should be performed to verify the finding and find more predictors. The high C-index and the good coherence between the predicted probability and actual observation of the model indicate that the modeling has better performance. The model with highest Youden index defined EF risk as 0.207 and predicted extubation failure with a sensitivity of 0.83 and a specificity of 0.67. The study of Shalish et al. (13) shows that spontaneous breathing trials with highest Youden index predicted extubation success with a sensitivity of 0.93 and a specificity of 0.39. The two can complement each other as extubation preparations. The strengths, limitations and improvement directions of our model are listed as follows. The main strengths: (1) The nomogram can rely on a user-friendly digital interface to assist doctors in making clinical decisions. And this is the first study that used a nomogram to predict EF in preterm infants, which was demonstrated good discrimination and calibration power in the validation; (2) The model could facilitate the individualized prediction of EF using five available variables. Clinicians could combine the infant's individual condition with available predictors to prepare for extubation. The main limitations: (1) Our cohort was not representative of all neonates with MVEI. Different countries might have different intubation/reintubation and extubation practices that limit the generalizability of the model; (2) Though we have tried to make inclusion criteria and data collection more rigorous, undiscovered bias and better predictors might exist since it is a retrospective study. Improvement directions: (1) The model established in this study is the prediction model of all-cause EF. Establishing predictive models, respectively, based on the causes of EF will make it possible to predict the EF and its causes, which is conducive to take targeted preventive measures for infants at high risk of EF. (2) Discovering and incorporating more and better predictors in the model. Our study noted that neonates who failed extubation had higher rates of mortality, VAP, pulmonary hemorrhage, and other negative outcomes among survivors, including longer time on MVEI, higher antibiotic administration rate, and higher hospitalization costs. These findings are consistent with published researches that suggest a relatively poor prognosis for EF. CONCLUSIONS Our study suggested that a lower 5-min Apgar score, EOS, lower pH before extubation, lower HB before extubation, and non-use of caffeine treatment before and after extubation were the key factors that determined EF for infants treated with MVEI. Based on our research, we conclude that the following measures might reduce the rate of EF for a preterm infant with a gestational age of 25 0/7 ∼29 6/7 weeks: (1) In the delivery room, the best resuscitation teams and facilities should be ensured to improve the 5-min Apgar score; (2) After transferring to the NICU, it is important to take measures to prevent sepsis, especially for neonates with a low 5-min Apgar score; (3) Reduce unnecessary venous blood collection. When a neonate passes the spontaneous breathing trial, clinicians should monitor the neonate's HB level before weaning. If the 5-min Apgar score is found to be low or sepsis has been diagnosed, and the hemoglobin is at a relatively low level, indications for red blood cells transfusion could be appropriately broadened; (4) Extubation at pH <7.25 is not recommended; (5) It is recommended to administer caffeine before extubation and maintain it for over 2 days after extubation. Our study provides a reference basis for clinicians to choose the timing of extubation and improves the situation that extubation relies on the subjective judgment of clinicians in some way. Our cohort was not representative of all preterm infants with invasive mechanical ventilation; thus, qualified medical centers could try to organize a multi-center prospective study to verify the conclusion and explore better models. DATA AVAILABILITY STATEMENT The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. ETHICS STATEMENT The studies involving human participants were reviewed and approved by Ethics Committee of the First Affiliated Hospital of Anhui Medical University. Written informed consent from the participants' legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements. AUTHOR CONTRIBUTIONS ZC, ZD, and QZ: conception and study design, collection, analysis, and interpretation of data, and writing the first draft of the manuscript. JZ, SH, and JG: collection, analysis, and interpretation of data and drafting the manuscript for important intellectual content. YW: conception and study design, drafting the manuscript for important intellectual content, review and editing of the manuscript, and the decision to submit the paper for publication. All authors contributed to the article and approved the submitted version.	2021-10-08T05:11:49.163Z	2021-09-22T00:00:00.000	{ "year": 2021, "sha1": "cf2b568263d6d00f55d883b192e347e80ea36a45", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fped.2021.693320/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "cf2b568263d6d00f55d883b192e347e80ea36a45", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
233797928	pes2o/s2orc	v3-fos-license	Ring-opening Polymerization Reaction Mechanism of ε-Caprolactone Catalyzed by Bis(dibenzoylmethanato) zirconium(IV) Using PM3 Semi-Empirical Method Polycaprolactone (PCL) is a semi-crystalline polymer and belongs to a biodegradable plastic. PCL also can naturally degraded, has a high degree of adjustment with other polymer, and has great mechanical and thermal properties. By using the Lewis acid catalyst, PCL can be yielded through the ring opening polymerization (ROP) of caprolactone (ε-CL). The objective of this study was to explore the most likely reaction mechanism of the ROP of ε-CL using bis(dibenzoylmethanato) zirconium(IV) chloride as a catalyst. Here, the Hyper Chem 8.0 program was performed to compute the monomer molecule, intermediate molecules, and polymer molecule. The program can be launch on the Windows 07 systems. The method used was the PM3 semi-empirical method. This program also used to show the results of optimized structures. The calculation results show that to generate PCL required bis(dibenzoylmethanato) zirconium (IV) catalyst. The ROP reaction mechanism of ε-CL can be occurs through direct coordination ε-CL on the Zr complex. After that, deprotonation and insertion monomer. Last step was chain propagation of the χ–CL. Introduction Due to its advantages in the medicinal fields and packaging, poly(ԑ-caprolactone) (PCL) and the other natural polymer has been extensively studied for use in a variety of field [1][2][3][4]. In the biomedical field, PCL can be used as a drug delivery system. In the packaging field, PCL can be used as an environmentally friendly packaging [1][2][3]. Ring Opening Polymerization (ROP) technique is commonly used to synthesize polymers and this technique can also be used to synthesize polyester. Ring Opening Polymerization (ROP) of cyclic esters can be performed using cationic, anionic, active monomeric, enzymatic, and organocatalytic methods [1], [ [1], [6], aluminum metal [7], and lanthanide [8]. In ROP, the most commonly used catalysts are complex compounds. In the academia and the catalysts industry, complex compund is still also widely used until today as catalyst for ROP of cyclic esters including for ԑ-caprolactone (ԑ-CL). In these complex compounds, transition metals are used as the central atom. Some transition metals that are commonly used as the central atom include zirconium [1], [6], zinc, cuprum [9], copper [10] and titanium [11]. Apart from using transition metals, compounds with the element Sn are also widely used as catalysts in ROP [12][13]. This research will study the mechanism of ε-CL ring opening polymerization reaction using bis(dibenzoylmethanato) zirconium(IV) chloride (bis(dibzm)Zr) catalyst. The method of proof is carried out using computational calculations. The result will be obtained in the form of energy from the optimized structural geometry. The lower the energy obtained, the more likely it is to occur when laboratory experiments are carried out [5]. The results of these calculations will be input for various researchers to conduct further research at the experimental level. Materials and Method The ROP ε-CL mechanism can be predicted using semi empirical method (PM3). The method serves to calculate the geometry, energy, reaction profiles, and some stationary points (reactants, complexes, intermediate and products). These molecules are calculated in the Hyperchem 8.0 software which operated on the Windows operating system [1], [14][15][16]. Zr atoms are used as metal center in the bis(dibzm)Zr catalyst. Meanwhile, C, H, and O atoms are used as non-metal atoms in the complex compounds. In our previous research the semi empirical method is general and effective in predicting geometry and minimum energy. The procedure is performed using an RMS gradient of 0.1 kcal/(Åmol) with a maximum cycle around 2767. The calculation of the minimum energy confirms the characteristics of the reactant, intermediate and dimer ε-CL product [5]. Results and Discussion In the Figure 1, the ROP mechanism of the cyclic ester using bis(dbzm)Zr as the catalysator has been described. In the diagram, the Zr atom acts as a metal in the complex, while the ԑ-caprolaktone acts as an ester monomer. We have observed the energy of complex, intermediate, and dimer using the PM3 Semi-Empirical approach. The energy value at 0 kJ/mol represents the bis(dibzm)Zr energy. Furtheremore, the energy value of int-1, linear ε-CL, and dimer are above of the bis(dibzm)Zr energy i.e 260.52 kJ/mol, 365.08 kJ/mol and 265.28 kJ mol, respectively. Next, the energy value of complex and int-2 are on under the bis(dibzm)Zr energy i.e -225.11 kJ/mol and -54.12 kJ mol, respectively. The comparisons of energy value between the complex, intermediate state, and product were also discussed in this investigated. We have found that the ε-CL linear has the higher energy value compared with other molecule. It is due to the ε-CL linear has electron lone pair on its oxygen atom. Meanwhile, the dimer has also electron lone pair on its oxygen atom. However, because the open chain is longer, its energy value is more stable. The energy changes of the molecules which involved in the ε-CL polymerization reaction catalyzed by bis (dbzm) Zr describes in the Figure 3. Conclusion The ROP of ε-CL with bis(dibzm)Zr as a catalyst have been investigated using PM3 semi empirical method. The energy of -CL linear molecules is higher than other compounds due to the presence of electron lone pairs of oxygen atoms in the -CL linear molecules. The same thing with the int1 molecules and dimer product, the energy is higher than bis(dibzm)Zr. In contrast, the energies of int-2 and complex molecules are lower than bis(dibzm)Zr. This is because the ε-CL monomers in the molecule have lost their electron lone pair on the its oxygen atom.	2021-05-07T00:04:22.187Z	2021-03-01T00:00:00.000	{ "year": 2021, "sha1": "cc8d133547db07f6c40e3f250c0eb6ffbc84c4e4", "oa_license": null, "oa_url": "https://doi.org/10.1088/1742-6596/1811/1/012057", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "153069d0b00bafb5dfa77f33b436487eff81c7c2", "s2fieldsofstudy": [ "Materials Science", "Chemistry" ], "extfieldsofstudy": [ "Physics", "Chemistry" ] }
210998895	pes2o/s2orc	v3-fos-license	Evaluation of the effect of GM-CSF blocking on the phenotype and function of human monocytes Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a multipotent cytokine that prompts the proliferation of bone marrow-derived macrophages and granulocytes. In addition to its effects as a growth factor, GM-CSF plays an important role in chronic inflammatory autoimmune diseases such as multiple sclerosis and rheumatoid arthritis. Reports have identified monocytes as the primary target of GM-CSF; however, its effect on monocyte activation has been under-estimated. Here, using flow cytometry and ELISA we show that GM-CSF induces an inflammatory profile in human monocytes, which includes an upregulated expression of HLA-DR and CD86 molecules and increased production of TNF-α and IL-1β. Conversely, blockage of endogenous GM-CSF with antibody treatment not only inhibited the inflammatory profile of these cells, but also induced an immunomodulatory one, as shown by increased IL-10 production by monocytes. Further analysis with qPCR, flow cytometry and ELISA experiments revealed that GM-CSF blockage in monocytes stimulated production of the chemokine CXCL-11, which suppressed T cell proliferation. Blockade of CXCL-11 abrogated anti-GM-CSF treatment and induced inflammatory monocytes. Our findings show that anti-GM-CSF treatment induces modulatory monocytes that act in a CXCL-11-dependent manner, a mechanism that can be used in the development of novel approaches to treat chronic inflammatory autoimmune diseases. Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a multipotent cytokine that stimulates the proliferation of bone marrow-derived macrophages and granulocytes. Various cell types produce this cytokine, including activated T cells, monocytes/macrophages, B cells, NK cells, endothelial, epithelial, and fibroblasts cells 1 . GM-CSF has been identified as a major cytokine in chronic inflammatory autoimmune diseases such as multiple sclerosis (MS) and rheumatoid arthritis (RA) 2,3 GM-CSF plays a crucial role in RA progression and augments inflammatory immune responses in synovia 4,5 . Moreover, GM-CSF-producing CD4 + T cells in the blood and lesions of untreated MS patients correlate with disease severity 6 . We have shown that GM-CSF is necessary for the pathogenicity of Th17 cells in experimental autoimmune encephalomyelitis, the prototypical animal model for MS 7 . GM-CSF exerts its function by binding to its receptor, which is composed of two different subunit α (CD116; GM-CSF Rα) and β chains (CD131; GM-CSF Rβ) with low and high affinity, respectively. The alpha subunit is involved in ligand-specific binding while the beta chain plays a central role in the signal transduction pathway 8 . GM-CSF signaling affects the survival and activation of myeloid cells, dendritic cell (DC) differentiation and M1 macrophage phenotype polarization; it boosts antigen presentation, induces phagocytosis, recruits monocytes and other myeloid populations from bone marrow to circulation and promotes chemotaxis 9, 10 . It has been recently demonstrated that CCR2 + Ly6C hi inflammatory monocytes are a target of GM-CSF in CNS autoimmunity by stimulating inflammatory monocytes and their conversion into pathogenic macrophage-derived dendritic cells [11][12][13] . GM-CSF-activated monocytes migrate across the blood-brain barrier (BBB) and mediate BBB rupture by increasing expression of the endothelial adhesion molecules ICAM-1 and VCAM-1 14,15 . GM-CSF also induces CCR2 expression in monocytes, which gives them an increased ability to cross the BBB. In EAE and MS, the CCR2-CCL2 axis has been previously shown to be a significant driver of inflammatory leukocyte infiltration into the CNS, and its activity positively correlates with disease pathogenesis [16][17][18] . Migration of leukocytes into the CNS is also mediated by CXCL9 and CXCL10 produced by glial cells 19 . Activated T lymphocytes in MS patients express CXCR3, which is the corresponding receptor of CXCL9, CXCL10, and CXCL-11 chemokines 20 . It has been previously shown that while CXCL9 is a homing chemokine in the CNS, CXCL10, and CXCL-11 are induced after inflammation, and their role in inflammation is less clear 21-23 . CXCL10 is involved in intrathecal inflammation 24 . Interestingly, CXCL-11 is upregulated in MS patients after IFN-β therapy and the decrease in the number of relapses may be linked to the increase in CXCR3 ligands in the serum of IFN-β-treated MS patients 25 . In this study, we analyzed the effect of GM-CSF on the phenotype and function of human monocytes. We found that GM-CSF treatment induces an inflammatory phenotype in monocytes, while endogenous GM-CSF blocking is accompanied by an immunomodulatory phenotype. Further, GM-CSF blockade promoted CXCL-11 expression, and recombinant CXCL-11 inhibited the GM-CSF-induced proinflammatory impact of monocytes on T cells. Our findings show that one of the mechanisms by which GM-CSF induces inflammatory monocytes is the inhibition of CXCL-11 production and that this chemokine may be harnessed to suppress deleterious inflammatory responses observed in chronic inflammatory diseases such as MS. Methods Isolation of human monocytes and culture treatments. All subjects gave informed consent before their participation in the current study. All human studies were approved by the Institutional Review Board (IRB) of Thomas Jefferson University, and all methods were performed in accordance with the relevant guidelines and regulations. Whole blood samples were collected from healthy donors and peripheral blood mononuclear cells (PBMCs) were enriched by gradient centrifugation in Ficoll. CD14 + monocytes were isolated by positive selection using magnetic beads following the manufacturer's instructions (Miltenyi Biotec, Bergisch Gladbach, Germany). The purity of cells was above 90%, measured by flow cytometry. Monocytes were seeded (1 × 10 6 /ml) in 24-well plates and cultured in Iscove's Modified Dulbecco's Medium (IMDM) (Gibco, Gaithersburg, MD, USA) supplemented with 10% FBS, 1% penicillin/streptomycin antibiotic (Gibco), 2 mM glutamine and 2β-Mercaptoethanol (50 µg/ml, Gibco). Monocytes were activated with lipopolysaccharide (100 ng/mL, Sigma-Aldrich) for 18 h at 37 °C in the presence of recombinant human GM-CSF (10 ng/mL, R&D Systems, Minneapolis, MN, USA) or anti-GM-CSF (10 µg/mL, Biolegend, San Diego, CA). LPS-activated cells (mature monocytes) cultured with PBS were used as controls and culture of monocytes without LPS stimulation were considered as immature cells. RNA extraction, cDNA synthesis, and qPCR array. RNA was extracted using the RNeasy Mini Kit (Qiagen, Hilden, Germany), and the RNA concentration and quality were determined with Nanodrop (Thermofisher Scientific, Waltham, MA, USA). cDNA synthesis was performed from 1 µg of RNA using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. Real-time PCR was performed using the TaqMan ™ Array Human Immune Response (Applied Biosystems). Real-time PCR for CXCL-11 (Hs03003631_g1) was conducted according to the manufacturer's instructions using TaqMan reagents (ThermoFisher). Relative expression was calculated following the 2 -ΔΔCT method, where 18 s (Hs03003631_g1) was considered the housekeeping gene. Samples were acquired on a BD FACS Aria Fusion (BD Biosciences) flow cytometry instrument and data analyzed using Flowjo software 10. The instrument calibration was examined before running the samples with BD FACSDiva ™ CS&T research beads (CS&T research beads, BD Biosciences). Co-culture experiments. To examine the effects of GM-CSF, CXCL11 and their blockade on T cell responses, human monocytes and naïve T cells were isolated from peripheral blood of healthy volunteers by magnetic cell isolation according to the company's instructions (Miltenyi Biotec). Purified monocytes at 2 × 10 4 were seeded to U bottom 96 well plates and cultured in IMDM culture medium, which contained 10% FBS and 1% penicillin-streptomycin, in 7 groups with different treatment conditions including PBS, GM-CSF (10 ng/ml), Anti-GM-CSF (10 µg/ml), RHCXCL11(15 ng/ml) (R&D Systems), RHCXCL11 + GM-CSF, Anti CXCL11 (10 µg/ ml) (R&D Systems, USA), and Anti-GM-CSF + Anti-CXCL11 for 24 hours. The following day, isolated naïve T cells were labeled with Cell Trace Violet (Thermo Fisher) following the manufacturer's instructions. T cells were then stimulated with the soluble anti-CD3 antibody at a concentration of 1 µg/ml. T cells were then added (1 × 10 5 cells per well) to the same monocyte wells for cell-cell interactions and kept humidified in a 5% CO2 incubator at 37 °C for 72 hours. After that time, cell culture supernatant was carefully removed and immediately frozen at −20 °C. To measure the proliferative capacity of T cells, CFSE intensity in the cells was assessed by flow cytometry. Statistical analysis. Statistical analyses were done using Graph Pad Prism version 7. Comparison between groups was performed using Student's t-test (two groups) and one-way ANOVA tests (three or more groups). Data are shown as mean ± SEM. P values lower than 0.05 were considered statistically significant. GM-CSF induces an inflammatory profile in monocytes. To examine the effect of GM-CSF on monocyte maturation, we stimulated freshly isolated monocytes with LPS in the presence or absence of GM-CSF for 18 h and analyzed the expression of molecules associated with monocyte maturation and antigen presentation. We found that GM-CSF induced a significant increase in the expression of MHC-II and CD86 compared to the PBS-treated groups. Inversely, the expression level of PDL-1 was decreased after GM-CSF treatment. However, no significant differences were found in CD80, CD83 expression between treatment and control groups (Fig. 1A,B) Given that HLA-DR and CD86 are necessary first and second signals in antigen-presentation 26 , we investigated whether cytokines, which are the third signal, would be affected by GM-CSF treatment. We found that GM-CSF significantly decreased the level of anti-inflammatory cytokines like IL-27 and Il-10 ( Fig. 2A,B). On the other hand we observed the increased levels of TNF-α and IL-1β in GM-CSF treated compared with PBS-treated cultures ( Fig. 2A-C). The level of IL-1β and TNF-α was found to be significantly increased in culture supernatants as detected by ELISA (Fig. 2C). Of note, IL-10 and IL-27 was significantly reduced after GM-CSF treatment (Fig. 2C). Blockade of endogenous GM-CSF induces an immunomodulatory phenotype in monocytes. Our results showed that GM-CSF induced a pro-inflammatory profile in monocytes, an effect that has been previously described in the mouse system 27 . Given that monocytes also produce small amounts of GM-CSF 28 , we wanted to test if blockage of endogenous GM-CSF would affect the maturation of monocytes after exposure to LPS. To block endogenous GM-CSF, LPS-activated monocytes were cultured in the presence of anti-GM-CSF monoclonal antibodies. Our results showed that anti-GM-CSF treatment of monocytes significantly reduced expression of HLA-DR and CD86 compared to the control group (Fig. 3A,B). Interestingly, PD-L1 expression remained unaltered (Fig. 3A,B). PD-L1 induced modulation or anergy of PD1 + cells 29 . The fact that GM-CSF supplementation or blockage had little or no effect on PD-L1 expression suggests that GM-CSF plays no role in the PD-L1-PD1 signaling axis. We also found that anti-GM-CSF treatment decreased the expression of TNF-α, IL-1β compared with controls (Fig. 4A,B). Conversely, anti-GM-CSF-treated monocytes significantly produced more IL-10 than controls while, there was no significant difference in IL-27 level between groups (Fig. 4A,B). Collectively, our data show that GM-CSF induced pro-inflammatory phenotype in monocytes and its blockage induced anti-inflammatory cells. CXCL-11 is suppressed by GM-CSF. Our results confirmed that GM-CSF induces the expression of antigen-presenting molecules and stimulates the production of inflammatory cytokines by monocytes. We also showed that blockage of endogenous GM-CSF had the opposite effect by inducing anti-inflammatory monocytes. We then sought to investigate which genes are negatively regulated by GM-CSF. We extracted RNA from monocytes treated with GM-CSF. Monocytes treated with PBS were used as controls. The RNA was reverse transcribed to cDNA, and the cDNA was run in a PCR Array plate that analyzed 96 genes. Our results showed that GM-CSF altered the total gene expression levels in human monocytes (Fig. 5A). We found that CXCL-11 was among the www.nature.com/scientificreports www.nature.com/scientificreports/ genes least expressed in GM-CSF-treated monocytes (Fig. 5A). Conversely, we confirmed that anti-GM-CSF treatment upregulated CXCL-11 at the gene and protein levels (Fig. 5B,C, respectively). CXCL-11 is involved in the generation of regulatory T cells and in the suppression of inflammation 30 . To investigate the contribution of CXCL-11 as a modulatory chemokine of anti-GM-CSF-treated monocytes, we performed a co-culture experiment between syngeneic monocytes and naive T cells. Monocytes were treated with GM-CSF and LPS or with LPS only. To assess their ability to stimulate T cell proliferation, GM-CSF-treated monocytes were cultured with Cell Trace Violet (CTV)-labeled naïve CD4 + T cells in the presence or absence of www.nature.com/scientificreports www.nature.com/scientificreports/ recombinant human CXCL-11. After the incubation period, cells were collected, and the dye decay was measured by flow cytometry. Our data showed that while GM-CSF-treated monocytes stimulated a high proliferative response from T cells, the addition of rhCXCL-11 significantly suppressed this response (Fig. 6A). We also analyzed the phenotype of T cells that were cultured with monocytes and found that IFN-γ production was significantly suppressed in cultures conducted in the presence of rhCXCL-11 (Fig. 6B). Interestingly, the expression of FOXP3 in T cells, which is related to suppressive activity of Treg cells was increased in existence of the CXCL11 (Fig. 6C). Moreover, the Th17-related transcriptional factor RORγt was considerably decreased, in T cells cultured in the presence of rhCXCL-11 (Fig. 6D). These results demonstrate that CXCL-11 is an inhibitor of T cell activation and can overcome the inflammatory effect that GM-CSF exerts on monocytes. CXCL-11 is a significant regulatory chemokine produced by anti-GM-CSF-treated monocytes. We found that GM-CSF directly suppresses the production of CXCL-11 while anti-GM-CSF treatment had the opposite effect (Fig. 5). We then showed that addition of rhCXCL-11 to co-cultures of GM-CSF-treated monocytes overcame the inflammatory phenotype of monocytes and modulated T cells towards an immunoregulatory phenotype (Fig. 6). We then investigated whether blockage of CXCL-11 would hamper the beneficial effect of anti-GM-CSF treatment on monocytes. We performed a co-culture experiment with anti-GM-CSF-treated monocytes and CTV-labeled T cells obtained from the same donors. Culture conditions included cells cultured in the presence or absence of blocking antibodies to CXCL-11. Our results showed that while anti-GM-CSF-treated monocytes had little impact on the proliferation of T cells, blockage of CXCL-11 rescued their proliferation (Fig. 7A). Also, we figured out that blockage of CXCL-11 would result in increase in the production of www.nature.com/scientificreports www.nature.com/scientificreports/ www.nature.com/scientificreports www.nature.com/scientificreports/ IFN-γ (Fig. 7B). Moreover, we found that CD4 + T cells expressed higher levels of FOXP3 when cultured with anti-GM-CSF-treated monocytes in comparison with controls, and this effect was reversed by anti-CXCL-11 (Fig. 7C). In addition, the expression of RORγt which is related to Th17 cells was increased significantly after neutralization of CXCL-11 (Fig. 7D). Taken together, these data confirm that CXCL-11 is a major modulatory chemokine produced by anti-GM-CSF-treated monocytes. Discussion The significant inflammatory role of GM-CSF in autoimmune disorders such as multiple sclerosis (MS) and rheumatoid arthritis (RA) has recently been shown 3,31 . Even though the mechanistic basis of the GM-CSF inflammatory role has not been fully elucidated, its biological and clinical implications are clear [32][33][34][35][36] . In this study, we show that CXCL-11 is a monocyte-derived modulatory chemokine that is suppressed explicitly by GM-CSF. We showed that GM-CSF-treated monocytes increased the expression of critical antigen-presenting molecules such as HLA-DR and CD86 as compared with the control group. Consistent with our findings, previous studies have shown that GM-CSF primes macrophages for the production of pro-inflammatory cytokines such as TNF-α and IL-6 in response to LPS 37,38 . In addition, Croxford et al. have indicated that GM-CSF signaling induces an inflammatory signature in CCR2 + Ly6C hi monocytes and their progeny, which plays a critical role in tissue destruction 11 . In agreement with the known mechanism of action of GM-CSF, in our study, the increase in HLA-DR and co-stimulatory molecule expression in the presence of GM-CSF and its decrease after GM-CSF blockage suggest that the inflammatory impact of GM-CSF is mediated in part through class II molecules along with co-stimulatory markers 39 . Next, we investigated and provided evidence of the role of GM-CSF blocking on T cell responses. We hypothesized that CXCL-11 would be increased after GM-CSF blockade. We next asked whether the CXCL-11 level might affect immunomodulatory pathways of T cell responses, and we found that CXCL-11 suppressed T cell proliferation induced by GM-CSF-treated monocytes. Moreover, CXCL-11 inhibited GM-CSF-induced pro-inflammatory effects of monocytes on CD4 + T cells. On the other hand, we demonstrated that CXCL-11 is involved in suppression of T cell proliferation caused by GM-CSF and also that this chemokine is required for the immunomodulatory effects of monocytes on CD4 + T cells. CXCL-11 is a chemokine that regulates cell trafficking via communication with a specific 7-transmembrane G protein-coupled receptor (GPCR) 40 . This chemokine plays a vital role in the induction of chemotaxis, tissue extravasation, and leukocyte differentiation 41 . Our findings of increased expression of PD-L1, CD39 (data not shown), and IL-10, with a decreased level of T cell proliferation and RORγt expression in treatments with rhCXCL-11 and anti-GM-CSF, show a critical relationship between GM-CSF and CXCL-11 in inflammation regulation. CXCL9 (MIG), CXCL10 (IP-10), and CXCL-11 (I-TAC) can bind to a chemokine receptor called CXCR3, which is expressed on effector T cells, Th17 and also NK cells [42][43][44][45][46] . The binding epitope site of CXCL-9 and CXCL-10 on CXCR3 is different from that of CXCL-11 45,46 . These three chemokines are largely secreted by monocytes, endothelial cells, fibroblasts, and cancer cells 47 . Notably, CXCL-11 has a much greater affinity for binding to CXCR3 than CXCL-9 and CXCL-10, resulting in desensitization of the receptor 45,46 , which makes it a possible antagonist of two other ligands. In a mechanistic study, Paterka et al. assessed the role of CD11c + cells in neuroinflammation. They indicated that IL-17 production by Th17 cells is induced by GM-CSF, and more interaction between Th17 cells and dendritic cells locally reactivates Th17 cells. This loop stimulates Th17 cells for the production of higher levels of GM-CSF, which is required for CD11c + cell induction 48 . Furthermore, the role of GM-CSF in the induction of experimental autoimmune encephalomyelitis (EAE) is not related to IL-17 and IFN-γ 7,48 . CXCL-10 potentially drives Th1 cell differentiation, whereas it has been suggested that CXCL-11 induces FOXP3-negative regulatory T cells that suppress autoimmune encephalomyelitis 30,49 . The robust production of CXCL10 and CXCL9 by CD11c + cells in EAE mice and the depletion of T cell retention or accumulation in the CNS of GM-CSF-deficient (Csf2 −/− ) and CD11c + -depleted mice also demonstrate the inflammatory effects of GM-CSF 48,50 . In addition, suppression of CXCL10 and CXCL-11-induced chemotaxis does not affect IL-10 and IFN-γ production by CXCL-11 and CXCL10, respectively 51 . Accordingly, it thus appears that CXCL-11 binding to CXCR3 regulates inflammatory immune responses that occur in the absence of GM-CSF 49 . In summary, our findings indicate that GM-CSF blockade not only inhibited the pro-inflammatory profile of monocytes, but they also suggest an immunomodulatory role. Our study also strengthens the possibility that GM-CSF may contribute to inflammatory responses through suppression of CXCL-11 production. Induction of CXCL-11 after GM-CSF neutralization and enhanced production of immunoregulatory markers such as IL-10 and PD-L1 after CXCL-11 treatment indicate that CXCL-11 promotes anti-inflammatory responses and that GM-CSF likely interferes with that function. More studies are warranted to test if CXCL11 treatment can suppress chronic inflammatory autoimmune diseases, such as MS and RA.	2020-02-02T14:04:03.348Z	2020-01-31T00:00:00.000	{ "year": 2020, "sha1": "07cb308f976346227eb6f98e8a77395640e7530a", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/s41598-020-58131-2.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "9e0c04690e441236294ddda461685732b0da8204", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Chemistry", "Medicine" ] }
163167571	pes2o/s2orc	v3-fos-license	Association study between matrix metalloproteinase‐3 gene (MMP3) polymorphisms and ankylosing spondylitis susceptibility Abstract Background Ankylosing spondylitis (AS) is the second most common cause of inflammatory arthritis worldwide affecting the axial skeleton. Single nucleotide polymorphisms (SNPs) of matrix metalloproteinase‐3 (MMP3) in the development of AS has few been investigated in Chinese population. Methods A total of 362 patients with AS and 362 healthy controls were enrolled in the study. Five SNPs in MMP3 genotypes were identified by Agena MassARRAY. Chi‐squared tests and genetic model were used to evaluate associations. Results rs522616 had a significant risk of AS development compared to those with the TT genotype (p = 0.008). By multiple logistic regression models analysis, in codominant model, rs522616 CT genotypes also had a 1.44‐fold risk (95% CI = 1.06–1.96, p = 0.008) for AS development compared to those with TT genotypes. In recessive model, the CC genotypes was a significantly reduced AS risk for individuals with TT/CT genotype (OR = 0.64; 95% CI = 0.41–0.99, p = 0.040). Conclusion The present study suggests that MMP3 rs522616 polymorphism is associated with AS susceptibility and MMP3 might be a potential diagnostic biomarker for AS. Further independent studies with larger cohorts are warranted to validate our findings in different populations. \| Subjects Our study recruited 362 patients with AS from the HongHui Affiliated Hospital of Xi'an Jiaotong University College of Medicine Medical University Hospital and The Second Affiliated Hospital of Inner Mongolia Medical University. The study was approved by the Ethics Committee of The Second Affiliated Hospital of Inner Mongolia Medical University. All patients were diagnosed using the modified New York criteria. Patients were informed consent to participate. As a control group, healthy controls were matched 1:1 with AS patients by age and sex. A total of 362 potential controls were randomly selected from subjects with regular health examinations in the center, and they had no rheumatic Demographic characteristics and clinical features of patients with AS and healthy controls. \| Genotyping The gene associated with AS were selected using UCSC (http://genome.ucsc.edu/) database. We found that MMP3 gene was associated with several diseases including AS. We then searched the SNPs in dbSNP database and 1,000 Genomes database (http://www.inter natio nalge nome.org/) to obtain the genetic data of them. We selected five SNPs of the MMP3 gene based on the minor allele frequencies of all the selected SNPs were >5% in the 1,000 Genomes Project (http://www.inter natio nalge nome.org/) Chinese population. All of the selected SNPs in the study were successfully genotyped with an average call rate of 99.38%. Blood samples were collected in tubes containing ethylene diaminetetraacetic acid (EDTA). DNA was extracted from whole blood using GoldMag-Mini Whole Blood Genomic DNA Purification Kit (GoldMag Co. Ltd. Xi'an City, China). We use NanoDrop 2000 (Thermo Scientific, Waltham, Massachusetts, USA) to measure DNA concentration. The design of SNP genotyping and data processing were performed by Agena MassARRAY platform Software (Agena Co. Ltd., San Diego, California, USA). Genotype calling was carried out with 3.0 version MassARRAY RT software and analyzed by 3.4 version MassARRAY Typer software (Gabriel, Ziaugra, & Tabbaa, 2009). Agena Typer 4.0 software was used for data management and analysis. We listed the primer in \| Statistical analysis Microsoft Excel (Microsoft, Redmond, WA) and SPSS Statistics (version 20.0, SPSS, Chicago, IL) were used for statistical analyses. SNP genotype frequencies in the case and control groups were calculated by Chi-square Test, and the Hardy-Weinberg equilibrium (HWE) was used to check the genotype frequency of the control group. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were tested using unconditional logistic regression analysis with adjustment for age and gender (Bland & Altman, 2000). Haploview version 4.2 was used to identify the linkage disequilibrium (LD) block and haplotypes (Barrett, Fry, Maller, & Daly, 2005). The significance level for all statistical analyses was 0.05. \| RESULTS Tables 2 give the candidate SNP genotypes and allele frequency. These data conformed to the Hardy-Weinberg equilibrium (HWE) in that the allele frequencies (p > 0.05). By study design, there were no SNPs statistically significant. In Table 3 we displayed the genotype and AS risk. We found rs522616 had significant risk of AS development compared to those with the TT genotype (p = 0.008). To determine whether SNPs of MMP3 were associated with susceptibility to AS, multiple logistic regression analysis while adjusting for age and gender was conducted. Multiple logistic regression models (codominant models, dominant models, recessive models and additive model). As shown in metalloproteinase-3 (MMP3, also known as human fibroblast stromelysin) is a secreted metalloprotease produced predominantly by connective tissue cells (Lièvre et al., 2006). Together with other MMPs, it can synergistically degrade the major components of the extracellular matrix (Johansson et al., 2000) and is also capable of degrading proteoglycan, fibronectin, laminin and type IVcollagen (Jin et al., 2005). The exact biological mechanisms are unknown, but tissue degradation of biochemical mediators, especially MMPs, has been identified as an important factor (Zade, Gosavi, Hazarey, & Ganvir, 2017). Recent studies have shown that it plays an important role in AS. Serum MMP3 levels were significantly higher in patients than in healthy subjects, and to a greater extent in patients with high disease activity (Chen et al., 2006). In addition to digesting components of ECM, MMP3 activates a number of pro-MMPs and is critical in the full generation of active MMPs (Nagase, 1997;Visse & Nagase, 2003). It plays a key role in cartilage damage and joint destruction. Serum MMP3, originating directly from inflamed joints, can be specific markers of inflammation in joint activity (Vandooren, Kruithof, & Yu, 2004). At present, there are few studies on MMP3 polymorphism. We found rs522616(MMP3) was associated with AS risk, as far as we know, no other studies have been reported the SNP associated with AS risk. In addition, we demonstrated that the CC genotype of rs522616, which is located in the promoter region of MMP3, was associated with a lower risk of developing AS. It is possible that a variant in the promoter region of MMP3 could affect the production of proteolytic enzymes. Meanwhile, it may have an effect on the risk of AS occurrence. It may be the reason that the transcription factor can bind to rs522616 C allele of the MMP3 promoter, activate its transcription, and lead to a higher expression of this gene. Although AS is thought to be caused by a complex interaction of environmental and genetic factors, the polymorphisms identified in this study might be useful for predicting the susceptibility to the disease. In conclusion, the present study suggests that MMP3 rs522616 polymorphism is associated with AS susceptibility and MMP3 might be a potential diagnostic biomarker for AS. Further independent studies with larger cohorts are warranted to validate our findings in different populations.	2019-05-25T13:03:02.591Z	2019-05-23T00:00:00.000	{ "year": 2019, "sha1": "928c4350dae229390e5e6f8d0d9b4c535a990349", "oa_license": "CCBYNCND", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/mgg3.752", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "928c4350dae229390e5e6f8d0d9b4c535a990349", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
269859963	pes2o/s2orc	v3-fos-license	Effect of Prunus cerasus (Sour Cherry) on Nephrolithiasis in Children: The First Noninferiority Two-Arm Randomized Clinical Trial Objective: The present study evaluated the effect of Prunus cerasus (sour cherry) on children with nephrolithiasis. Methods: We conducted a randomized noninferiority controlled trial to evaluate the therapeutic efficacy of P. cerasus among children with nephrolithiasis. Subjects in the intervention group received 1.25 mL/kg of cherry concentrate once daily for 2 months, while the control group received 1 mL/kg Polycitra-K, which consists of 220 g citrate potassium and 68 g citric acid in 1000 mL sterile water. The major outcome was sonographically determined number and sizes of kidney stones, which were assessed before and after the trial. Results: Sixty-eight children completed the study. At trial onset, both groups were similar in baseline characteristics (P > .05). In within-group analysis, the number of stones significantly decreased in both groups (P < .05). After 2 months, the number of nephrolithiasis was 1.55 ± 0.49 and 1.47 ± 0.67 in the control and intervention groups, respectively (P value = .56). The percentage of change in calculi number was 44.11 ± 11.12 and 38.14 ± 14.08 in the control and intervention groups, respectively (P value = .08). At the end of the study, the urine pH was 6.46 ± 0.99 and 6.14 ± 0.83 in the control and intervention groups, respectively (P value = .19). Urine calcium and uric acid concentrations were 32.00 ± 12.32 and 28.95 ± 10.96 mg/mm (P value = .68) and 24.11 ± 10.58 and 30.03 ± 11.39 mg/mm (P value = .012) in control and intervention groups, respectively. Conclusion: Our clinical data supported the efficacy of sour cherry in the treatment of nephrolithiasis compared to Polycitra-K. Future randomized controlled trials are needed to confirm the present observation. Introduction Nephrolithiasis is a common health problem with multifactorial etiologies.The exact prevalence of urolithiasis in children. is unknown.It has been estimated that children account for 2%-3% of stone formers worldwide. 1According to epidemiological studies, the prevalence of kidney stones is about 36 to 145 per 100 000 children. 1 Several factors such as genetics, nutrition, and hydration status of hydration affect stone formation. 2lycitra-K is commonly used to treat several types of nephrolithiasis because it forms a soluble calcium-citrate complex, thereby preventing the precipitation of calcium from urine as either calcium phosphate or calcium oxalate.Polycitra-K prevents the formation of uric acid and cystine stones by increasing urine pH. 3 Alkali citrates are effective in the treatment of calcium oxalate, calcium phosphate, cystine, and uric acid stones.Citrates reduce urinary calcium via binding to calcium.This leads to the prevention of the crystallization of calcium oxalate and calcium phosphate stones. 4An old randomized clinical trial confirmed the efficacy of potassium citrate in calcium nephrolithiasis compared to placebo. 5 Adherence of children to the prescribed Polycitra-K is generally challenging due to its bitter taste.In one Iranian study, only 62% of children treated with Polycitra-K showed adequate adherence. 6e popularity of natural products for the treatment of various illnesses appears to be rising.People believe that natural products are safer and have fewer side effects than synthetic drugs.Thus, there is a demand to study the effects of natural products with scientific rigor.Sour cherries (Prunus cerasus) are native to Europe and some parts of Asia.They are also grown in most parts of Iran.They are rich in vitamins A, C, and E, anthocyanins, and isoflavonoids. 7[9][10] Their consumption has been found to be beneficial for muscle soreness after exercise, gout, arterial and cardiovascular diseases. 11,12Nevertheless, scientific evidence to support the clinical use of sour cherries is scarce.Animal experiments in rats suggest a therapeutic effect of Cerasus avium on calcium oxalate stones.For example, Azaryan et al. found that it reduced the number and sizes of calcium oxalate stones in rats. 13ur cherries have a considerable content of citrate compared to other fruits. 14This led us to hypothesize that sour cherry may be effective in preventing or treating patients with nephrolithiasis.If so, the use of sour cherry extract could increase adherence due to its better palatability compared with Polycitra-K.However, we were unable to find a study demonstrating the efficacy of sour cherry compared with Polycitra-K in patients with nephrolithiasis. We therefore conducted the present randomized two-arm singleblind clinical study to evaluate the therapeutic efficacy of an oral sour cherry concentrate in children with nephrolithiasis in the form of a noninferiority trial with Polycitra-K treatment as the reference drug. Material and Methods The present study was a single-blind randomized clinical trial.It was conducted between 2020 and 2021 in a tertiary pediatric hospital, affiliated with the affiliated with the Mashhad University of Medical Sciences (MUMS), Mashhad, Iran.The study population comprised children with incident urolithiasis who had not been previously treated.The study was approved by the Medical Ethics Committee of the Mashhad University of Medical Sciences (Approval Number: IR.MUMS.fm.REC.1396.376;Date: September 27, 2017) and registered in the Iranian registry of clinical trials (IRCT) (Code: IRCT20170415033428N5, Date: June 29, 2020).Informed consent was obtained from all subjects and/or their legal guardian(s).Seventy children aged 1-18 years were included.All clinical examinations were performed by a single pediatric nephrologist.Patients with urolithiasis and metabolic disorders including hyperuricosuria, hypercalciuria, hyperoxaluria, cystinuria, and hypocitraturia were included.Exclusion criteria were the presence of obstructive or infected calculi and lack of consent.A Consolidated Standards of Reporting Trials (CONSORT) flowchart is shown in Figure 1.The control treatment was Polycitra-K (Sepidaj Pharmaceutical Company).Controls recieved 1 mL/kg Polycitra-K, which consists of 220 g citrate potassium and 68 g citric acid in 1000 mL sterile water.One milliliter of Polycitra-K corresponds to 1 mmoL (2 mmEq) alkali.The intervention was sour cherry concentrate, which was purchased from Khoosheh Sorkhe Shargh Agro Industrial® (Tehran, Iran).The citrate content in the cherry concentrate was 8%, as determined by high-performance liquid chromatography (HPLC) (Testa Quality Control Laboratories, Mashhad, Iran).Cherry concentrate dosing in the intervention group was adjusted to match the amount of citrate consumed by control patients.Polycitra-K had a citrate content of 10%, while the cherry concentrate had 8% citrate.Therefore, the citrate content of 1 mL Polycitra-K was equivalent to 1.25 mL of cherry concentrate. Accordingly, the intervention group received 1.25 mL/kg of concentrate once daily for 2 months.All patients were instructed to follow a low-sodium diet and maintain high water intake.The primary outcome was the size and number of calculi determined by sonography (MyLab™X8 Doppler sonography machine, Esaote, Genova, Italy).An expert sonographer performed all ultrasound scans and was unaware of the patient groupings.Urine analysis and sonography of the kidneys and urinary tract were performed at the time of enrollment and 2 months after intervention for all patients.The following demographic and clinical data were recorded in a data collection sheet: sex, age at diagnosis, urine analysis (UA), and urine culture (UC) results, presence or absence hematuria (gross or microscopic), chief complaint (restless, dysuria, flank pain, vomiting, frequency, urinary tract infection MAIN POINTS • Prunus cerasus (sour cherry) is a common and highly used fruit worldwide which has a high content of citrate. • Use of sour cherry for 2 months decreased the number and size of kidney stones in children. • Sour cherry leads to a decrease in the urine uric acid during the intervention time. • Sour cherry leads to an increase in the urine pH during the intervention time. • Sour cherry is not inferior to poly-citrate potassium in the treat- ment of urolithiasis in children.(UTI), poor growth and change in voiding habits), abnormal clinical exam findings (flank mass, suprapubic pain, flank pain), family history of urolithiasis, blood test results (serum creatinine, uric acid, calcium, phosphorus, parathyroid hormone), and sonography findings.Urine pH was determined using dipstick chemistry.The urine sample was collected in the morning before breakfast.In toilet-training children, midstream urine was collected, while in non-toilet-training children, the sample was taken via a urine bag.All urine samples were sent to the laboratory within the first hour of sampling. Randomization was based on a commercial computer-generated random sequence.Allocation concealment was done via sealed envelopes.A person who was not on the research team did the patient's assignment.An expert who was not involved in the study evaluated primary outcomes.A 15% lesser response to treatment compared to the control group was considered as noninferior treatment.A sample size of 35 patients per group was considered appropriate for the initial study (a total of 70 patients). Statistical Analysis Data were analyzed using SPSS version 16.0 (SPSS Inc.; Chicago, IL, USA).Quantitative data were presented as mean ± standard deviation (SD), and categorical data were presented as frequency and percentage.The association of qualitative variables was tested using Fisher's exact test or chi-square.Quantitative variables were compared using an independent-sample t-test.Within-group changes were assessed using paired t-tests.A P-value less than .05was considered statistically significant. Baseline Characteristics Among the 70 enrolled patients, 63 patients completed the study, with 34 in the control group and 29 in the intervention group.Seventeen patients (50%) and 14 patients (48.3%) were male in the control and intervention groups, respectively (P = .89).The mean age of patients in the control and intervention groups was 5.7 ± 2.31 and 5.29 ± 3.6 years respectively (P = .58).All patients were similar in baseline clinical and laboratory data (P > .05).Additionally, all patients had normal renal function at the time of enrollment (P > .05).Data are presented in Table 1 . Number of Stones The number of urolithiasis at the baseline was 2.79 ± 1.06 and 2.39 ± 1.05 in the control and intervention groups, respectively (P value = .22).After 2 months, the number of nephrolithiasis was 1.55 ± 0.49 and 1.47 ± 0.67 in the control and intervention groups, respectively (P value = .56).In within-group analysis, the number of stones significantly decreased in both groups (P < .05),but the percent of change in calculi number was 44.11 ± 11.12 and 38.14 ± 14.08 in the control and intervention groups, respectively (P value = .08,95% CI: 1.75, 12.27). Urine pH The urine pH before the trial was 5.56 ± 0.27 and 5.44 ± 0.31 in the control and intervention groups, respectively (P value = .12).After 1 month, the urine pH was 6.14 ± 0.47 and 6.7 ± 0.3 in the control and intervention groups, respectively (P value = .57).After 2 months, the urine pH was 6.46 ± 0.99 and 6.14 ± 0.83 in the control and intervention groups, respectively (P value = .19).The data are presented in Table 2.The data are illustrated in Figure 2. Urine Calcium In a random sample, the urine calcium before the trial was 34.54 ± 29.32 and 29.39 ± 19.53 mg/mL) in the control and intervention groups, respectively (P value = .42).After 1 month, the urine calcium was 32.22 ± 23.06 and 28.75 ± 10.39 mg/mL in the control and intervention groups, respectively (P value = .45).After 2 months, the urine calcium was 32.00 ± 15.32 and 28.95 ± 18.96 mg/mL in the control and intervention groups, respectively (P value = .68).The data are presented in Table 2.The data are illustrated in Figure 3. Urine Uric Acid In a random sample, the urine uric acid before the trial was 37.4 ± 24.77 and 34.07 ± 16.48 mg/mL in the control and intervention groups, respectively (P value = .54).After 1 month, the urine uric acid was 31.11± 24.95 and 33.44 ± 15.46 mg/mL in the control and intervention groups, respectively (P value = .66).After 2 months, the urine uric acid was 24.11 ± 14.58 and 30.03 ± 12.39 mg/mL in the control and intervention groups, respectively (P value=0.012).The data are presented in table 2. The data are illustrated in Figure 4. Side Effects There were no side effects associated with the intervention.However, in the control group, 2 patients (5.88%) had mild to moderate diarrhea. Discussion In a randomized two-arm single-blind clinical study, 68 children with urolithiasis were studied evaluating the effect of sour cherry concentrate on the number of nephrolithiasis compared to Polycitra-K.Results showed that sour cherry concentrate is not inferior to Polycitra-K in the treatment of nephrolithiasis.Despite some evidence regarding the traditional use of P. cerasus in the management of kidney stones, 15 this study is the first clinical trial supporting the beneficial effect of sour cherry concentrate in the treatment of nephrolithiasis.Our results are particularly important in children because the compliance of Polycitra-K in children is rather low.7][18] However, the current study demonstrates its clinical efficacy for the treatment of nephrolithiasis for the first time.Many medicinal plants have been used to treat urolithiasis in animal and clinical experiments 1,2 with variable results. 15,19ccording to Shirani, Arnebia euchroma root extract efficacy dissolved calcium oxalate and calcium phosphate stones in vitro. 20In a meta-analysis of clinical trials, it has been shown that Cystone® (ayurvedic proprietary medicine) decreased the size and number of kidney stones compared to placebo. 19In another systematic review, the authors concluded that there is insufficient data to support the use of herbal products for the treatment of urolithiasis. 21ingh et al conducted a randomized clinical trial comparing the efficacy of another ayurvedic plant extract, Calcury®, with potassium citrate in stone-forming patients and concluded that phytotherapy effectively reduces stone size and improves biochemical parameters, such as hypercalcemia, oxaluria, and urinary citrate and pH. 22e beneficial effect of Prunus mahaleb extract on kidney stone formation was studied in BALB/c mice.In this study, 500 mg of P. mahaleb effectively prevented the formation of kidney stones and improved serum creatinine and urea levels compared to the control group without adverse effects. 23Azaryan reported that 200 and 400 mg/kg aqueous extract of C. avium stem powder have a therapeutic effect on calcium oxalate stones in rats with nephrolithiasis and reduce the number of calcium oxalate deposits. 13Mehmet et al reported that patients with nephrolithiasis consumed less cherry (as a fruit) compared to healthy controls. 2In another clinical study on healthy volunteers, it was reported that the capsules of Prunus avium (cherry) stalk increase the volume of urine and that it could be used as a mild diuretic without identified side effects. 24In our study, P. cerasus was as effective as Polycitra-K in reducing the stone load.Despite this, the measured urine calcium and uric acid excretion did not decrease as in the control group.Polycitra-K is considered the first-line treatment in metabolic nephrolithiasis.It alkalinizes the urine which leads to the dissolution of calcium, uric acid, and other crystals. 3According to our results, P. cerasus also raises the urine pH which is clinically valuable, but less than potassium citrate. Our clinical data support the hypothesis that sour cherry concentrate is not inferior to the standard treatment of nephrolithiasis with Polycitra-K.Sour cherry concentrate led to a 38% decrease in the number of calculi, which was similar to the treatment effect of Polycitra-K.Additional clinical studies are warranted, in children and adults, to confirm the efficacy and the benefit of the described treatment. Limitations Because of the distinctive taste and appearance of sour cherry concentrate, it was not possible to blind the patients.The follow-up time in this study was rather short, which may have affected the results.Longer durations of treatment and follow-up are suggested in future studies. Figure 1 . Figure 1.CONSORT diagram of the study. Figure 2 .Figure 3 . Figure 2. Changes in urine pH in groups over time. Figure 4 . Figure 4. Changes in urine uric acid in groups over time. Table 1 . Demographic, Clinical, and Laboratory Characteristics of Patients with Urolithiasis RBC, red blood cell; UTI, urinary tract infection; WBC, white blood cell.# More than 5 cells in high power field.Chi-square or Fisher's exact test.+ Independent t-test. Table 2 . The urine pH, Calcium, and Uric Acid in Children with Nephrolithiasis Before and After the Study Repeated measure ANOVA.	2024-05-19T15:11:00.621Z	2024-03-01T00:00:00.000	{ "year": 2024, "sha1": "b56ea3cf6709c8d0625efdfd385590621701812e", "oa_license": "CCBY", "oa_url": null, "oa_status": "CLOSED", "pdf_src": "PubMedCentral", "pdf_hash": "804328cbdee568d65b481ab5e63e20f2f39baeca", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
249314077	pes2o/s2orc	v3-fos-license	Quantitative relaxometry using synthetic MRI could be better than T2-FLAIR mismatch sign for differentiation of IDH-mutant gliomas: a pilot study This study aimed to determine whether quantitative relaxometry using synthetic magnetic resonance imaging (SyMRI) could differentiate between two diffuse glioma groups with isocitrate dehydrogenase (IDH)-mutant tumors, achieving an increased sensitivity compared to the qualitative T2-fluid-attenuated inversion recovery (FLAIR) mismatch sign. Between May 2019 and May 2020, thirteen patients with IDH-mutant diffuse gliomas, including seven with astrocytomas and six with oligodendrogliomas, were evaluated. Five neuroradiologists independently evaluated the presence of the qualitative T2-FLAIR mismatch sign. Interrater agreement on the presence of the T2-FLAIR mismatch sign was calculated using the Fleiss kappa coefficient. SyMRI parameters (T1 and T2 relaxation times and proton density) were measured in the gliomas and compared by the Mann–Whitney U test. Receiver operating characteristic curve analysis was used to evaluate the diagnostic performance. The sensitivity, specificity, and kappa coefficient were 57.1%, 100%, and 0.60, respectively, for the qualitative T2-FLAIR mismatch sign. The two types of diffuse gliomas could be differentiated using a cutoff value of 178 ms for the T2 relaxation time parameter with 100% sensitivity, specificity, accuracy, and positive and negative predictive values, with an area under the curve (AUC) of 1.00. Quantitative relaxometry using SyMRI could differentiate astrocytomas from oligodendrogliomas, achieving an increased sensitivity and objectivity compared to the qualitative T2-FLAIR mismatch sign. www.nature.com/scientificreports/ In the differential diagnosis of these diffuse gliomas, the so-called "T2-fluid-attenuated inversion recovery (FLAIR) mismatch sign" was reported as a specific imaging marker for IDH-mutant astrocytomas 5 . This simple visual sign has a specificity of 100%; however, it has low sensitivity (12-51%) 5,8 and is considered subjective since its diagnosis is based on its appearance in imaging evaluations. We hypothesized that quantitative relaxometry would show increased sensitivity and objectivity compared to the qualitative T2-FLAIR mismatch sign. Synthetic MRI (SyMRI) can directly evaluate T1 and T2 relaxation times and proton density (PD) values during the clinical scan time and hence was considered for quantitative relaxometry 9 . To the best of our knowledge, no study has quantitatively evaluated the diagnostic accuracy of relaxometry in patients with IDH-mutant lower-grade gliomas. Hence, the purpose of this study was to determine whether quantitative relaxometry using SyMRI could differentiate these diffuse gliomas, achieving an increased sensitivity when compared to the qualitative T2-FLAIR mismatch sign. Results Patient demographics and characteristics. A total of 13 patients (median age, 43 years; interquartile range, 36.8-51.2 years; six male and seven female patients) were included in this study. There were seven patients with IDH-mutant astrocytomas (range, median age: 29-49, 41 years; three males and four females), and six with IDH-mutant and 1p/19q-codeleted oligodendrogliomas (37-63, median age: 48 years; three males and three females). An overview of the patient demographics and characteristics is presented in Table 1, and the patient selection process is depicted in Fig. 1. There was a significant difference observed in the T2-FLAIR mismatch sign between the two tumor groups (p = 0.02). The patients' demographics and pathological diagnoses, based on the 2016 World Health Organization (WHO) classification 4 , were as follows: seven patients with astrocytomas, including four with diffuse astrocytomas (WHO grade II; age, 33-44 years; two males and two females), one with anaplastic astrocytoma (WHO grade III; age, 29 years; one male), and two with glioblastomas (WHO grade IV; age, 45 and 49 years; two females), and six patients with oligodendrogliomas, including four with oligodendrogliomas (WHO grade II; age, 40-63 years; two males and two females) and two with anaplastic oligodendrogliomas (WHO grade III; age, 37 and 43 years; one male and one female). All the patients were genetically verified as having IDH-mutant type tumors. Patients with postoperative recurrences (three astrocytomas and two oligodendrogliomas) were included, and the recurrences were confirmed in the pathological specimens obtained during the second surgery. Qualitative evaluation. Table 2 shows the qualitative results from the five neuroradiologists. The mean sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were 57.1%, 100%, 76.9%, 100%, and 67.0%, respectively. Moderate interrater agreement was observed (kappa value = 0.60). Table S1 show the histograms of each parameter over all the pixels in the tumor regions of interest (ROIs). T1 and T2 relaxation times and PDs from the astrocytomas all exhibited a slight rightward shift relative to those from the oligodendrogliomas. T1 and T2 relaxation times and PDs were larger for astrocytomas than for oligodendrogliomas (mean ± standard deviation, p value: 2047 ± 454 vs. 1290 ± 274 ms, p < 0.0001 for T1 relaxation time; 254 ± 124 vs. 109 ± 16 ms, p < 0.0001 for T2 relaxation time; 89.4 ± 5.3 vs. 80.0 ± 4.4%, p = 0.01 for PD, respectively). There were also significant differences in the 10-90th percentiles for T1 and T2 relaxation times and PDs (all p < 0.05). Table 3 shows the diagnostic performance in differentiating the two glioma groups; the most useful values of each parameter are shown in Discussion As anticipated, our results confirmed that quantitative relaxometry using SyMRI could differentiate IDH-mutant gliomas, leading to an increased sensitivity compared to the qualitative T2-FLAIR mismatch sign. The qualitative results showed high specificity; however, the sensitivity was low, and the interrater agreement was moderate. Quantitative results showed that there were significant differences in the 10-90th percentiles and the mean between the gliomas for all SyMRI parameters. The receiver operating characteristic (ROC) curve analysis revealed that the T2 relaxation time 10th and 50th percentile and mean values showed the highest diagnostic ability in differentiating gliomas. Previous studies have reported that the T2-FLAIR mismatch sign has been shown to predict IDH-mutant astrocytoma with 95-100% specificity 10,11 ; however, the sensitivity has been low at 22-51% 8,10,12 . The interrater agreement has been shown to have a wide range, κ = 0.38-0.88 5,8,10,11 . Jain et al. 13 mentioned that strict application of the criteria is necessary to maintain high specificity, which tends to result in low sensitivity. Our results are consistent with those of these previous results. The low interrater agreement is probably due to the binary scoring system used, whereas subtle changes may result in different interpretations across readers 8 . Our quantitative results showed long T1 and T2 relaxation times and increased PDs within astrocytomas compared to oligodendrogliomas. To our knowledge, this is the first study to quantitatively evaluate IDH-mutant gliomas using SyMRI. These prolonged T1 and T2 relaxation times imply fluid changes in the IDH-mutant astrocytomas. Deguchi et al. revealed that abundant microcysts were observed upon hematoxylin-eosin staining of specimens from the T2-FLAIR mismatched region in IDH-mutant astrocytomas 11 , which may reflect the T1 and T2 relaxation time prolongations. Kinoshita et al. evaluated gliomas quantitatively using MP2RAGE images calculated via Bayesian inference modeling and showed that the T2-FLAIR mismatch region exhibited extremely long T1 and T2 relaxation times 12 . Their results also support fluid changes within IDH-mutant astrocytomas. Our quantitative results showed that there were significant differences in the mean values for all SyMRI parameters between the two glioma groups. These findings may support the clinical utility of SyMRI to diagnose gliomas because only the placement of a simple ROI in the tumor is required, not a histogram analysis. The 10th and 50th percentile values are not as good as the mean value owing to the simplicity of the measurement for actual clinical use and the fact that they are supplementary indicators. Furthermore, as expected, quantitative evaluation significantly increased the sensitivity compared to the qualitative T2-FLAIR mismatch sign. Advanced MR techniques, such as dynamic perfusion MRI 14 and MR spectroscopy 15 , have also revealed differences between IDH-mutant and wild-type gliomas. However, dynamic perfusion MRI requires contrast media, and 2-hydroxyglutarate MR spectroscopy has been shown to be associated with false-positive cases with intratumoral hemorrhage 16 . In Table 3. Diagnostic performance of parameters in differentiating between IDH-mutant astrocytic tumors and IDH-mutant and 1p/19q-codeleted oligodendroglial tumors. AUC area under the curve, IDH isocitrate dehydrogenase, NPV negative predictive value, PD proton density, PPV positive predictive value. Data in parentheses are numerators/denominators; data in brackets are 95% confidence intervals. 17 . This study has several limitations. First, the sample size was small, and our study included postoperative cases. Although postoperative changes may affect the results of the qualitative evaluation, the quantitative evaluation was still useful to differentiate the gliomas in this study. Our results demonstrate that different subtypes of diffuse gliomas have different relaxation properties. Second, our study did not evaluate IDH-wild-type astrocytomas. It would be desirable to proceed with a subsequent study including patients with IDH-wild-type astrocytomas. Third, we did not include the whole tumor volume in the histogram analysis. Instead, we used the maximum section of the tumor, with its boundary defined by the hyperintensity on T2-weighted imaging (T2WI). However, in previous studies on the T2-FLAIR mismatch sign, whole-volume histogram analysis was not performed, and only the maximum section of the tumor was utilized. Simple evaluation based on the maximum-sized slice of the tumor could be sufficient since the T2-FLAIR mismatch sign criteria are designed such that high specificity is maintained rather than increasing sensitivity by employing these strict criteria 13 . If quantitative volumetry were conducted, a histogram of the entire tumor would be provided, and the optimal cutoff value would presumably be different from the present results; however, it is likely that astrocytomas would have shown longer T1 and T2 values than oligodendrogliomas. In conclusion, relaxometry using SyMRI could differentiate IDH-mutant astrocytomas from IDH-mutant and 1p/19q-codeleted oligodendrogliomas. Quantitative relaxometry could increase sensitivity and objectivity compared to the qualitative T2-FLAIR mismatch sign; therefore, this objective evaluation provides a helpful, noninvasive diagnostic method for differentiating IDH-mutant gliomas. Prospective multicenter validation is needed to confirm our findings. Materials and methods Patient selection and clinical data. All patients consecutively diagnosed with glioma at our institution from May 2019 to May 2020 were eligible for this study. Inclusion criteria were (1) patients with IDH-mutant and/or 1p/19q-codeleted glioma based on the WHO classification 4 ; (2) MRI scans that had been performed www.nature.com/scientificreports/ within two weeks before surgery; and (3) patients for whom SyMRI was acquired. The exclusion criterion was image distortion, such as motion artifacts or noise. Ethics approval. The institutional review board of Kyushu University Hospital approved this retrospective study, and the requirement for informed consent was waived. All methods were performed in accordance with the relevant guidelines and regulations. Histopathological analysis. All tissue samples were analyzed based on the WHO 2016 classification 4 . Immunohistochemistry for IDH1 R132H, ATRX, p53, and Ki67 was routinely performed. IDH1/2 was analyzed using high-resolution melting with DNA extracted from frozen tissue samples. The 1p/19q codeletions were evaluated using a microsatellite-based loss of heterozygosity analysis with 18 markers to detect the loss of the entire chromosome arm 19 . MRI. MRI was performed using a 3 T MR scanner (Ingenia 3.0 T CX; Philips Healthcare, Best, Netherlands) with a 15-channel head coil. Quantitative MRI was performed using the two-dimensional axial quantification of relaxation times and PDs by the multiecho acquisition of a saturation recovery using a turbo spin-echo readout (QRAPMASTER) pulse sequence with two echo times (TEs; 13 and 100 ms) and four delay times to generate eight real images and eight imaginary images 9 www.nature.com/scientificreports/ Qualitative evaluation. Five board-certified neuroradiologists (with 23, 21,19,8, and 6 years of experience) were blinded to the patient information of the evaluated T2-FLAIR mismatch sign 5,8 . The T2-FLAIR mismatch sign was defined by the presence of 2 distinct MRI features as follows 5 : (1) The tumor displayed a complete or nearly complete and nearly homogeneous hyperintense signal on T2WI. (2) The tumor displayed a relatively hypointense signal on the FLAIR sequence except for a hyperintense peripheral rim. These two criteria should be strictly used to maintain high specificity for the diagnosis of IDH-mutant astrocytomas 13 . Jain et al. 13 introduced additional imaging features aiding in the accurate identification of the T2-FLAIR mismatch sign: (3) Necrotic cavities do not represent the T2-FLAIR mismatch sign; small cysts do not meet the criteria for the T2-FLAIR mismatch sign. (4) The T2-FLAIR mismatch sign is typically accompanied by little or no contrast enhancement. (5) The degree of FLAIR signal suppression could be inhomogeneous within the tumor. (6) Common imaging correlates include homogeneous hypointensity on noncontrast T1WI, markedly elevated apparent diffusion coefficient values, low blood volume on perfusion maps, and diffuse hypodensity on CT. After independent data collection, the interreader agreement was calculated, and discordant results for the T2-FLAIR mismatch sign were resolved by consensus 5 . Five radiologists read both the T2WI and FLAIR images based on whether the T2-FLAIR mismatch sign was present or absent. The sensitivity, specificity, PPV, NPV, and accuracy were calculated. Quantitative evaluation. The DICOM data of the T1 and T2 relaxation time and PD maps were extracted by SyMRI software. We used a single maximum section of each tumor for the ROI analysis on the T2-prolonged region in the tumor using an ImageJ plugin (ImageJ/Fiji; version 2.0.0-rc-59/1.51 k, National Institutes of Health, Bethesda, MD). The maximum section of the tumor was visually decided as the largest orthogonal cross product of the tumor on the axial T2/FLAIR scans 20 . Using the ROI manager tool of ImageJ/Fiji, the ROI mask from the T2-prolonged region on conventional T2WI scans was copied and placed on each parameter map (T1 and T2 relaxation times and PD) to obtain pixel-by-pixel values for the histogram analyses. The 10th, 25th, 50th, 75th, and 90th percentiles and the mean, skewness, and kurtosis of each parameter were recorded from the histograms. Statistical analysis. Chi-square tests were used to compare the patients' categorical variables (e.g., sex, WHO grade, tumor location, presence or absence of enhancement, calcification, cyst, hemorrhage, T2-FLAIR mismatch sign). In the qualitative evaluation, the interrater agreement for the T2-FLAIR mismatch sign among the five observers was evaluated using Fleiss's kappa coefficient 21 . The kappa value was interpreted as follows: almost perfect agreement, 1.00-0.81; substantial agreement, 0.80-0.61; moderate agreement, 0.60-0.41; fair agreement, 0.40-0.21; slight agreement, 0.20-0.01; and poor agreement, < 0 22 . In the quantitative evaluation, the percentiles, mean, skewness, and kurtosis of each parameter (i.e., T1 and T2 relaxation times and PD) were compared between astrocytomas and oligodendrogliomas by the Mann-Whitney U test. The diagnostic performance of each parameter was evaluated by ROC curve analysis. The comparison of simulated T2-FLAIR mismatch signs was made by the Mann-Whitney U test. All statistical analyses were performed using commercial software programs (JMP, version 15.0.0; SAS Institute, Cary, NC, USA; Prism 7.0, GraphPad Software, La Jolla, CA, USA). P values < 0.05 were considered statistically significant. Data availability The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.	2022-06-04T06:23:10.038Z	2022-06-02T00:00:00.000	{ "year": 2022, "sha1": "8c3aab4e56ada8bb982ce171570d4ba312ce3e32", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "f721918e1fe0b32b7d7854a6d7cc6e10da15d00d", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
14067638	pes2o/s2orc	v3-fos-license	Linear Algorithms for Radioelectric Spectrum Forecast : This paper presents the development and evaluation of two linear algorithms for forecasting reception power for different channels at an assigned spectrum band of global systems for mobile communications (GSM), in order to analyze the spatial opportunity for reuse of frequencies by secondary users (SUs) in a cognitive radio (CR) network. The algorithms employed correspond to seasonal autoregressive integrated moving average (SARIMA) and generalized autoregressive conditional heteroskedasticity (GARCH), which allow for a forecast of channel occupancy status. Results are evaluated using the following criteria: availability and occupancy time for channels, different types of mean absolute error, and observation time. The contributions of this work include a more integral forecast as the algorithm not only forecasts reception power but also the occupancy and availability time of a channel to determine its precision percentage during the use by primary users (PUs) and SUs within a CR system. Algorithm analyses demonstrate a better performance for SARIMA over GARCH algorithm in most of the evaluated variables. Introduction Radioelectric spectrum occupancy is widely studied due to its importance for the construction of new spectrum assigning policies in emerging technologies, as well as in monitoring activities both in licensed and unlicensed bands.Real measurements for spectrum use within a determined band allow the corresponding authorities to guarantee that licenses meet local and regional spectrum regulations [1].On the other hand, precise parameter estimates like time quantity and geographical region where the different spectrum band is actually used bring useful information to determine spectral opportunities for variant technologies within a domain.In this paper, such technologies correspond to a global systems for mobile communications (GSM) technology variant in the time domain [2]. The spectrum sensing in cognitive radio (CR) provides the necessary information about the status of the wireless channels, modeling and prediction of communications activity.This could contribute to spectral efficiency improvement efforts [3][4][5].The prediction information of the channel status can be used by secondary users (SUs) to decide the sensing periods and channel occupancy duration for a single channel sensing scenario [6].Besides, based on prediction information, SUs can select the channels with higher probability of vacancy in multi-channel wideband sensing scenarios [7], and also primary user (PU) occupancy models can be used as empty channel indicators replacing the spectrum sensing procedures [2,8].A key issue for spectrum sensing operations is the evaluation of the so-called hidden node margin (HNM).The HNM has been evaluated in sharing scenarios between white space devices and digital terrestrial television systems [9]. Different initiatives for radioelectric spectrum channel modeling, mainly using deterministic and stochastic models, have been proposed [3,[10][11][12][13][14][15][16].Differences among those proposals and the present work lie in that in the first initiatives, the duty cycle (the average percentage of the channel utilization) time series is modeled for different types of channels, whereas the proposal presented here models the time series for received power in three GSM channels with different occupancy levels.For this purpose, firstly we used seasonal autoregressive integrated moving average (SARIMA) algorithm, which is adequate for analyzing time series with seasonality, in accordance with different studies [13][14][15].Secondly, the generalized autoregressive conditional heteroskedasticity (GARCH) algorithm, which had been applied in traffic modeling and forecast for different communications networks, was also used [17][18][19]. The evaluation of the results obtained in algorithm forecasts is based on the following variables: channel availability time (the time interval where the channel is not used by the PUs), channel occupancy time (time interval where the channel is used by the PUs), observation time and error criteria analysis (symmetrical mean absolute percentage error, SMAPE, mean absolute percentage error, MAPE, and mean absolute error, MAE) [20][21][22]. Theory and Background GARCH and SARIMA are deployed in a time series that is assumed as linear and with a known statistical distribution.As presented below, this is partially met in long-term analysis of time series measurements. Seasonal Autoregressive Integrated Moving Average Algorithm In general, if a time series exhibits potential seasonality indexed by s, then using a multiplied SARIMA(p,d,q)(P,D,Q)s algorithm is advantageous, where d is the level of non-seasonal differencing, p is the autoregressive (AR) non-seasonal order, q is the moving average (MA) non-seasonal order, P is the number of seasonal autoregressive terms, D is the number of seasonal differences, and Q is the number of seasonal moving average terms.The seasonal autoregressive integrated moving average algorithm of Box and Jenkins [23] is given in the Equation (1), where B is the backward shift operator, x t is the observed time series of load at time t, e t is the independent, identical, normally distributed error (random shock) at period t; ∇ D s x t = (1 − B s ) D x t , Φ p (B s ) and Θ Q (B s ) are the seasonal AR(p) and MA(q) operators, respectively, which are defined in Equations ( 2) and (3), where Φ 1, Φ 2 , . . ., Φ p are the parameters of the seasonal AR(p) model, Θ 1 , Θ 2 , . . ., Θ Q are the parameters of the seasonal MA(q) [24].The Box-Jenkins methodology consists of four iterative steps [25]: • Step 1: Identification.This step focuses on the selection of d, D, p, P, q and Q.The number of the order can be identified by observing the sample autocorrelations (ACF) and sample partial autocorrelations (PACF). • Step 2: Estimation.The historical data is used to estimate the parameters of the tentative model in Step 1. • Step 3: Diagnostic checking.Diagnostic test is used to check the adequacy of the tentative model. • Step 4: Forecasting.The final model in Step 3 is used to forecast the values [26]. Generalized Autoregressive Conditional Heteroskedastic Algorithm An important number of models, most of which have the property that conditional variance depends on past, have been proposed for capturing special data characteristics.Algorithms commonly used are those with autoregressive conditional heteroskedasticity (ARCH) introduced in [27] and generalized ARCH (GARCH) given by [28].Modeling ARCH-GARCH considers conditional error variance as a compression function of the past of the series. ARCH modeling usually requires a great number of lags (q), and therefore a great number of parameters.This might yield a model with a great number of parameters, which is in opposition to the parsimony principle.This fact drives many times to difficulties when using the model to describe data in an adequate way.On the contrary, a GARCH model uses an inferior quantity of parameters, which makes it preferable to an ARCH model [29][30][31].In this paper, the GARCH algorithm with order p ≥ 0 and q ≥ 0 for the discrete-time stochastic process r t is expressed in Equations ( 4) and ( 5) where ε t is an independent and identically distributed process with a zero mean and one standard deviation, µ is the mean constant offset, σ t 2 is variance, and α 0 is the constant in the conditional variance.Unknown parameters for model are α 0 , α i and β j for some positive integer p, q. Just as in an ARIMA model, ACF and PACF are useful for p and q order identification in a GARCH(p,q) process [30]. Case Study and Experiment Procedure The decision to carry out this study was made during the spectrum measurements campaign held in Bogota-Colombia where we obtained the measurements employed here from spectrum occupancy study previously carried out [1,32].The band analyzed was the GSM 850 MHz, as it is a band constantly used and viable for analysis in time function with conventional equipment, like a spectrum analyzer.The measurements used in this study correspond to a week, from 23 December to 29 December 2012.In some studies [33], it has been indicated that a reasonable option to obtain representative data without any a priori information about a band is to consider measurement periods of at least 24 h in order to avoid under or overestimating frequency bands occupancy with some temporary patterns.While a 24-h measurement period could be thought of as adequate in order to properly characterize the activity of determined spectrum bands [34], in this research 7 days were analyzed, including patterns for workdays and weekends.Additionally, this time period is sufficient to measure occupancy in mobile networks with low use, as indicated in [2,34]. The channels to be modeled were selected after measuring the duty cycles of 60 channels at GSM band.From these, three channels with different occupancy levels (high, medium and low), were chosen.Figure 1 presents results of power measure for three downlink channels during a week, with different power level.Spectrum analyzer configuration for this band was the following: a resolution bandwidth of 100 kHz with a sweep time of 290 ms, which guarantees GSM signal detection with a bandwidth of 200 kHz.Daily duty cycles from PUs at selected channels are shown in Figure 2. Threshold (λ) used, which for this event is of −89 dBm, was obtained from Equation ( 6) with a probability of false alarm (Pfa) of 1% [35].λ which is above of the detected noise floor of −102 dBm [1], where Г(.) y Γ(. , .) are complete and incomplete gamma functions, respectively, and m is the product of time times bandwidth.In this work the HNM is calculated from the differences between measurements of received power performed outside at street level and indoors, in a building.These measurements were performed with a discone-type antenna and spectrum analyzer.The calculation allows analysis of the margin between a non-licensed device indoors, and a PU outdoors without interference due to shadowing.Table 1 shows the HNM found for each channel. Low Channel HNM (dB) Medium Channel HNM (dB) High Channel HNM (dB) 9.2 11.9 6.8 Figures 3-5 present histograms corresponding to opportunities distribution during time periods of GSM band channels; it is observed that such opportunities have an exponential behavior, whose approximate equations and the coefficient of determination (R 2 ) are exhibited in each figure.Thus, the occurrence increases for the channel occupancy, especially for shorter time periods of use.For low, medium and high occupancy channels, total times of opportunities were approximately 84 h, 81 h and 78 h, respectively, which indicates relatively low occupancy [2].In this work the HNM is calculated from the differences between measurements of received power performed outside at street level and indoors, in a building.These measurements were performed with a discone-type antenna and spectrum analyzer.The calculation allows analysis of the margin between a non-licensed device indoors, and a PU outdoors without interference due to shadowing.Table 1 shows the HNM found for each channel.In this work the HNM is calculated from the differences between measurements of received power performed outside at street level and indoors, in a building.These measurements were performed with a discone-type antenna and spectrum analyzer.The calculation allows analysis of the margin between a non-licensed device indoors, and a PU outdoors without interference due to shadowing.Table 1 shows the HNM found for each channel. Table 1.Hidden node margin (HNM) for three different frequency channels for an urban environment. Low Channel HNM (dB) Medium Channel HNM (dB) High Channel HNM (dB) 9.2 11.9 6.8 Figures 3-5 present histograms corresponding to opportunities distribution during time periods of GSM band channels; it is observed that such opportunities have an exponential behavior, whose approximate equations and the coefficient of determination (R 2 ) are exhibited in each figure.Thus, the occurrence increases for the channel occupancy, especially for shorter time periods of use.For low, medium and high occupancy channels, total times of opportunities were approximately 84 h, 81 h and 78 h, respectively, which indicates relatively low occupancy [2].Following this, we proceeded to analyze the time series of measured channels over a week, which was equivalent to 1,062,514 samples.To do this, ACF is initially presented, as observed in Figure 6.ACF diagrams for the three channels presents forms which are alternately positive and negative, decaying to zero, the values are in 95%-confidence intervals, shown with the blue lines.Therefore, this indicates that there is correlation [2,26].Following this, we proceeded to analyze the time series of measured channels over a week, which was equivalent to 1,062,514 samples.To do this, ACF is initially presented, as observed in Figure 6.ACF diagrams for the three channels presents forms which are alternately positive and negative, decaying to zero, the values are in 95%-confidence intervals, shown with the blue lines.Therefore, this indicates that there is correlation [2,26].Following this, we proceeded to analyze the time series of measured channels over a week, which was equivalent to 1,062,514 samples.To do this, ACF is initially presented, as observed in Figure 6.ACF diagrams for the three channels presents forms which are alternately positive and negative, decaying to zero, the values are in 95%-confidence intervals, shown with the blue lines.Therefore, this indicates that there is correlation [2,26].Following this, we proceeded to analyze the time series of measured channels over a week, which was equivalent to 1,062,514 samples.To do this, ACF is initially presented, as observed in Figure 6.When analyzing channels stationarity of Figure 6, it is observed that the mean and variance are constant and similar to each other, on each one of the days from Monday to Friday.Therefore, measurements at the weekend are not taken into consideration when training the analyzed models, because the mean and variance are not similar and change in a significant way with regard to the measurements from Monday to Friday. Design of SARIMA Algorithm In Figure 7 the trend and seasonality are presented in occupancy level for the three channels.Seasonality had a period of 24 h, practically without trend and with stationary components, which makes possible the use of a SARIMA model to forecast behavior of the GSM channels [26]. Delay difference s, which for this event is selected as five (∆5), was equivalent to the number of days of the week in which the signal was stationary [15].Applying the augmented Dickey-Fuller test [36], in the series of three channels from Monday to Friday, the null hypothesis of unit root is rejected, which indicates stationarity.In order to find the parameters of SARIMA(p,d,q)(P,D,Q)s model, ACF and PACF were calculated for ∆5 of respective channels, as shown in Figure 8.When analyzing channels stationarity of Figure 6, it is observed that the mean and variance are constant and similar to each other, on each one of the days from Monday to Friday.Therefore, measurements at the weekend are not taken into consideration when training the analyzed models, because the mean and variance are not similar and change in a significant way with regard to the measurements from Monday to Friday. Design of SARIMA Algorithm In Figure 7 the trend and seasonality are presented in occupancy level for the three channels.Seasonality had a period of 24 h, practically without trend and with stationary components, which makes possible the use of a SARIMA model to forecast behavior of the GSM channels [26].When analyzing channels stationarity of Figure 6, it is observed that the mean and variance are constant and similar to each other, on each one of the days from Monday to Friday.Therefore, measurements at the weekend are not taken into consideration when training the analyzed models, because the mean and variance are not similar and change in a significant way with regard to the measurements from Monday to Friday. Design of SARIMA Algorithm In Figure 7 the trend and seasonality are presented in occupancy level for the three channels.Seasonality had a period of 24 h, practically without trend and with stationary components, which makes possible the use of a SARIMA model to forecast behavior of the GSM channels [26]. Delay difference s, which for this event is selected as five (∆5), was equivalent to the number of days of the week in which the signal was stationary [15].Applying the augmented Dickey-Fuller test [36], in the series of three channels from Monday to Friday, the null hypothesis of unit root is rejected, which indicates stationarity.In order to find the parameters of SARIMA(p,d,q)(P,D,Q)s model, ACF and PACF were calculated for ∆5 of respective channels, as shown in Figure 8. Delay difference s, which for this event is selected as five (∆ 5 ), was equivalent to the number of days of the week in which the signal was stationary [15].Applying the augmented Dickey-Fuller test [36], in the series of three channels from Monday to Friday, the null hypothesis of unit root is rejected, which indicates stationarity.In order to find the parameters of SARIMA(p,d,q)(P,D,Q)s model, ACF and PACF were calculated for ∆ 5 of respective channels, as shown in Figure 8.Using Box-Jenkins methodology [23], Figure 8 shows that PACF of ∆5 decays to zero with a seasonal pattern, and crosses confidence level initially in lag 5 for negative side.This suggests that a term non-seasonal AR(1) could be used, and a seasonal MA(5) could be added. Design of GARCH Algorithm When analyzing in detail the large amount of acquired information, the existence of standard deviation was observed; therefore the GARCH algorithm was used to forecast the behavior of measured series.Stochastic models ARIMA and SARIMA are methods for univariate modeling.The main difference among former models and GARCH model lies in the constant variance assumption. Even though for the developed algorithm there is stationarity in original signal from Monday to Friday, for this case the fifth difference is developed because there is a greater degree of stationarity.In Figure 9 the difference for each channel is presented.Channel measurements are converted into returns by logarithmic transformation.The logarithmic returns are defined in Equation (10) Using Box-Jenkins methodology [23], Figure 8 shows that PACF of ∆ 5 decays to zero with a seasonal pattern, and crosses confidence level initially in lag 5 for negative side.This suggests that a term non-seasonal AR(1) could be used, and a seasonal MA(5) could be added. Design of GARCH Algorithm When analyzing in detail the large amount of acquired information, the existence of standard deviation was observed; therefore the GARCH algorithm was used to forecast the behavior of measured series.Stochastic models ARIMA and SARIMA are methods for univariate modeling.The main difference among former models and GARCH model lies in the constant variance assumption. Even though for the developed algorithm there is stationarity in original signal from Monday to Friday, for this case the fifth difference is developed because there is a greater degree of stationarity. In Figure 9 the difference for each channel is presented.Channel measurements are converted into returns by logarithmic transformation.The logarithmic returns are defined in Equation ( 10), r t = ln P t P t−1 (10) where P t is power value in time t and P t−1 is power value in time t−1 . Algorithms 2016, 9, 82 8 of 18 where Pt is power value in time t and Pt−1 is power value in time t−1.Now we present a formal statistical test in order to establish the presence of ARCH effects in the data and correlation.H = 0 implies that there exist no significant correlation as well as H = 1 indicates that there exists a significant correlation.In Tables 3 and 4 If the time series is the result of a completely random phenomenon, the autocorrelation should be close to zero for all time-lag separations.Otherwise, one or more of the autocorrelations will be significantly different from zero.Another useful way to examine dependencies of the series is to revise the PACF, where the dependence of intermediate elements (those within the lag) is eliminated.In Figure 10, graphs of ACF, PACF and ACF of square returns present the existence of correlation in data of channel occupancy.Now we present a formal statistical test in order to establish the presence of ARCH effects in the data and correlation.H = 0 implies that there exist no significant correlation as well as H = 1 indicates that there exists a significant correlation.In Tables 3 and 4, all the p values show that Ljung-Box-Pierce Q-Test and Engle ARCH Test in lag 10, 15 and 20 are significant, revealing the presence of ARCH effects (heteroskedasticity), indicating that GARCH modeling is appropriate.Dependence in data x 1 , . . ., x n was determined by computing correlations.This was done by representing the ACF. If the time series is the result of a completely random phenomenon, the autocorrelation should be close to zero for all time-lag separations.Otherwise, one or more of the autocorrelations will be significantly different from zero.Another useful way to examine dependencies of the series is to revise the PACF, where the dependence of intermediate elements (those within the lag) is eliminated. In Figure 10, graphs of ACF, PACF and ACF of square returns present the existence of correlation in data of channel occupancy.Below, in Tables 5-7, the evaluation and selection of the GARCH model for each channel was performed.Below, in Tables 5-7, the evaluation and selection of the GARCH model for each channel was performed.The GARCH model selection for each channel was done by fulfilling α i + β i < 1 criterion, so the model is stationary, and then taking into account the more proximate values to zero of MAE, MAPE and SMAPE from Tables 5-7.Therefore, the selected models for low, medium and high channel are GARCH(2,2), GARCH(0,2) and GARCH(0,1), respectively. Parameters for low channel model were estimated and are presented in Table 8.GARCH (2,2), where Thus, the model according to Table 8 is presented in Equations ( 11) and ( 12), For medium channel, GARCH(0,2), model values presented in Table 9 are estimated.Therefore, Equations ( 13) and ( 14) are obtained, For high channel, GARCH(0,1), the following parameters were obtained, as shown in Table 10.Then the model is described in Equations ( 15) and ( 16), ARCH-GARCH model analysis is based on evaluation of standardized residuals [31].One assumption with GARCH model is that for a good model, residuals should follow a white noise process.This is to say that it is expected that residuals be at random, independent and identically distributed, following a normal distribution.Figure 11 presents the relationship between innovations (residuals) derivate from adjusted model, the corresponding conditional standard deviations and returns.Figure 11 shows that both innovations and returns exhibit variations.In the following we intend to find out if by performing GARCH the autocorrelation of the standardized innovations disappears, which would indicate the effectiveness of GARCH model.distributed, following a normal distribution.Figure 11 presents the relationship between innovations (residuals) derivate from adjusted model, the corresponding conditional standard deviations and returns.Figure 11 shows that both innovations and returns exhibit variations.In the following we intend to find out if by performing GARCH the autocorrelation of the standardized innovations disappears, which would indicate the effectiveness of GARCH model.In Tables 11 and 12, results of Ljung-Box-Pierce Q-Test and Engle ARCH test for later analysis are presented using standardized innovations.These tests indicate no presence of correlation or ARCH effects.We have GARCH effects and also correlation between innovations that disappear after treating the data.Therefore, the GARCH model is a proper model for explaining the variances of the three channels.distributed, following a normal distribution.Figure 11 presents the relationship between innovations (residuals) derivate from adjusted model, the corresponding conditional standard deviations and returns.Figure 11 shows that both innovations and returns exhibit variations.In the following we intend to find out if by performing GARCH the autocorrelation of the standardized innovations disappears, which would indicate the effectiveness of GARCH model.In Tables 11 and 12, results of Ljung-Box-Pierce Q-Test and Engle ARCH test for later analysis are presented using standardized innovations.These tests indicate no presence of correlation or ARCH effects.We have GARCH effects and also correlation between innovations that disappear after treating the data.Therefore, the GARCH model is a proper model for explaining the variances of the three channels.In Tables 11 and 12, results of Ljung-Box-Pierce Q-Test and Engle ARCH test for later analysis are presented using standardized innovations.These tests indicate no presence of correlation or ARCH effects.We have GARCH effects and also correlation between innovations that disappear after treating the data.Therefore, the GARCH model is a proper model for explaining the variances of the three channels.Normality verification was performed by analyzing histograms of residuals and normal probability graph, as shown in Figure 13.The histograms of the three channels shows that the residuals are normally distributed.In turn, the probability graph confirms that residuals respond to a normal distribution, since most of data are spread along the straight line.Normality verification was performed by analyzing histograms of residuals and normal probability graph, as shown in Figure 13.The histograms of the three channels shows that the residuals are normally distributed.In turn, the probability graph confirms that residuals respond to a normal distribution, since most of data are spread along the straight line. Results and Discussion In Figure 14 an example is displayed where there is application of the designed time series algorithms.Here, the interaction between the CR user and the primary base station (BS) is shown by received power from the primary transmitter.This is represented by the oval and the direction of the arrows.The CR user can forecast the power level it will receive from the primary BS. In the example of Figure 14, in order to analyze the SARIMA and GARCH algorithms, the forecast of the power is performed by the CR user, making a comparison with the spectrum analyzer in which the measurements were made.However, this depends on the architecture of CR deployed in the environment.Due to the processor and power consumption being more limited in the CR user's computer, use of infrastructure architecture is recommended, where the forecast is carried out by the CR BS.This provides a better processor than that of the CR user, and has no limitations on power consumption.However, there is a time period between data capture in the environment and the processing, which adds a delay to the response.This must not be ignored, but the forecast helps to reduce the negative impact of the delayed response. Results and Discussion In Figure 14 an example is displayed where there is application of the designed time series algorithms.Here, the interaction between the CR user and the primary base station (BS) is shown by received power from the primary transmitter.This is represented by the oval and the direction of the arrows.The CR user can forecast the power level it will receive from the primary BS. In the example of Figure 14, in order to analyze the SARIMA and GARCH algorithms, the forecast of the power is performed by the CR user, making a comparison with the spectrum analyzer in which the measurements were made.However, this depends on the architecture of CR deployed in the environment.Due to the processor and power consumption being more limited in the CR user's computer, use of infrastructure architecture is recommended, where the forecast is carried out by the CR BS.This provides a better processor than that of the CR user, and has no limitations on power consumption.However, there is a time period between data capture in the environment and the processing, which adds a delay to the response.This must not be ignored, but the forecast helps to reduce the negative impact of the delayed response. The analysis of the precision of the forecast made with the SARIMA and GARCH algorithms is presented, as follows.Figure 15 shows the SARIMA algorithm forecasts obtained from Equations ( 7)-( 9), and from GARCH algorithm based on Equations ( 11)-( 16).This was contrasted with the power measured data of Friday from 5:00 p.m. to 6:00 p.m.This period was chosen since during this time an increased use of the channels by PUs was perceived.In Figures 16 and 17, availability and occupancy times of the measured and forecast channels are presented.Availability time allows us to analyze the precision with which SUs could use availability time in GSM channels for a CR system.In the same way, occupancy time examines the precision during time in which PUs use GSM channels.Average precisions obtained between actual and forecast data for availability times were: 82%, 54% and 60%, for the SARIMA algorithm; and 31%, 30% and 43%, for the GARCH algorithm, corresponding to channels with low, medium and high occupation, respectively.Average precisions for occupation times between real data and forecast data are equivalent to 58%, 77% and 78% for the SARIMA algorithm; and 44%, 46.6% and 44.2%, for the GARCH algorithm, corresponding to channels with low, medium and high occupancy levels, respectively.Additionally, as expected, for each algorithm there is an inversely proportional relationship between channel occupancy and availability time, as well as a directly proportional relationship between occupancy probability and occupancy time of the channels.The analysis of the precision of the forecast made with the SARIMA and GARCH algorithms is presented, as follows.Figure 15 shows the SARIMA algorithm forecasts obtained from Equations ( 7)-( 9), and from GARCH algorithm based on Equations ( 11)-( 16).This was contrasted with the power measured data of Friday from 5:00 p.m. to 6:00 p.m.This period was chosen since during this time an increased use of the channels by PUs was perceived.The analysis of the precision of the forecast made with the SARIMA and GARCH algorithms is presented, as follows.Figure 15 shows the SARIMA algorithm forecasts obtained from Equations ( 7)-( 9), and from GARCH algorithm based on Equations ( 11)-( 16).This was contrasted with the power measured data of Friday from 5:00 p.m. to 6:00 p.m.This period was chosen since during this time an increased use of the channels by PUs was perceived.In Figures 16 and 17, availability and occupancy times of the measured and forecast channels are presented.Availability time allows us to analyze the precision with which SUs could use availability time in GSM channels for a CR system.In the same way, occupancy time examines the precision during time in which PUs use GSM channels.Average precisions obtained between actual and forecast data for availability times were: 82%, 54% and 60%, for the SARIMA algorithm; and 31%, 30% and 43%, for the GARCH algorithm, corresponding to channels with low, medium and high occupation, respectively.Average precisions for occupation times between real data and forecast data are equivalent to 58%, 77% and 78% for the SARIMA algorithm; and 44%, 46.6% and 44.2%, for the GARCH algorithm, corresponding to channels with low, medium and high occupancy levels, respectively.Additionally, as expected, for each algorithm there is an inversely proportional relationship between channel occupancy and availability time, as well as a directly proportional relationship between occupancy probability and occupancy time of the channels.In Figures 16 and 17, availability and occupancy times of the measured and forecast channels are presented.Availability time allows us to analyze the precision with which SUs could use availability time in GSM channels for a CR system.In the same way, occupancy time examines the precision during time in which PUs use GSM channels.Average precisions obtained between actual and forecast data for availability times were: 82%, 54% and 60%, for the SARIMA algorithm; and 31%, 30% and 43%, for the GARCH algorithm, corresponding to channels with low, medium and high occupation, respectively.Average precisions for occupation times between real data and forecast data are equivalent to 58%, 77% and 78% for the SARIMA algorithm; and 44%, 46.6% and 44.2%, for the GARCH algorithm, corresponding to channels with low, medium and high occupancy levels, respectively.Additionally, as expected, for each algorithm there is an inversely proportional relationship between channel occupancy and availability time, as well as a directly proportional relationship between occupancy probability and occupancy time of the channels.In Table 13 forecast and measured data are compared to different methods for error estimation such as SMAPE, MAPE and MAE.Results of Table 13 point that only in data forecast for the medium occupancy channel, the GARCH algorithm presents smaller errors than the SARIMA algorithm.Comparison of performance in forecast is shown in Figure 18, developed on a computer with dual-core processor 2.4 GHz and 4 GB of random access memory (RAM) memory.Here is observed that in general, for each model, the higher the observation time the lower the prediction error.However, this entails no significance.Prediction error is lower in forecast of GARCH algorithms, but In Table 13 forecast and measured data are compared to different methods for error estimation such as SMAPE, MAPE and MAE.Results of Table 13 point that only in data forecast for the medium occupancy channel, the GARCH algorithm presents smaller errors than the SARIMA algorithm.Comparison of performance in forecast is shown in Figure 18, developed on a computer with dual-core processor 2.4 GHz and 4 GB of random access memory (RAM) memory.Here is observed that in general, for each model, the higher the observation time the lower the prediction error.However, this entails no significance.Prediction error is lower in forecast of GARCH algorithms, but In Table 13 forecast and measured data are compared to different methods for error estimation such as SMAPE, MAPE and MAE.Results of Table 13 point that only in data forecast for the medium occupancy channel, the GARCH algorithm presents smaller errors than the SARIMA algorithm. Conclusions The SARIMA and GARCH algorithms have been evaluated in this paper in order to forecast the reception power in channels of a GSM band.Even though GARCH algorithm presented lower prediction errors than SARIMA algorithm, the use of a SARIMA algorithm is more convenient for a CR system, because it has higher precisions with respect to availability and occupancy times, with which the use of spectrum efficiency is improved and the interference level and collisions between PUs and SUs will be reduced.Additionally, the SARIMA algorithm employs lower observation times than the GARCH algorithm.As noted, for the best of cases, observation times lower than four seconds could be obtained for the three GSM channels, which is an advantage in practical CR systems.For a CR system, the forecast developed in the GSM band could help to improve the use of spectral efficiency, since it would allow CR users to share channels and avoid collisions with PUs in the found opportunities. The SARIMA and GARCH algorithms forecast not only the reception power; but the occupation and availability times for GSM channels.It would also be feasible to use the training data from one day for the forecast of a CR user's received power from a primary BS. The significance of the forecast of received power is that CR users can save energy in the process of detecting the spectrum and take advantage of spectral opportunities, thereby increasing the rate of successful transmission and transmission opportunities, reducing the time to find an available channel, and adjusting transmission power levels to protect against collisions and interference with the PUs. Figure 1 . Figure 1.Power measurements for three global systems for mobile communications (GSM) band downlink channels. Figure 2 . Figure 2. Duty cycles for three GSM band downlink channels. Figure 1 . Figure 1.Power measurements for three global systems for mobile communications (GSM) band downlink channels. Figure 2 . Figure 2. Duty cycles for three GSM band downlink channels. Figure 2 . Figure 2. Duty cycles for three GSM band downlink channels. Figure 4 . Figure 4. Time period opportunities distribution for medium channel. Figure 5 . Figure 5.Time period opportunities distribution for high channel. Figure 3 . 18 Figure 3 . Figure 3.Time period opportunities distribution for low channel. Figure 4 . Figure 4. Time period opportunities distribution for medium channel. Figure 5 . Figure 5.Time period opportunities distribution for high channel. Figure 4 . 18 Figure 3 . Figure 4. Time period opportunities distribution for medium channel. Figure 4 . Figure 4. Time period opportunities distribution for medium channel. Figure 5 . Figure 5.Time period opportunities distribution for high channel. Figure 5 . Figure 5.Time period opportunities distribution for high channel. Figure 6 . Figure 6.Autocorrelation for three GSM band downlink channels. Figure 7 . Figure 7. Seasonality and trend components of the GSM channels. Figure 6 . Figure 6.Autocorrelation for three GSM band downlink channels. Figure 6 . Figure 6.Autocorrelation for three GSM band downlink channels. Figure 7 . Figure 7. Seasonality and trend components of the GSM channels. Figure 7 . Figure 7. Seasonality and trend components of the GSM channels. Figure 8 . Figure 8. Simple and partial autocorrelation for GSM channels. Figure 8 . Figure 8. Simple and partial autocorrelation for GSM channels. Figure 9 . Figure 9. Fifth difference of measured powers in channels of GSM band. , all the p values show that Ljung-Box-Pierce Q-Test and Engle ARCH Test in lag 10, 15 and 20 are significant, revealing the presence of ARCH effects (heteroskedasticity), indicating that GARCH modeling is appropriate. Figure 9 . Figure 9. Fifth difference of measured powers in channels of GSM band. Figure 10 . Figure 10.Correlation graphs for GSM band channels. Figure 10 . Figure 10.Correlation graphs for GSM band channels. Figure 11 . Figure 11.Innovations, conditional standard deviations and returns of GSM channels. Figure 12 Figure 12 corresponds to the autocorrelation of the squared standardized innovations, in which correlation was not observed. Figure 12 . Figure 12.Autocorrelation of the squared standardized innovations of GSM channels. Figure 11 . Figure 11.Innovations, conditional standard deviations and returns of GSM channels. Figure 12 Figure 12 corresponds to the autocorrelation of the squared standardized innovations, in which correlation was not observed. Figure 11 . Figure 11.Innovations, conditional standard deviations and returns of GSM channels. Figure 12 Figure 12 corresponds to the autocorrelation of the squared standardized innovations, in which correlation was not observed. Figure 12 . Figure 12.Autocorrelation of the squared standardized innovations of GSM channels. Figure 12 . Figure 12.Autocorrelation of the squared standardized innovations of GSM channels. Figure 13 . Figure 13.Histogram of residuals and normal probability for GSM channels. Figure 13 . Figure 13.Histogram of residuals and normal probability for GSM channels. Figure 14 . Figure 14.Example of application to forecast the received power from the base station (BS).CR: cognitive radio. Figure 15 . Figure 15.GSM channel series and forecast series with SARIMA and GARCH algorithms. Figure 14 . Figure 14.Example of application to forecast the received power from the base station (BS).CR: cognitive radio. Figure 14 . Figure 14.Example of application to forecast the received power from the base station (BS).CR: cognitive radio. Figure 15 . Figure 15.GSM channel series and forecast series with SARIMA and GARCH algorithms. Figure 15 . Figure 15.GSM channel series and forecast series with SARIMA and GARCH algorithms. Table 1 . Hidden node margin (HNM) for three different frequency channels for an urban environment. Table 1 . Hidden node margin (HNM) for three different frequency channels for an urban environment. Table 5 . Generalized autoregressive conditional heteroskedasticity (GARCH) model comparison for low channel.SMAPE: symmetrical mean absolute percentage error; MAPE: mean absolute percentage error; MAE: mean absolute error. Table 6 . GARCH model comparison for medium channel. Table 7 . GARCH model comparison for high channel. Table 5 . Generalized autoregressive conditional heteroskedasticity (GARCH) model comparison for low channel.SMAPE: symmetrical mean absolute percentage error; MAPE: mean absolute percentage error; MAE: mean absolute error. Table 6 . GARCH model comparison for medium channel. Table 7 . GARCH model comparison for high channel. Table 8 . Parameters estimation for low channel model. Table 9 . Parameters estimation for medium channel model. Table 10 . Parameters estimation for high channel model. Table 11 . Ljung-Box-Pierce Q-Test in standardized innovations for GSM channels. Table 12 . Engle ARCH test in standardized innovations for GSM channels. Table 11 . Ljung-Box-Pierce Q-Test in standardized innovations for GSM channels. Table 12 . Engle ARCH test in standardized innovations for GSM channels. Table 13 . Error variables comparison for forecasted values. Table 13 . Error variables comparison for forecasted values. Table 13 . Error variables comparison for forecasted values. Table 16 . Result of mean squared error for high channel with different number of training days.	2017-01-15T08:35:26.413Z	2016-12-02T00:00:00.000	{ "year": 2016, "sha1": "9f4e2cfe8bf6ff09105107844a7004254aa75d5b", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1999-4893/9/4/82/pdf?version=1480675407", "oa_status": "GOLD", "pdf_src": "Crawler", "pdf_hash": "9f4e2cfe8bf6ff09105107844a7004254aa75d5b", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Computer Science", "Mathematics" ] }
73913223	pes2o/s2orc	v3-fos-license	The effects of incisional bupivacaine infusions on postoperative opioid consumption and pain scores after total abdominal hysterectomy Abstract Background: The aim of this study was to determine opioid requirements and pain intensity scores in patients after a total abdominal hysterectomy (TAH) administered with a bupivacaine infusion for a 30-hour period, and then to compare the data with that of a control group. Method: This was a prospective, parallel, single-blinded randomised trial which took place at the Rahima Moosa Mother and Child Hospital, Johannesburg. Thirty-six consenting patients, who underwent a TAH, were randomised to either having a 0.39% bupivacaine infusion in the incisional site or not. Morphine was administered via a patient-controlled analgesia pump (PCA) for rescue analgesia. Dynamic, static and worst pain scores were assessed one, six and 30 hours after surgery by using a visual analogue scale (VAS). Morphine consumption was recorded at set intervals. Results: There were statistically significant differences between the two groups' dynamic VAS scores in the first hour and at 24 hours and 30 hours; in the static VAS score in the first hour; and in the VAS scores for the worst pain experienced since the patients were last seen in the first hour and six hours after the operation. There was no statistical difference between the two groups' opioid consumption at all set observation points. Conclusion: The opioid requirements of the two groups were comparable, although participants who had the bupivacaine infusion experienced reduced pain intensity which lasted until six hours postoperatively, and also had reduced pain intensity when moving around 30 hours after the operation. Introduction Pain is rated as a highly undesirable postoperative outcome. Postoperative pain, especially when poorly controlled, results in acute adverse physiological responses and chronic effects. 1 The two modalities of pain relief are systemic (opioid and non-opioid) analgesia and regional analgesia. Local analgesia is a well-recognised component in multimodal analgesia. It is inexpensive and relatively safe and simple to use. Infusing local anaesthetics through catheters is a new and evolving area of postoperative pain management. 2,3 In the international arena of postoperative pain control, many pain control protocols are available for total abdominal hysterectomy (TAH). The Procedure-Specific Postoperative Pain Management (PROSPECT) website contains recommendations of several postoperative pain protocols for patients undergoing TAH. 4 This committee has made recommendations based on evidence collected from randomised controlled trials. The committee suggests that after TAH, patients should be given "strong" opioids via an intravenous patient-controlled analgesia pump (PCA), or by fixed intravenous dosing titrated to pain intensity. Continuous wound infiltration with a local anaesthetic after closure is not recommended, as there is limited procedurespecific evidence that is currently available. 5 Pain relief provided by a local anaesthetic that is continuously infused in the incisional site has not yet been tried out in a public hospital. This technique of acute pain management has also not been studied in a healthcare setting in a developing country such as South Africa. This was evident from a PubMed medical subject headings (MESH) search, in October 2008, for the following terms: "South Africa", "local anaesthetic infusions" and "postoperative pain management in South Africa". 6 The aim of this study was to determine whether there is a reduction in patients' opioid requirements after TAH, and a decrease in the patients' pain intensity as a result of a bupivacaine infusion into the incisional site. Method Approval to conduct the study at the Rahima Moosa Mother and Child Hospital in Johannesburg was obtained from the ethics committee of the University of the Witwatersrand, the postgraduate committee of the University of the Witwatersrand, and the hospital superintendent. Thirty-six patients enrolled for the study, which was a contextual, prospective, parallel and single-blinded randomised trial. The selected participants were patients with an American Society of Anesthesiologists (ASA) physical status of one or two, scheduled for an elective TAH that required a Pfannenstiel incision. The following patients were excluded from the clinical trial: those who had a contraindication to general anaesthesia, an allergy to any of the study medications, a history of alcohol/drug abuse, a major medical disease such as cardiovascular, pulmonary, metabolic, renal, neurological or psychiatric disease, and patients with a clinically significant bacterial infection. Patients scheduled for a TAH were assessed preoperatively by the investigator. If the inclusion criteria were met, informed consent to participate in the trial was obtained. Patients who consented to participate in this clinical study were randomly assigned to receive either a bupivacaine infusion into their incisional sites and a morphine PCA pump, or a morphine PCA pump only. A consecutive convenience sampling method was used, and the control device was inserted into alternate patients. The patients were assessed the day before the operation, required to fast, and given no analgesic premedication. Intraoperatively, the study participants were given a uniform general anaesthesia with opioid analgesia only. All patients were subjected to standard monitoring (noninvasive arterial blood pressure, heart rate, peripheral oxygen saturation, end-tidal gas monitoring and an electrocardiogram). After intravenous cannulation, anaesthesia was induced with opioids (either fentanyl, alfentanil or morphine) and propofol until loss of consciousness was established. The doses were titrated to effect on the patient and varied accordingly. Tracheal intubation was performed after muscle relaxation with a non-depolarising muscle relaxant of the anaesthetists' choice. Anaesthesia was maintained with air and oxygen. The inhalational anaesthetic was either isoflurane or sevoflurane. Mechanical ventilation was used in a low-flow system to maintain end-tidal CO 2 of 35-45 mmHg. Doses were titrated according to the patients' body mass and effect. At the end of the operation, muscle relaxation was reversed with glycopyrrolate and neostigmine in adequate doses, and the inhalational anaesthetic was turned off. After satisfactory spontaneous ventilation and awakening, the patients were extubated and transferred to the recovery area. Standard postoperative observations took place, and face mask oxygen was supplied via Venturi masks. Once the recovery sisters were satisfied with the patients' condition, they discharged the patients to the ward, where further standard postoperative observations were performed by the nursing staff. Surgery was preformed in a standardised manner using a Pfannenstiel incision of approximately 10-15 cm, depending on the patients' body habitus. None of the participants experienced extensive blood loss requiring blood transfusion. The device group participants had On-Q PainBuster Soaker ™ 6.5 pain relief system [270 ml volume, 4 ml/hour; I-Flow Corporation, USA (all within their expiry dates)] devices inserted. This was accomplished at wound closure, when a multi-holed catheter was inserted by the gynaecologist along the length of the incisional site under the abdominal fascia. The placement method was as per the manufacturer's instructions, using the Z-track method. The fascial layer was closed with sutures over the catheter. After closure of the fascia, the second catheter was inserted in a similar manner as the subfascial catheter, but from the opposite side, and above the subfascial catheter. Once the skin was closed, a 5 ml bolus of 0.5% bupivacaine was injected through each catheter, infiltrating the incision. The catheters were secured to the skin by coiling the catheter with tape. Using an aseptic technique, these catheters were then connected to a 270 ml elastomeric disposable balloon pump with 0.39% bupivacaine solution. Opening the clamps on the catheter started the drug infusion. The drug was infused at 4 ml/hour (15.6 mg/hour) for 30 hours. The control participants had a sterile bandage placed over the wound site and a catheter (positioned on top of the bandage that was coiled) connected to apparatus similar to that of the trial group. The catheter was taped and covered by another bandage. The catheter neither penetrated the wound site, nor infused any substance. Both groups had the pump apparatus concealed in a black bag. After surgery, all the patients were connected to a morphine CADD-Legacy ® PCA Pump Model 6300s (Smith Medical). The PCA protocol was 1 mg morphine bolus dose, with a lockout of six minutes for breakthrough pain. The maximum dose of morphine was 10 mg/hour. This PCA pump was set up to manage breakthrough pain. Before surgery, the patients were given instructions on how to use it. No other pain analgesia was prescribed. If the patients complained of nausea or vomiting, prochlorperazine 12.5 mg was administered intramuscularly. Promethazine 25 mg intramuscularly (eight hourly) was prescribed for patients with itchiness. After 30 hours, the catheter was withdrawn and the intravenous line was taken down, along with the morphine PCA pump. Regular diclofenac suppositories (100 mg every 18 hours) and paracetamol (1 g orally every six hours) were prescribed for analgesia. The surgical team then discharged the patients. Data collection was performed by two assessors at one, six, 24 and 30 hours postoperatively. With regard to the measurement taking, the time at which the infusion of the study drug was started was considered to be Time 0. Parameters assessed were visual analogue scale (VAS) scores for the worst pain experienced since the last observation, static pain intensity at the time of observation, and dynamic pain intensity at the time of observation. Total morphine consumption was recorded for the periods 0-1 hour, 1-6 hours, 6-24 hours and 24-30 hours postoperatively. In addition, opioid adverse effects, namely nausea, vomiting and itchiness, were recorded. The day of discharge and any perioperative complications were noted. Data was collected by the investigator and compiled on a Microsoft Excel (2003) data table. Results Data analysis was conducted in consultation with the biostatistician. Testing was carried out at the 0.05 level of significance. The opioid consumption and the VAS scores of the two groups for the set observation periods were verified with the two sample t-tests with unequal variances, and two sample Wilcoxon rank-sum (Mann-Whitney) tests. The noncontinuous variables, namely the adverse effects of the opioids (nausea, vomiting and itchiness) were tested with the Fischer exact test. An analysis of the demographic information, namely the study participants' age, body mass index, type of operation and racial group, and a comparison of the two groups using parametric and non-parametric testing, indicated that there were no statistical differences between these two groups. There was no statistical difference between the two groups regarding opioid consumption. The mean opioid consumption in the device group was 3.35 mg, 10.83 mg, 25.11 mg and 5.78 mg in the first, second, third and fourth periods respectively. This can be compared to the mean opioid consumption in the control group, which was 4.89 mg, 8.22 mg, 24.67 mg and 8.33 mg in the respective observation periods. The following p-values are obtained when equal variance is assumed and tested again for unequal variance. The p-values at the first set period were 0.134 and 0.145. The p-values at the second set period were 0.302 and 0.369. At the third set period, the p-values were 0.922 and 0.544, and for the last set period, the p-values were 0.094 and 0.090 respectively. Figure 1 is a line graph illustrating the opioid consumption of the participants at the set observation periods. Opioid adverse effects in the control and device groups were analysed with the Fischer's exact test and showed no statistical difference between the groups for the incidence of nausea, vomiting and itchiness. The mean dynamic VAS scores of the device group in the first, third and fourth periods were 39.42, 39.17 and 35.36 respectively. This is in comparison with the mean dynamic VAS scores of the control group in the first, third and fourth periods, which were 67.17, 60.89 and 54.31 respectively. The p-values were 0.013, 0.019 and 0.023 respectively. This is illustrated in Figure 2. The significant p-values are asterisked. The mean static VAS score of the device group in the first period was 34.89, compared to the mean static VAS score of the control group of 59.25. This resulted in a statistically significant p-value of 0.038. The p-values for the second set period were 0.887 and 0.596 respectively. At the third set period, the p-values were 0.405 and 0.921 respectively, while the p-values for the last set period were 0.339 and 0.231 respectively. These data are depicted in Figure 3. The VAS scores of the control group, in the first and second observation period, for the worst pain experienced since the patients were last seen, were 72.45 and 60.43 respectively. The mean VAS scores of the device group, for the same periods, for the worst pain experienced since the patients were last seen were 46.89 and 35.86 respectively. The p-values were 0.008 and 0.023 respectively. At the third set period, the p-value value was 0.704, while it was 0.711 in the last set period. These data are illustrated in Figure 4. Discussion The two clinical study groups shared a similar demographic profile, to ensure the validity of the measured study findings. Statistical differences were seen between the dynamic VAS scores in the first (one hour postoperatively), third (24 hours postoperatively) and fourth (30 hours postoperatively) observation periods. There was also a statistical difference in the static VAS score in the first observation period. There was a statistical difference in the VAS scores for the worst pain experienced since the patients were last seen in the first and second observation periods. Movement by a patient elicits somatic pain, rather than visceral pain. The decrease in dynamic VAS scores could be an indication that the bupivacaine infusion decreased the somatic pain at the incisional site. By contrast, from six hours postoperatively, the static VAS scores were not different, as the bupivacaine infusion did not provide any pain relief from the visceral component originating mainly from the peritoneum. The reason for the difference in VAS scores, but similarity in opioid consumption, is not clear. One could speculate that the bupivacaine infusion helped relieve somatic pain at the incisional site, but not the visceral component of pain after the operation, which needed morphine boluses to be eased. Psychological factors, beliefs and expectations were not tested in this trial. The VAS scores did not correlate with the participants' opioid consumption and require further evaluation. Limitations and logistical issues The study population was not representative of the group of women undergoing TAH in South Africa as a whole. However, the study sample addressed a clinical setting at a public hospital in central Johannesburg, which is relevant locally. The clinical setting of the study resulted in single-blinded conditions while the evaluations were being performed. The patients did not know whether or not they were receiving a bupivacaine infusion in their incisional sites, as all participants had the same external apparatus. The researchers and surgeon knew this. Scientific guidelines advocated the placement of a subfascial catheter in all the participants, whereas evident ethical reasons restrained us in applying this method. Bias may have occurred as the design of the trial was not double-blinded. The application of the subfascial catheter prolonged the anaesthetic time minimally, and the surgeons found the technique of inserting the catheters to be simple. The time required for the insertion of these catheters was less than five minutes. There was one complication with the subfascial catheter, when the surgeon transected the first catheter with a scalpel on insertion. It was noted immediately and the catheter parts were removed and replaced with a complete catheter, with no further problems. The participants reported no adverse effects or hindrances as a result of the elastomeric pumps. The nursing staff in the recovery holding area and wards required only basic education and training in the use of these pumps. They were highly satisfied with the ease of use of the elastomeric pumps and required minimal re-education. Overall, the nurses felt that the participants had received an increased standard of patient care, and that there was no extra nursing burden with regard to care of the elastomeric pumps and their application. Generally, the surgical teams were pleased with the level of patient postoperative pain control. Conclusion Continuous bupivacaine infusion into the incisional site increases options for postoperative pain relief. In this study, it was noted that participants who received the bupivacaine infusion experienced decreased pain intensity in the first hour postoperatively. They also experienced a decreased level of dynamic pain intensity in three of the four time periods, and decreased worst pain intensity up to six hours postoperatively. This pain relief option could provide better analgesia, although we did not show that it helps to reduce patients' opioid consumption.	2018-11-30T12:48:53.268Z	2011-04-01T00:00:00.000	{ "year": 2011, "sha1": "46daeed6727130a220d70a643500af3c4d1d2f94", "oa_license": null, "oa_url": "https://www.tandfonline.com/doi/pdf/10.1080/22201173.2011.10872786?needAccess=true", "oa_status": "GOLD", "pdf_src": "TaylorAndFrancis", "pdf_hash": "931bc300314011e401f7c42653687f2d51b8ca3a", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
237506671	pes2o/s2orc	v3-fos-license	Probing allosteric regulations with coevolution-driven molecular simulations Description INTRODUCTION The free-energy landscape of proteins drives the functioning and regulation of cellular processes (1)(2)(3). Underlying these highly controlled activities is the balance among different conformational states, which is often regulated by allosteric effectors-ligands that produce a structural change in the target in a region distant from their binding site (2,4). Despite notable advances (3,(5)(6)(7), the quantitative characterization and prediction of allosteric mechanisms continue to posit formidable challenges for both experiments and computations. The molecular underpinnings of allosteric changes in protein structures are often hindered by averaged metrics, the transient nature of the species involved (8), the difficulty to reproduce biological events in vitro (9), or the great deal of computational power required to model uncharted free-energy surfaces (9,10). Nonetheless, allosteric regulations are widespread in biology (11)(12)(13)(14)(15)(16)(17)(18), and the characterization of the underlying free energy is critical to understand and control the functional conformational landscape of biomolecules (19). The objective of this work is thus to devise a computational procedure to efficiently explore the functional freeenergy landscape of a protein with and without the allosteric effector bound, thereby gaining insight into how conformational equilibria translates into function. We focus on the conserved regulation of adenylyl cyclase (AC) by stimulatory and inhibitory G proteins-a prototypical example of protein-protein allostery at the center of the G protein-coupled receptor (GPCR) signaling cascade, the target of nearly half of all drugs. Despite extensive molecular research conducted on this system (20)(21)(22), there are still open questions. When hormones or drugs bind to GPCRs, they activate G proteins, which, in turn, activate or inhibit AC (20)(21)(22). ACs control the rate of conversion of adenosine 5′-triphosphate (ATP) into 3′,5′-cyclic adenosine monophosphate (cAMP)-the second messenger that, by interacting with protein kinase A, triggers the phosphorylation of a myriad of downstream targets (23). The most conserved regions of ACs are the pseudo-symmetric catalytic domains (called C 1 and C 2 ) that work as obligate dimers with the active site at the dimer interface, a feature rarely found in other enzymes (24). This unique feature makes the catalytic rate markedly affected by structural movements at the seam of the two domains (22,24). X-ray crystallography (25)(26)(27) and cryo-electron microscopy (cryo-EM) (22,28) experiments have detailed two conformational states in AC structure. They are coupled to the formation of the complex with stimulatory G s protein and include the reorientation of AC catalytic domains, together with a small displacement of an  helix opposite to the binding site of stimulatory G s (22,(24)(25)(26)(27). The high versatility of signal transduction encoded in AC structure (22,24,29) suggests, however, that the known states are only snapshots of a much wider and diverse set of functional ensembles that still have to be captured and quantified. In this context, computer simulations (30) could bring about a major productivity leap. Recently, insight has been generated from long molecular dynamics (MD) simulations, confirming the high structural plasticity of AC and providing a general blueprint of AC dynamics in a variety of biological contexts (31)(32)(33)(34)(35)(36)(37). Although extremely informative, these studies have offered only a qualitative description of the complicated link between effector binding and AC dynamics, raising challenging questions on the mechanisms and the free-energy landscape governing the functional ensemble of AC structures. To face these challenges, we devised a multiscale MD approach guided by coevolutionary data (38) that allowed the free-energy landscape of a fully solvated atomistic model of AC with and without the bound G protein to be reconstructed. Coevolutionary-based approaches have been used to predict native structures in proteins (39,40) and nucleic acids (41,42) and to generate conformational ensembles (43)(44)(45). Our work is unique in that it leverages coevolutionary information to reduce the complexity of the configurational space in protein-protein regulations, thereby focusing only on the functional free-energy landscape obtained from residue coevolution. By doing so, we expand and manipulate AC structural ensembles consistently with-and beyond-experiments. We focus on human AC type 5 (hAC5) as quintessential allosteric unit controlled by both stimulatory and inhibitory G proteins (20). We find AC populating two main conformational ensembles with all the existing experimental structures falling into just one of these ensembles. Notably, AC shifts from one ensemble to the other depending on which G protein it binds to. The results complement, and are in agreement with, structural (22) and biochemical (29,46) data, and extend our understanding of the mechanisms of protein-induced allostery in AC. Furthermore, they provide a general framework for investigating and manipulating complex biomolecular regulations. Coevolutionary analysis captures functional domain reorientations We performed direct coupling analysis (DCA; see Methods) (40,47) on a multiple sequence alignment of hAC5 homologs, selecting those coevolving pairs that are not in contact in the native structure, and which thus bear information on possible alternative, unknown, conformations (i.e., nontrivial pairs; Fig. 1A, red dots) (43). Nontrivial pairs appear between C 1 and C 2 catalytic domains (Fig. 1A, left), suggesting that interdomain movements are linked to function. DCA scores were filtered and introduced as ensemble restraints in coarse-grained discrete MD (dMD) simulations (43,(48)(49)(50)(51), which allowed us not only to detect functionally relevant conformations but also to generate trajectories connecting them (43). The trajectories resulted in hAC5 conformational transitions dominated by the opening and closing of the catalytic site as shown by the analysis of its principal components (Fig. 1B). The closure movement of C 1 and C 2 domain modulates the shape and accessibility of the catalytic site to ATP, providing a molecular glimpse on the major regulation mechanism of AC activity (22,(24)(25)(26)(27). Functional free-energy landscape from path metadynamics To probe the energetic feasibility of the coevolution-based dMD transition, we back mapped the coarse-grained transition into a fully atomistic one, characterizing its free-energy landscape with path-based (52) metadynamics (53,54) simulations in explicit solvent (see Methods and the Supplementary Materials). We observed that the apo (ligand-free) hAC5 populated two free-energy basins with a similar well-depth of 4 kcal/mol ( Fig. 2 and fig. S3). Such free-energy minima corresponded to a sparse "open" ensemble ( Fig. 2, A and B) and to a well-defined closed conformation of the catalytic domain and active site. Notably, the open ensemble included the experimental structures of activated AC bound to G s together with an ATP analog (P-site inhibitor) and the activator forskolin (1cjk) (26), a forskolin derivative alone (1azs) (25), and a guanosine 5′-triphosphate (GTP)-based substrate analog and forskolin (6r4o) from cryo-EM experiments (28) of the full-length membrane AC ( Fig. 2A, gray circles). As the above AC structures were experimentally solved in complex with the stimulatory G s , we label this open ensemble as the "active conformation"; the residues forming the active site are apart from each other and appear in an ATP-binding competent state. Although ATP was not present in our simulations, note that an ATP-bound conformation is a free-energy minimum encoded in AC intrinsic dynamics. Moving along the functional transition, we observed a metastable state (labeled 2 in Fig. 2) with reduced interdomain distance than the open ensemble. This intermediate conformation features a salt bridge between D518 and R1208 from C 1 and C 2 domain, respectively. D518 has been experimentally observed interacting with catalytic Mg 2+ and R1208 with the -phosphate of ATP analogs and is thus essential for ATP binding and catalysis (26). The desolvation of D518 and R1208 side chains likely contributes to the energy barrier for reaching this intermediate state from the open ensemble. Further progression along the functional reaction coordinate leads the system to a compacted closed conformation (labeled 3 in Fig. 2) with the interdomain distance dropping below 25 Å. Such a closed conformation is characterized by an additional bidentate salt bridge between D474 and E596 and the catalytic K1244. This interaction "seals" AC into a conformation with no accessible binding site for ATP (Fig. 2, A and B). As ATP binding is essential for cAMP production, this conformation is catalytically inactive. We remark that closed conformations of AC have not yet been observed experimentally, likely because the structure of AC could only be solved either in complex with the stimulatory G s (25,26,28,(55)(56)(57) or with the active site occupied by the activator forskolin (27). These results motivate and challenge the design of future experimental research. Overall, path metadynamics allowed the atomistic description of functional transitions in hAC5 and the efficient reconstruction of the underlying free-energy landscape with an estimated error in the order of k B T (Fig. 2C). Population shifts upon G protein binding To assess the effect of stimulatory and inhibitory G proteins on the regulation of hAC5 structural plasticity, we perturbed the free-energy landscape of hAC5 with the presence of either stimulatory G s or inhibitory G i (Fig. 3). The stimulatory G protein, G s , binds within a cleft in the C 2 domain of AC, while G i binds within the opposite cleft in the C 1 domain (Fig. 3) (29). When G s binds to hAC5, the conformational ensemble of AC is shifted to the open conformation that becomes ~6 kcal/mol more stable than the closed state (Fig. 3A, blue lines, and figs. S1 and S3). Notably, this open conformation belongs to the same ensemble comprising the experimental structures of AC complexed with stimulatory G s , thus validating the theoretical framework used here (see Fig. 2A for reference). In notable contrast, however, when we simulated hAC5 bound to the inhibitory G i , the open/closed equilibrium shifted toward the closed conformation of AC that becomes ~8 kcal/mol more stable than the open state (Fig. 3A, black lines, and figs. S2 and S3). The closure of the ATP-binding site is consistent with biochemical data indicating that P-site inhibitors (ATP analogs) bind with greatly reduced affinity to AC in the presence of G i (46). This behavior suggests a straightforward mechanism of AC regulation by G proteins. That is, when G s binds to hAC5 (Fig. 3B), the open ensemble of conformations is selected and the cyclase becomes competent to host ATP in its active site; vice versa, when G i binds to hAC5 (Fig. 3C, the binding of G i is symmetrically opposed to G s ), the ensemble of hAC5 conformations shifts toward a closed state that impedes or perturbs ATP binding, thus resulting in the inhibition of cAMP production. Multiple pathways of information transfer The structural dynamics of ACs systems was further investigated with community network analysis (see Methods) (58) to identify and compare major pathways of signal transduction and allosteric communication (Fig. 4). Analyzing the community network of the hAC5/G s system, a major communication path is found between the binding site of G s (community #1) and the community (#6) including the small 4 and 7 helices, which line the extension of the catalytic site in the C 1 domain (Fig. 4A). The communication is mediated by correlated motions in the  sheet (1, 4, and 5) connecting the two communities, as found in other globular proteins (59), and in the hAC5/G s system, they further mediate a weaker allosteric communication between communities 3 and 4, also lining the catalytic site. The dynamical network of the hAC5/G i system shows quite different features, as the communication from the binding site of inhibitory G i (community #3) propagates to the rest of the protein through the -helix 3 of C 1 domain (Fig. 4B). Particularly, G i allosterically modulates the binding site of G s (community #1), suggesting a mechanism to inhibit cAMP production by interfering with binding of the stimulatory G s , as it has been proposed previously (29,32,35). Therefore, the binding of G i not only selects the closed state of hAC5 catalytic domain but also triggers information transfer to the binding site of G s , likely modulating its capability to activate the cyclase. Furthermore, the residues participating in the salt bridges observed in the intermediate and closed state are involved in different communication pathways depending on which G protein AC binds to. In the hAC5/G s complex, residues D474 and E596 are the critical nodes connecting communities 6 and 4, respectively, while D518 is the critical node connecting communities 4 to 3. Residues K1244 and R1208 belong to community 5 (Fig. 4A, left). There is no direct communication between communities 5 and 6, 4, likely reflecting the observation that the free-energy landscape of the hAC5/G s complex is shifted toward the open state where no salt bridges are formed. In the hAC5/G i complex, residues K1244 and R1208 fall into communities 5 and 4, respectively (Fig. 4B, left). In this complex, we observe a direct communication between communities 3 and 4 that could mediate the formation of the salt bridge between D518 and R1208 featured in the intermediate state (Fig. 2B). Notably, community 5, including the catalytic K1244, appears to communicate mostly with community 4, suggesting a possible flow of information between intermediate and closed state. We further note that the asymmetry in the communication pathways activated when G s binds AC compared to G i binding could be related to the asymmetry in the relative (de)stabilization of open/closed states by the two G proteins (with G s stabilizing by ~6 kcal/mol the open conformation, while G i stabilizing by ~8 kcal/mol the closed one). In particular, a qualitative comparison of the networks shown in Fig. 4 (A and B) suggests that the binding of G i to AC triggers a much wider flow of information with high correlation (or energy of interaction) (58) between nodes than the binding of G s to AC. Last, the large number of highly conserved residues at the interface between communities (critical nodes) suggests that the response to external stimuli (e.g., to inhibitory G i ) among different AC isoforms can be achieved by 1-6 and 3-4). The modulation of hAC5 dynamics by G i is achieved through a larger network reaching the C 1 domain mainly passing through the helix 3 (intercommunity connection 3-2). As highlighted by the community network, the binding of one G subunit can produce allosteric effects at the binding site for the other, with the binding of G i having the major allosteric modulation on the dynamics of G s binding site. sequence changes as well as by specific binding to effectors that, in turn, modulate intracommunity dynamics (60). By combining recent breakthroughs in coevolutionary analysis with multiscale modeling and free-energy methods, we have dissected the allosteric regulation of AC intrinsic dynamics by G proteins-a fundamental process in biology and pharmacology. Our model reveals quantitative details on the regulation mechanism and demonstrates that signal transduction in AC/G protein systems operates through the selective (de)stabilization of the particular state to which ATP preferentially binds following the classical "Monod-Wyman-Changeux" population-shift paradigm (2, 61). These results create a common reading frame among multiple lines of experimental data and provide an unprecedented spatiotemporal resolution on the molecular mechanisms regulating cAMP generation. A possible limitation of the present work is that coevolutionary information was encoded only in the dynamics of AC, thus assuming that the activated GTP-bound state of G proteins fluctuates around the observed crystallographic structure (22). Overall, the general procedure outlined here can be used to probe other allosteric regulations in uncharted conformational space for a wide range of complex systems. Advances in genomic sequencing make the procedure applicable to thousands (43) of macromolecules for which functionally relevant transitions can be efficiently perturbed. Molecular design strategies can be combined with the approach presented here to interrogate desirable functional intermediates, change the kinetics, or reprogram the mechanism of these systems with far-reaching consequences for pharmaceutical and biotechnological applications. 3D structural models The catalytic domains of hAC5 (UniProt: O95622) were built by homology modeling using SWISS-MODEL (62), based on up to 98% sequence identity with the crystal structure of the mammalian hybrid AC5/AC2 in complex with G s [Protein Data Bank (PDB) ID: 1AZS]; GTP-activated human G s (UniProt: P63092) in complex with hAC5 was built from the same PDB template. The crystal structure of GTP-activated human G i (UniProt: P63096; 2GTP) was used to generate the complex with hAC5 that resulted in G i binding the C 1 domain in an orientation symmetrically opposed to G s with respect to AC5 in the 1AZS complex, as suggested by mutagenesis data (29). Myristoylation of G i was modeled following the procedure described in (63). Coevolutionary analysis and coarse-grained simulations The pairwise coevolved amino acid positions correlate strongly with spatial proximity in the three-dimensional (3D) space (38)(39)(40)47) and can be used to fuel the search for biologically relevant conformational ensembles (43)(44)(45) and to identify functionally relevant transitions in proteins (43,44) and nucleic acids (41,42). Multiple sequence alignment (1261 hits) from the clustered UniProt database (uniclust30_2018_08) (64) was generated using hAC5 (UniProt code: O95622) as query with HHblits (65), and DCA (47) was used to measure residue-residue coevolution with default parameters. DCA outputs a direct information (DI) score per pair of residues that was then filtered and used as input for the coarse-grained dMD simulations as described by Sfriso et al. (43). Briefly, given a DI-ranked list of coevolution pairs, we keep for further analysis only the first n pairs (n = 1000 in the case of AC) that maximize the Matthew's correlation coefficient resulting from the prediction of contacts (<10 Å) in the initial structure. Intuitively, in this step, we extend to the number of DCA contacts that are still informative about the initial structure. For the n pair selected, we test the accessibility of each residue pair in the structure by means of dMD pulling simulations where DCA pairs are brought close in space (one independent dMD simulation for each pair). By doing so, we filter the DCA output to remove uninformative or impossible contact pairs; individual trajectories are accepted only if they show better coincidence with coevolution information than a given threshold. For this, we check whether coevolutionary contacts are spontaneously established along the pulling trajectory and compute receiver operating characteristic (ROC) curves to quantify the agreement between the conformations generated and the list of n coevolution pairs. The area under the resulting ROC curve (AUC) provides a means to compare and rank the coherence between trajectories and the coevolutionary fingerprint. We retain instances exceeding 1.5 of the interquartile range in the AUC distribution (16 generated models in the case of AC), and the corresponding pairs of residues are incorporated as weighed (43) square potentials (904 wells were added to the native 9631 wells in the case of AC) into a multiple structure-based model (SBM). Coevolution pairs are thus reflected in the multiple SBM by favorable energy interactions that guide an ad hoc sampling strategy (43) in the exploration of the conformational landscape with dMD. Implicitly, this approach filters noise in the DCA signal and reveals the protein ensemble encoded by coevolution. The dMD simulations are clustered and analyzed to provide a trajectory that is representative of the functional conformational landscape. Last, the trajectory was discretized in 80 equidistant frames (66), which were sufficient to describe the functional transition with the necessary resolution for metadynamics. We note that our approach is similar to the one proposed by Morcos et al. (44), with the main difference being the filtering step of the DCA contacts. While Morcos et al. (44) directly incorporate top-ranked DCA pairs as energy minima of the SBM, our protocol includes the filtering of coevolution pairs with the dMD pulling simulations mentioned above. We have shown (43) that filtering of DCA contacts is not critical when abundant sequences (>10,000) are available, yielding strong evolutionary signal. However, when fewer homologs are aligned (in the order of few thousands), conformational transitions could not be modeled without filtering DCA contacts due to noise in the coevolution map. In our experience, few high-quality coevolved pairs are thus necessary to robustly guide protein dynamics, making the detection of these constraints decisive. System setup and MD simulations Each system was solvated with a 1-nm-thick truncated octahedron box of TIP3P (67) water molecules with periodic boundary conditions. Equations of motion were integrated with a time step of 2 fs. For all nonbonded interactions, the real space cutoff was set to 1.0 nm, and the electrostatic long-range interactions were treated using the default particle-mesh Ewald settings (68). Bonds involving hydrogen atoms were constrained using the LINCS algorithm (69). After minimization and thermalization in the canonical ensemble (NVT), each system was further equilibrated at constant pressure and temperature (1 atm, 298 K) for 100 ns; metadynamics production runs were then generated in the isothermal-isobaric ensemble (NPT) using the stochastic velocity rescaling thermostat (70) and the Parrinello-Rahman barostat (71). The active site of G proteins was modeled with one crystallographic Mg 2+ ion and a GTP molecule. N-terminal myristoylation was present in G i and modeled following the procedure described in (63). AMBER parameters for Mg 2+ (72), GTP (73), and myristoyl group (63) were used. All simulations were run using GROMACS 2018.3 (74) and PLUMED 2.4.3 (75) with the Amber ff99SB-ILDN force field (76). Atomistic metadynamics simulations Well-tempered metadynamics (77) was used to reconstruct the underlying free energy as a function of two collective variables (CVs) or reaction coordinates based on the functional path generated from the coevolutionary-driven coarse-grained simulations. In this framework, the microscopic coordinates of the system, q, are mapped in the CV space by s(q), which measures the progress along the functional path (52), and z(q), which measures the distance from the preassigned path. Using these variables, one can explore the free-energy landscape between an initial and final state and can find low free-energy pathways connecting them-pathways that, in turn, can be different from the originally assigned one (52). Proper reweighting (78) allowed the resulting free-energy landscape to be projected on the interdomain distance and s(q) (see also figs. S1 to S3). We ran 2.3 s of well-tempered metadynamics simulations for each system. Community network analysis Allosteric network in the AC systems was examined through community network analysis (58) and visualized (79) with VMD (80). Briefly, the metadynamics trajectory of the AC systems is used to group C carbons (nodes) into communities-a set of nodes that move in concert with each other. Edges between pair of nodes are drawn if the average internode distance is below a certain threshold (5 Å); edge distances between node i and j have correlation-based weights, w ij = −log(\|C ij \|), which define the probability of information transfer across a given edge. Major communication pathways are identified by the edge betweenness, the number of shortest paths that cross a given edge (58). SUPPLEMENTARY MATERIALS Supplementary material for this article is available at https://science.org/doi/10.1126/ sciadv.abj0786 View/request a protocol for this paper from Bio-protocol.	2021-09-15T06:17:58.569Z	2021-09-01T00:00:00.000	{ "year": 2021, "sha1": "3be3f6bcb3a394ba35cc9dbee7aaebcf9857a930", "oa_license": "CCBYNC", "oa_url": "https://www.science.org/doi/pdf/10.1126/sciadv.abj0786?download=true", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "ac53c62f5afd1582fb01f85ad1d1d50c15ee0402", "s2fieldsofstudy": [ "Chemistry", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
225746518	pes2o/s2orc	v3-fos-license	The key factors affecting the polish investment attractiveness for foreign investors The article researches the role of formation of favorable investment climate in increasing investment activity in the territory of Poland. It focuses on the creation of favorable macroeconomic conditions for attracting direct (real) foreign investments to implement innovative projects. As a result of the study of investment conditions on the basis of PEST-analysis, the key factors of IA development were identified and their possible future impact (both positive and negative) was assessed. The results of the PEST-analysis allow their implementation in the program development and also in the investment activity management mechanisms Introduction The main condition for increasing the level of national competitiveness and increasing the pace of innovation-oriented development of the Polish economy is to attract new money.Under the conditions of crisis phenomena in the global and national economic system and the inability of domestic investors and the state to fully invest in the modernization of production as well as infrastructure potential, it is relevant and important to attract foreign investments. Of course, in their decisions on capital investment foreign investors focus on the indicators of the investment climate in Poland.This means that the efficiency of investment activities in general directly depends on the state policy and the mechanism of attracting foreign investment capital.They are the key factors in creating a favorable investment climate, which signals the governmentʼs support for investment initiatives. Many scientists devote their research to the problems of regulation and attraction of foreign A VUZF University, Sofia, Bulgaria, master, doctorant VUZF, e-mail: monika_artman@o2.pl,ORCID: 0000-0001-7163-982X B State University of Applied Sciences in Konin, Poland, Dr of Economic Sciences, Professor, e-mail: t.a.cherniavska@gmail.com,ORCID: 0000-0002-4729-2157 investments in general, as well as in the economy of the European Union countries and Poland in particular.Thus, in the paper Major and Szilágyi (2009) researches the importance of investing in the countryʼs infrastructure in order to boost investment activity.Famous Ukrainian scientist Amosha (2011) calls for concentration of attention in the strategy of investment development on creation of effective institutional environment, which will contribute to increase of investment attractiveness.The national investment climate depends to a large extent on how transparent and understandable the procedures for registration and operation of investors are.The article speaks about prioritizing and state support for innovative investment projects. In our article we focus on creating enabling macroeconomic conditions for attracting direct (real) foreign investments into the Polish economy.We have aroused interest in in-depth analysis of exogenous and endogenous factors for the intensification of investment activity in the Nakije Kida article.Scientists have paid great attention to the study of models that explain the role of foreign direct investment in GDP growth as well as the activation of scientific and technological development.The author reveals the causal links between the final result (foreign investment inflow) and key factors (Kida, 2016).We share the scientific position and believe that sound macroeconomic policy will create a more effective environment for economic activity and promote economic growth. Material and Method One of the popular methods in assessing the investment climate of the country, namely to identify strengths and weaknesses, threats and opportunities is SWOT-analysis.In particular, a group of authors consisting of Gavatiuk, Sokrovolska, Korbutiak & Karvatskyi (2020), this technique was used to identify key factors, threats and opportunities for the development of investment activities in Ukraine.We offer PEST (STEP) analysis as a structural component of SWOT analysis for the general parametric identification of external and internal factors of investment activity in Poland.We were especially inspired to use this method by Matyrkoʼs (2016) scientific paper "Basic Approach for Evaluation of Foreign Marketʼs Potential and Attractiveness", which describes in detail the possibilities of its application. We should note that the scientific literature uses several variants of the name of this method, depending on the number of factors under study.Thus, the name PESTLE includes an expanded number of factors means an abbreviation of the key six categories of macroeconomic variables analyzed in the model political, economic, socio-cultural, technological, legal and environmental (Del Marmol, 2015).It should also be noted that PESTLE can be used as a tool to investigate micro-environment factors, for example, when setting up a new business or to test strategic maneuvers (Allen, 2001).In our opinion, PESTanalysis can be considered both as an independent and as a basic methodological method for conducting an analytical study of the investment climate and attractiveness of a particular country for foreign investors. Originally developed in 1967 by Harvard professor Francis Aguilar, PEST analysis is a strategic planning tool that helps organizations identify and evaluate threats and opportunities for the business (Aguilar, 1967).PEST is an acronym describing four primary external factors that influence the business environment: Political, Economic, Socio-cultural, and Technological. The entire PEST analysis process is guided by the following steps: 1) Identify the factors that may influence the implementation of the companyʼs plans.The easiest way to select a limited number of the most important factors from the proposed list of factors in each category and use them for further analysis; 2) Gather information about the dynamics and nature of each factor change.This is the main part of the work.Based on these data, the experts will form their opinion; 4) To analyze the significance and degree of influence of each factor.Develop a structured questionnaire for the experts, make estimates of the significance of the external environment factor and the probability of its change. Based on the results of the expert survey, it is necessary to: calculate the average value of the significance indicator for each factor; estimate the weight of the factor relative to the sum of significance indicators of all factors of all categories; estimate the average probability of an event (a synthetic indicator of the assessment of the influence and probability of a factor is the probability of an event multiplied by the weight of the factorʼs significance); optionally, it is possible to estimate the quality of the factorʼs influence on the results: positive (+) or negative (-); make summary tables of the PEST analysis. The summary table of the PEST analysis looks as follows (Table 1).The result of the PEST analysis is a tabular model or matrix, which reflects significant factors and for each of them an estimate of the force of influence in points and the probability of change is given.The factors are assigned ranks (Table 2).A very convenient tool for visualizing the results of the PEST analysis is the radar chart, which clearly shows the most important factors for each category of PEST analysis. As a strategic analysis tool, PEST allows for competent prioritization and identification of potential for improvement of the current situation (in our case -the state of investment activity of foreign investors in Poland -authorʼs note). With it you can make long-term strategic plans with annual data updates.A more effective result of STEP analysis can be obtained with regular use with time fixing to track the dynamics of changes in indicators as well as their impact, which in turn will accurately assess the external environment (Shtal, 2018). The purpose of the PEST analysis proposed in this paper is to develop an in-depth understanding on the context (e.g. a country) that is the original target of the study and subsequently identify a narrower context (e.g. a IA of foreign investors) in which the study can generate more in-depth and meaningful findings.However, PEST is far from being a precise and clearly circumscribed analysis framework.There are an almost unlimited number of variables that may emerge from each dimension.Therefore there is the need to prioritise those variables that have highest impact on the industry, sector, or country being studied (Peng, Nunes, 2007). The analytical method enables to evaluate the key trends in investment activity in Poland within the framework of this study.More importantly, it can be considered both as an independent and basic methodological technique for conducting analytical research according to the SWOTanalysis method. The fact that advantage of PEST analysis is not merely in identification of a complex of significant environmental factors, but in determination and analysis of their impact on the investment activity in the economic territory of Poland as well, should be noted.A tabular model reflecting and grouping all identified factors of a significant impact on the IA is the result of the Study.An evaluation of the effect and expectancy of changes in investment activity in the country will be presented for each of the selected complexes. The whole process of conducting PEST-analysis of the investment activity in Poland can be reduced to the following proceedings: identification of the most significant factors that may affect investment activity on the economic territory of the country; accumulation and aggregation of information on the dynamics and nature of changes in each of the complex of factors.That is, the optimal necessary maximum of information on each of the significant aspect of the IA is collected so that the experts can form their own opinion, which is necessary for implementation of the next stage in the analysis algorithm; evaluation of significance and analysis of the impact degree of each of the IA factors.Based on the data of the expertsʼ assessments, the average value of each factor is calculated and the significance indicator is determined, and then each factor significance is estimated relative to the significance of the indicators of all factors in the context of all categories; calculation of the synthetic index for evaluation the impact of a particular factor on the IA.It should be noted that the qualitative impact of a particular sampling frame factor can be estimated as positive (+) or negative (-); compilation of a tabulated summary of PESTanalysis (Table 3.).-national legislation and freedom of expression; -level of bureaucracy and corruption; -trends towards regulation or deregulation; -legislation in the sphere of labor and social protection of the population; -projected political changes for the next 3-5 years E (ECONOMICAL) -degree of development of the business structure; -rate of economic growth, interest rate, exchange rate and inflation rate; -degree of the economy globalization; -degree of development of banking, credit and tax systems; -labor market situation (labor cost, unemployment rate, etc.); -wage rates of the population; -projected economic changes for the next 3-5 years S (SOCIO-CULTURAL) -total population, its gender and age structure; -healthy lifestyle status; -educational level as well as social mobility; -public opinion, standards of conduct and restrictions; -wage rates and wealth divide; -most probable social changes for the next 3-5 years T (TECHNOLOGICAL) -level of technologies and equipment penetration into the sectors of national economy; -performance of scientific and technical, design and implementation activities; -degree of introduction of innovative technologies; -level of the State support for technological development; -most predicted changes in the technological sphere for 3-5 years Source: developed by the author based on Laptev (2017), Kocheshkov (2020). Results and discussion The Republic of Poland is one of the most favorable countries for the investment of foreign capital.It is significant that according to the definitive opinion of international experts, the Polish economic system was recognized as safe for organization and conduct of business; all necessary prerequisites for investment, including modern infrastructure support, have been created on its territory. Poland is a good place to invest -said foreign businesses in the "Investment climate in Poland" survey, conducted by Grant Thornton, PAIH and HSBC.As many as 65% of the respondents find the conditions for doing business in our country to be at least good.The businesses surveyed are happy with the investments they made in Poland.An impressive 94% of the respondents expressed satisfaction with having chosen Poland as the place to invest and declared they would make the same choice again.On a scale from 1 to 5 (with 5 being the highest mark), Poland was rated at 3.71 pts as a place for doing business, which is one of the highest scores in the 12-year history of the survey (Climate for Business, 2019). Poland has become a kind of "investment magnet", an attractive spot for investing foreign capital, where profits, according to the experts, can be up to 3 times more than in other EU countries, in many respects due to the comprehensive support of State bodies, as well as a wide range of investment incentives (Invest in Poland, 2015). Environment for doing business has been constantly improved by the State, and the level of national competitiveness has been growing over past decade.Thus, Poland was ranked the 24th in the world ranking of 190 countries in the nomination of Ease of Doing Business according to the latest report "Doing Business, 2017" published by the World Bank Group.The report noted that it is among the best manufactures in Europe and Central Asia, as well as occupies a leading position among the member States of the European Union and in Central Europe (Handbook of Investors, 2016). An analysis of the main positive factors of the investment attractiveness is presented in generalized Table 4.According to Doing Business 2010-2017 World Bank reports Poland is a leader in the nomination of Ease of Doing Business.The ranking compares the extent to which a given country has reduced the distance to the "border", which represents the best performance in terms of doing business achieved by the economy.For the period from 2010 to 2017, Poland reduced the distance by 14.67% overtaking such countries as Romania, Croatia, Slovenia and the Czech Republic Cost-effective industries for capital investment Tourism.Poland is one of the Central European countries most visited by the tourists.Development of investment activity in the tourism industry is inextricable with such factors as economic and geographical (location in the center of Europe); physicogeographical (healing mountain climate of local mountains and lakes in the North of the country); socio-economic (development of entire areas of tourism, combining high comfort and low cost recreation); historical-cultural (preservation of the native architecture and cultural values determine the attractiveness for capital investments); geopolitical (economic and political stability allows to keep the tourists traffic).Business services sector.Poland has already won the status of the European Center for Business Support Systems having involved about 150 thousand qualified personnel in its operation.High qualification and experience of the employees, multilingual business environment, availability of cultural compatibility and basic competitive costs make it an excellent place for localization of international business support service centers and development of innovative types of business services.Research and development sector.Progressive development of 40, as well as successful launch of 14 more Techno park zonal entities on the territory of Poland, recognition of Poland as one of the most attractive locations for foundation and launch of startups by Forbes Magazine mainstream the perspective trend for investment into the national economy. Motor industry.Attraction of potential investors into the industry is primarily based on availability of cost-effective and enhancing labor force, as well as on territorial proximity to the manufacturing facilities of complete equipment.What is more, capacity of the Polish automotive market as one of the largest in the EU serves as a positive factor.Along with that, Poland is a manufacturer of a wide assortment of automotive parts and engines; Fiat, GM and VW plants are located on its territory.Aerospace industry.Poland is traditionally considered as one of the best locations to implement projects in aerospace and aviation industries.All kinds of activities and production, including aircraft construction and manufacturing of spare parts, maintenance and repair, as well as scientific research and training, are developed here.Long experience of cooperation with such global companies as Airbus, Boeing and Embraer, as well as a well-deserved positive reputation as an exporter of products, are an indisputable indicator of the investment attractiveness of this sector of the economy.Electronics.Polish electronic industry is currently considered one of the most significant in the EU.The advantageous strategic location of Poland, successful implementation of investment projects in almost all segments of electronics (manufacturing of household appliances, computers, telecommunications, audio and video components, etc.), as well as creative staff and strong base for education and training, contribute to the competitive choice for investments into the macro-region.Food and processing industry.Poland justifiably occupies a leading position as one of the largest producers of agricultural, food and processing products in the EU and in the world.Products grown and produced on Polish soil has won the status of safe, tasty and healthy food, as well as high praise from domestic and foreign consumers.Priority positions in the structure of the national economy and comprehensive State support for the development of the industry contribute to its investment attractiveness. The fact that determination of the degree of impact from a sampling frame factor is a subjective expert evaluation should be noted.Results of a survey of the leading experts from the Polish Information and Foreign Investment Agency (PAIiIZ) were used for this purpose. Notably, despite the significant advances of Artificial Intelligence (AI) techniques, PESTLE analysis is currently performed manually.Usually, a large number of human experts, with different background and expertise, have to collect, select, and analyse large amount of information in order to suggest the best course of action to perform in response to the enquiry at hand (Vallati, Grassi, 2019). The impact of each of the selected significant factors was evaluated on a scale from 1 to 3, where: 1 point was assigned in case if the factor impact was estimated as insignificant, that is, any change in the factor scarcely affects the investment activity; 2 points -in case if the factor impact was of a moderate effect: only a significant change in the factor affects investment activity as well as any activity in Poland; 3 points -in case if a certain factor impact was critical (strong), that is, any fluctuation results in significant changes in regard to investment activity on the economic territory of the country.Factor Description Factor Impact Political factors 1. Frequent changes in Polish legislation 3 2. Lack of a dialog-oriented platform between investors and the Government 2 Economic factors 3. Reduced investment opportunities due to the reduction in EU funding 3 4. Dependence on external economic environment and political situation 2 5. Increased risks and threats to economic security due to the "liberal" supervision over investorsʼ activity.3 Social factors 6. Outflow of working population 1 Technological factors 7. Introduction of innovative technologies into the national infrastructure 1 Source: compiled by the author based on the analysis of expert reviews Further, expectancy of change in each of the selected sampling frame factors is estimated on a scale from 1 to 5 (Table 5.).Thus, according to the evaluation scale, 1 point means the minimum expectancy of changes in a particular factor of IA, and 5 points is the maximum (or 100%) expectancy of changes in investment activity. Degree of expectancy of changes in IA factors was further evaluated as follows: 1 point was assigned if the expectancy evaluated by the expert ranged from 0 to 20%; 2 points -if the expert evaluation of the changes expectancy ranged from 20 to 40%; 3 points -if the variance of the factor changes expectancy was estimated by the expert in the range from 40 to 60%; 4 points -if the expectancy score ranged from 60 to 80%; 5 points -if the estimated expectancy was equal to the maximum (strong) and the expert opinion ranged from 80 to 100%. Each of the five experts provided their evaluation of a particular sampling factor.Source: compiled by the author based on the analysis of expert reviews Further, actual significance of each sample factor adjusted for the weight was estimated (Table 7.).The indicator is calculated as the expectancy of change in a certain sampling frame factor weighted by the degree of its impact on the investment activity in an aggregate amount of impact of the sampling frame.The fact that the higher the significance of the factor, the more attention and efforts should be paid to level its negative impact on the investment activity should be emphasized.The given aspect acquires even more relevance in the context of this Study, since it ensures the use of a complex of measures to prevent risks and threats to the economic security.Apart from that, the given result of PEST-analysis allows to concentrate attention and strengthen managerial response of the regulatory bodies of IA on formation of mechanism for control and counteraction to negative factors, in other words, management of risks and threats from changes in the macro-environment of the investment activity.In detailing IA managerial process, subsequent appointment of the authorized body (authorized persons) responsible for monitoring fluctuations in the factor within the threshold values, as well as the corresponding further action plan, is quite necessary as well. The next stage of the analysis is presentation of the calculations results in a matrix form (Table 8.).As can be seen from the presented matrix, economic and political factors are the most important in the development of investment activity on the economic territory of Poland.In particular, the following positions are of the highest importance in the ranked list: reduced investment opportunities due to the reduction in EU funding (1.00); increased risks and threats to economic security due to the "liberal" supervision over investorsʼ activity (0.84); frequent changes in legislation (0.8); lack of dialogue between the Government and investors (0.37).Analysis of such a factor as reduction in funding under the Cohesion Policy program for the period of 2021-2027 is of particular note.Put that in context, the European Commission is planning reduction in the funding budget for Central and Eastern Europe, in particular Poland and Hungary, by EUR 37 billion in favor of allocating these funds to Greece, Spain and Italy in the period ahead.Thus, according to the new budget, the Republic of Poland will be allocated 23% or EUR 19.5 billion less funding as compared to the 2014-2020 budget, according to which EUR 83.9 billion was allocated. Based on the presented matrix, inferences of the most significant factors affecting investment activity in Poland aimed on focusing managerial impact on leveling their negative impact can be deduced. Thus, the undoubted advantage of the analysis is not only in identification of key factors in the development of IA and evaluation of their potential future impact (both positive and negative -authorʼs note), but in feasibility of developing programs, as well as mechanisms for managing different situations in the investment sphere.Despite the fact that the development of programs and formation of the efficient managerial mechanisms can take quite a long period, along with this, it is worth to agree that the change in the identified factors will no longer be sudden for the Governing bodies.And this means faster reaction and greater flexibility, as well as improvement in the efficiency of investment activity through the use of this strategic analysis tool.As it was noted earlier, results of PEST-analysis can serve as the basis for SWOT-analysis in terms of searching for prospects and resources for the development of IA. Conclusions In our opinion, a comprehensive study of the environment of investment activity on the basis of PEST and SWOT analysis methods allows to form a practical mechanism for the State management of the efficient investment development on the economic territory of Poland. Table 1 - PEST analysis chart Table 2 - Quadrilateral matrix of PEST analysis Table 3 - Matrix for Identification of External Factors of IA According to PEST Analysis Method Table 4 - Analysis of Positive Factors Effecting the Investment Attractiveness of Poland Poland at the very heart of Europe makes it an ideal place for investment into the activities of economic entities oriented both to the Eastern and Western parts of the Continent Political stability based on EU and NATO membership As a member of the EU and NATO the Republic of Poland has a long history of political stability and adheres to the free trade principles Considerable domestic market Poland is one of the largest EU member states.It is the sixth most populous country in the European Union and is the largest market in Eastern and Central Europe Stable economic development Poland's economy is one of the fastest growing economies in Europe.It is the only state in the EU that managed to withstand the 2009 recession.Population Poland is one of the largest consumer markets amounting to about 38 million people. Poland is the largest recipient of grants from the European Union budget.Thus, allocation of the budget funds in the amount of EUR 82.5 billion in the framework of implementation of the consolidation policy and EUR 32.09 billion for the development of agricultural industry has been planned for the period from 2014 to 2020.The funding will be allocated to such areas as research and its implementation, roads construction, Table 5 - Identification of Factors of IA and Evaluation of their Impact* Table 6 - Evaluation of Expectancy of Changes in IA Factors* Table 7 - Estimation of Impact of IA Factors Taking into Consideration Expectancy of their Changes* Table 8 - Matrix of Significant Factors for the Development of IA in Poland * *Source: compiled by the author based on the analysis of expert reviews	2020-07-02T10:25:42.316Z	2020-06-29T00:00:00.000	{ "year": 2020, "sha1": "fd0dbea0687a57d662df50d9e6c5bb95ac13eda0", "oa_license": "CCBY", "oa_url": "https://doi.org/10.38188/2534-9228.20.2.02", "oa_status": "CLOSED", "pdf_src": "ScienceParsePlus", "pdf_hash": "4ed278111b63898a593cc66f096b858dbc31d58d", "s2fieldsofstudy": [ "Economics", "Business" ], "extfieldsofstudy": [ "Business" ] }
263645668	pes2o/s2orc	v3-fos-license	On the Acoustic Radiation Force Affecting Two Liquid Drops Located in the Wave Field : A system of initially immobile spherical liquid drops placed in another liquid in the ﬁeld of a propagating acoustic wave is considered. The acoustic radiation force acting on each spherical liquid drop is determined as the function of the incident wave frequency, making use of the solution of a problem of plane harmonic wave scattering on a system of two spherical bodies. The problem is solved by the variable separation method. To satisfy the boundary conditions on spherical surfaces, the expansion of the incident and reﬂected wave potentials over the spherical wave functions are used. Required constants in the solution are calculated from an inﬁnite system of the algebraic equations, which is solved by a truncation method. It is established that the value of the acoustic radiation force affecting each liquid drop depends signiﬁcantly on the densities, speed of sound in the outer and internal liquid, as well as on the distance between drops. It is also found that the acoustic radiation force has the same or opposite direction as the incident sound wave depending on its frequency. As result, at different frequencies, the liquid drops can start moving towards or further away from each other. Introduction A harmonic wave propagating in the acoustic medium causes several remarkable effects.The periodic acoustic pressure and constant component of sound radiation are among them.The latter one is usually referred to as acoustic radiation pressure and is averaged over the incident wave period and over the surface of an object placed in the medium results in the acoustic radiation force (ARF) (see, for example [1][2][3]).From a mechanical point of view, the ARF is generated if an average impulse transferred by the wave over the correspondent period of oscillation varies in some volume of medium due to the presence of a foreign body located there. The phenomenon mentioned is of great interest for technological applications either already developed or potentially valuable.This has resulted in a large number of studies in the field (see, for instance [3][4][5]).It is worth mentioning here that the researchers applied several techniques to estimate the ARF.The range of methods comprise the exact analytical solution and its approximations (see, for instance [3,6,7]) as well as a vast variety of numerical simulations (see [8][9][10]). Various aspects of the ARF phenomenon were addressed in the scientific works over the decades: type of wave, properties of bearing medium, geometry and properties of submerged body, and the boundedness of the region filled with the acoustic medium, to name a few.For instance, generation of the acoustic radiation force by the plane traveling wave, Axioms 2023, 12, 940 2 of 14 plane standing wave, double orthogonal standing waves, acoustic beams with arbitrary wavefronts, and Bessel helicoidal beam were studied in several publications [5,7,[9][10][11].Due to practical applications, the ideal and viscous fluids were usually considered as the acoustic bearing medium [4,5].Obviously, the complex geometry of the submerged body (solid, liquid, or gaseous), complicated functions describing viscous properties, effects of acoustic streaming, etc., demand the application of sophisticated numerical techniques to attack the problem (see [8][9][10]12] for details).Nevertheless, usage of analytical approaches has significant advantage when investigating cylindrical and spherical bodies located in the infinite medium and has proved to be applicable to study ARF even for the bodies placed in the vicinity of a flat wall or free boundary [6]. As well as being of theoretical interest, the effect of the acoustic radiation force acting upon the body in the acoustic field finds its industrially valuable application in various technologies [1,2,13].Remote manipulation and control (or suspension) of the rigid particle, drop of liquid, or bubble of gas placed in the acoustic medium, non-crucible glass fusing, multi-components casting, preserving the cloud of the object at the targeted location and acoustic levitation, facilitation of coagulation and degassing processes of the liquids, ultrasonic cleaning, as well as crystal growing, etc., are among both well-established and prospective fields of ARF application [13]. Another level of complexity for the problem under consideration is introduced by the oscillatory and wave processes that occur in the liquid containing foreign solid, liquid, or gaseous inclusions [14,15].The interactions of particles, drops of liquid, and gas bubbles in the acoustic field, acoustic levitation, numerous cavitation effects in fuel pipelines and various propulsion systems, as well as bubble generation in a liquid by fast moving solids, are among them.The special significance of the ARF study is obvious for the medicine industry and health care.Being injected into the circulatory system of the living organisms, the microscopic solid particles, liquid drops, or gaseous bubbles provide great opportunities for morphology investigation and for visualization of a variety of processes in the organs.This approach can also be used for drug delivery applications to ensure the prescribed drug doses with specific localization at the targeted organ [16]. As mentioned above, the ARF affecting an object submerged in a liquid is determined as a time-averaged value of the acoustic pressure over the object surface.Thus, the Lagrange reference frame is more suitable to derive the problem statement for this case.To apply this approach, the acoustic pressure in the vicinity of the object considered has to be calculated with accuracy up to the second order of magnitude.It means the deviation of the acoustic pressure from the harmonic law has to be accounted for in the region around the object [1][2][3].It is obvious that the linear approximation of the pressure results in the zero value of the ARF.It is caused by the fact the pressure is a periodic function of time, and being averaged over the wave period yields zero.For details, see [17], for example.Therefore, the second order terms have to be preserved in the wave equation to provide a reliable estimation of the ARF.In other words, the second order approximation has to be used.Fortunately, as was shown by L.V.King in his pioneering work [3], to address the issue, the velocity field potential derived as the solution of a linear problem of an incident wave scattering on an object placed in an acoustic medium can be made use of.The remarkable result of [3] allows one to evaluate the acoustic pressure with accuracy to the quadratic terms, providing that the velocity field potential obtained from a linear diffraction problem is used in the present paper as well. The acoustic radiation force can be used as a quantity to estimate the level of interaction between the wave and the object placed in the acoustic field.As mentioned above, to characterize this interaction qualitatively, the sound pressure averaged over the field oscillation period and over the object surface has to be evaluated.Theoretical examination of the ARF effect on the system of solid particles, liquid drops, or gaseous bubble is a complicated problem [18,19].It is necessary to account for a set of instances to attack it properly: the shape of the object, its nature (solid, liquid, or gaseous), its size with regard to the wave length, the properties of a bearing liquid (or gaseous) medium, wave scattering on the other particles, and their mutual effect (if system of solids, drops, or bubbles are under investigation), etc.Some of these aspects were already addressed in the publications devoted to the estimation on ARF for a case of solitary rigid, liquid, or gaseous particles located in an infinite liquid space or in the vicinity of the plane boundary of a liquid (see papers [5,6,16]). In the present paper, the case of two spherical liquid drops placed in an ideal compressible liquid is under investigation.The plane acoustic wave propagates along the line passing through the centers of the spherical drops.The purpose of the work is to study the interaction regularities between liquid spherical particles and to estimate the ARF induced by the acoustic wave irradiation of the system of the drops.The technique of the problem solution is elaborated.It comprises two basic steps according to the version of approach developed in the paper by Guz and Zhuk [5].At the first step, solution to the problem of the incident wave scattering on the system of two liquid particles (drops) is addressed.Once the problem is solved and the velocity field potential is derived, the main features of the two-drop system and acoustic wave can be studied.At the second step, the hydrodynamic forces experienced by each of the particles are calculated.Then, the ARFs are obtained by the time-averaging over the period of the primary wave oscillation.The results of the ARF case studies for the variety of liquid media and spheres' liquid properties, as well as for different geometrical parameters of the system are presented.Conclusions on the main regularities for ARF affecting both liquid spherical particles are drawn. The Problem of the Velocity Potential Determination To clarify the problem statement even further, it is worth mentioning here that the velocity of the translational movement of each spherical particle is supposed to be negligibly small in comparison to the velocity of the incident acoustic wave.Therefore, the particle displacement is considered to be quite small.In fact, the solution of the problem is derived from a system of two immobile suspended liquid spherical drops in the present paper (see Figure 1).Accounting for the translational motion of the target results in the nonlinear problem formulation and essentially complicates the solution procedure.The linear problem formulation chosen provides a relatively straightforward way to derive the exact analytical solution for the velocity potential determination while accounting for the incident wave scattering on the spheres.As such, the boundary conditions on both spherical surfaces are satisfied exactly. to the wave length, the properties of a bearing liquid (or gaseous) medium, wavescattering on the other particles, and their mutual effect (if system of solids, drops, or bubbles are under investigation), etc.Some of these aspects were already addressed in the publications devoted to the estimation on ARF for a case of solitary rigid, liquid, or gaseous particles located in an infinite liquid space or in the vicinity of the plane boundary of a liquid (see papers [5,6,16]). In the present paper, the case of two spherical liquid drops placed in an ideal compressible liquid is under investigation.The plane acoustic wave propagates along the line passing through the centers of the spherical drops.The purpose of the work is to study the interaction regularities between liquid spherical particles and to estimate the ARF induced by the acoustic wave irradiation of the system of the drops.The technique of the problem solution is elaborated.It comprises two basic steps according to the version of approach developed in the paper by Guz and Zhuk [5].At the first step, solution tothe problem of the incident wave scattering on the system of two liquid particles (drops) is addressed.Once the problem is solved and the velocity field potential is derived, the main features of the two-drop system and acoustic wave can be studied.At the second step, the hydrodynamic forces experienced by each of the particles are calculated.Then, the ARFs are obtained by the time-averaging over the period of the primary wave oscillation.The results of the ARF case studies for the variety of liquid media and spheres' liquid properties, as well as for different geometrical parameters of the system are presented.Conclusions on the main regularities for ARF affecting both liquid spherical particles are drawn. The Problem of the Velocity Potential Determination To clarify the problem statement even further, it is worth mentioning here that the velocity of the translational movement of each spherical particle is supposed to be negligibly small in comparison to the velocity of the incident acoustic wave.Therefore, the particle displacement is considered to be quite small.In fact, the solution of the problem is derived from a system of two immobile suspended liquid spherical drops in the present paper (see Figure 1).Accounting for the translational motion of the target results in the non-linear problem formulation and essentially complicates the solution procedure.The linear problem formulation chosen provides a relatively straightforward way to derive the exact analytical solution for the velocity potential determination while accounting for the incident wave scattering on the spheres.As such, the boundary conditions on both spherical surfaces are satisfied exactly.The linear approximation problem statement for the first step mentioned above, i.e., the problem of determination of the fluid velocity potential field in a tube is described The linear approximation problem statement for the first step mentioned above, i.e., the problem of determination of the fluid velocity potential field in a tube is described below.The procedure adopted for this aim corresponds to the approach developed in other papers [5,15].It is assumed that an ideal compressible fluid of density ρ 0 fills the outer space continuously.The velocity of the plane sound wave propagating in the liquid is a 0 . Two spherical drops of another ideal liquid are placed in the medium.The distance between the drops' centers is denoted with .Furthermore, the number 1 is always assigned to the sphere which is placed first on the way of propagating acoustic wave, while number 2 is assigned to the second drop in a row (see Figure 1).Let us denote the density of the liquid of the first and second drop as ρ 1 and ρ 2 , respectively.Notations a 1 and a 2 are used for the speed of sound in the liquid of the first and second drops, respectively, while R 1 and R 2 are the drop radii, respectively.The main coordinate system is chosen to be the rectangular Cartesian coordinate system Oxyz with the origin located in the middle between the drop centers.Two other Cartesian rectangular O s x s y s z s and spherical O s r s θ s ϕ s , s = 1, 2 coordinate systems attached to the spherical drops are introduced in such a manner that their origins O s coincide with the drop centers and the axis Oz s runs along the line that passing through the sphere centers (see Figure 1).Let us apply the model of compressible barotropic liquid to describe the response of both the liquid medium and the drops.Conditions of a potential flow are assumed to be valid to determine the medium's effect on the drops. In this configuration, the steady wave process is under consideration.The incident plane harmonic wave propagating in a positive direction of the Oz axis is governed by a potential Φ i .Let us denote the potentials of the waves scattered on the spherical drops as Φ (s) d (s = 1, 2).The general potential of the velocity field Φ is, therefore, the combination of the abovementioned potentials.Then, the pressure p in the bearing liquid (medium) can be determined by making use of the potential Φ, with accuracy up to the squares of the Mach number according to the following formula [3,5]: where potential Φ is the solution of the following wave equation: Then, determination of the ARF acting upon the target in the liquid is reduced to the time-averaging of the hydrodynamic force, as follows: where N is a vector of a unit normal to the external surface S of the target.The linearity of the Equation (2) allows one to apply the superposition principle for the determination of the velocity field in the case of the system of objects placed in the liquid and, therefore, to determine the solution to the problem of the interaction of the bodies. Let the incident wave potential (the solution of the Equation ( 2)) be as follows: where A is the amplitude; k = ω/a 0 is the wave number; ω is the angular frequency; t is the time.An incident wave with potential Φ i propagates along the positive direction of the Oz axis. As mentioned above, the wave field in the bearing liquid (medium) is formed by the interference of the incident wave (4) and waves scattered on the spherical drops, as follows: (5) Furthermore, index s (either subscript or superscript) is used to denote the quantity related to the drop with number s. From the mathematical point of view, determination of the potentials (5) of the stationary wave field in a liquid is reduced to the solution of a linear problem of incident wave (4) scattering on the liquid spherical drops over the multiply connected region, i.e., the problem is reduced to finding solution to the linear differential Equation (2) that meets the boundary conditions on the surfaces of spherical drops. The boundary conditions demand continuity of the radial components of velocity in the outer and internal liquids on the drop surfaces, as well as continuity of pressure through the surfaces.They can be written in the following form: Let us emphasize here that potentials of the waves scattered on the spherical drops have to meet the Sommerfeld radiation condition for infinitely distant points.Furthermore, the drop surface tension is not taken into account in boundary condition formulations.It is also assumed the amplitude of the drop surface oscillations is small enough to consider the drop radius to be constant.In Equation (6), the bar over a symbol denotes the quantities characterizing the state of the liquid inside the corresponding drop. Wave Field Potential Construction: Determination of the Acoustic Pressure As mentioned above, the ARF affecting an object submerged in a liquid is determined as a time-averaged value of the acoustic pressure over the object surface.Thus, the Lagrange reference frame is more suitable to derive the problem statement for this case.To apply this approach, the acoustic pressure in the vicinity of the object considered has to be calculated with accuracy up to the second order of magnitude.It means that the deviation of the acoustic pressure from the harmonic law has to be accounted for in the region around the object [1][2][3].Let us emphasize once again that a linear approximation of the radiation pressure in the vicinity of the object yields zero ARF, which contradicts the experimental data.Therefore, preservation of the quadratic terms in the wave equation is needed to avoid occurrence of the zero ARFs.Evaluating the acoustic pressure with accuracy to the quadratic terms, thus, providing the velocity field potential, is achieved by using a linear diffraction problem in the present paper. To construct the solution to the problem (2), let us apply the variable separation method with respect to the local spherical coordinate systems associated with the corresponding drop.The potential of the incident wave (1) with respect to the spherical coordinate system O s r s θ s ϕ s , s = 1, 2 is as follows: where j n (kr s ) are the spherical Bessel functions of the first kind, and P n (cos θ s ) are the orthonormalized Legendre polynomials. The potential of the wave scattered on the sth drop, Φ d (s = 1, 2), (i.e., the solution of Equation (2) satisfying the radiation condition for infinitely distant points) is represented by the generalized Fourier series expansion over the spherical wave functions, as follows: where h (1) n (kr s ) is the spherical Hankel function of the first kind. Axioms 2023, 12, 940 6 of 14 The potential Ψ (s) , s = 1, 2, of the wave field in the liquid inside the corresponding spherical drop (which is solution of Equation (2) that meet the boundedness condition) can be written as follows: To find the expansion coefficients A n , s = 1, 2, in expressions ( 8) and ( 9), the boundary conditions (6) on the spherical surfaces are used.Therefore, the potential (5) of the acoustic field in the outer liquid should be rewritten with respect to each local spherical coordinate system.The addition theorems for spherical wave functions have to be implemented to realize it.As a result, the following expression can be derived: (10) where multiplier exp(−iωt) is omitted. The following notations are used in (10): 0n0p kR sq , θ sq , s, q = 1, 2; s = q; , if w even; Here R sq , θ sq , ϕ sq are spherical coordinates of the pole O q located at the sphere center with respect to sth local spherical coordinate system, i.e., they are coordinates of the center (pole) O 2 with respect to the center (pole) O 1 and vice versa.Thus, the relations R sq = R qs = are true for the particular geometry under consideration. At the first stage of the ARF determination (i.e., to solve the linear problem of the incident wave scattering on the spherical drops), the following formulae are used to calculate the pressure and disturbance of liquid inside the drops: Making use of the boundary conditions on the drop surfaces and taking into account the expressions for the potentials (9), (10), along with the formulae for pressure and velocity of liquid, yields an infinite system of algebraic equations with respect to the unknown coefficients A Let us emphasize there is no need to use the bar over symbols k s and R s in (12). The infinite system of algebraic Equation ( 12) has a unique solution A n , s = 1, 2 of truncated system (12) of the algebraic equations formally completes the procedure of determination of the velocity field potentials for both the outer liquid and the liquids inside the drops.The accuracy needed is achieved by the comparison of calculation results for a consecutively growing number of equations. Determination of the Acoustic Radiation Forces Acting upon Liquid Drops The acoustic radiation force affecting two spherical drop submerged in a liquid is equal to the hydrodynamic force acting on each spherical particle averaged over the period of an incident wave and is, in fact, the constant component of this force.In the system configuration under study, the hydrodynamic forces acting on each drop are directed along the Oz axis because the liquid velocity field is symmetrical with respect to this axis.This force can be calculated as an integral of the pressure p (s) over the correspondent drop surface [20], as follows: where pressure p (s) is calculated by making use of Expression (1) where the real part of the complex potential (10) has to be used.As a result, the formula can be derived as follows: Here, the following notations are used: n j n (kr s ). For the sake of brevity, the symbol Re denoting the real part ReΦ (s) of the potential Φ (s) will be omitted. The contribution of the first term in the righthand side of Expression ( 1) is zero because it varies sinusoidally over time [5,6].This term will, therefore, not be taken into consideration for ARF calculation any further.Finally, taking account of the reasoning concerning Equation (1), as mentioned, the expression for the pressure p takes the following form: Let us emphasize here that the pressure is calculated on the surface of the spherical drop (i.e., r s = R s ). Let us determine the contribution to the hydrodynamic force (13) (and, therefore, to ARF) of each additive from Formula (15) for the pressure. The contribution of the first term to the force F (s) z , s = 1, 2 is governed by an integral, as follows: Taking into account ( 14) and the following condition: 1 where µ = cos θ s , integration of ( 16) yields the expression for the contribution of the first term from (15) to the total value of the hydrodynamic force F z , as follows: The terms resulting in nonzero values being averaged over the period are preserved in Expression (18) only.Time-averaging of (18) over the period of the incident wave yields the contribution of the first term from the expression for pressure (15) to the total value of the acoustic radiation force, as follows: The contribution of second term from ( 15) is governed by an integral, as follows: Making use of Expression (14) and Condition (17), the integration of (20) yields the following formula for the estimation of the second term contribution to the force F (s) z : In Formula (21), the derivatives are taken with respect to α s = kr s at r s = R s .The terms resulting in zero values being averaged over the period are not taken into account in (21).Time-averaging of (21) over the period of the incident wave yields the contribution of the second term from the expression for pressure (15) to the total value of the ARF, as follows: The contribution of the third term from ( 15) is, in turn, determined by an integral, as follows: Taking account for ( 14) and the following condition: 1 integration of (23) yields the expression for the contribution of the third term from (15) to the total value of the hydrodynamic force F (s) z , as follows: The terms resulting in zero values being averaged over the period are not taken into account in (24).Time-averaging of (21) over the period of the incident wave yields the contribution of the third term from the expression for pressure (15) to the total value of the ARF, as follows: Summation of the expressions ( 19), ( 22), and (25) gives the final formula for the acoustic radiation force affecting each of the two spherical liquid drops placed in the outer liquid (bearing acoustic medium), provided that the acoustic wave propagates along the line passing through the centers of the drops, as follows: The formulae obtained through the mathematical manipulations above show that the expressions for the ARF are of the same form as the ones derived in [20].Nevertheless, let us emphasize here that the expressions are similar only in appearance.In fact, they predict absolutely different responses of the system depending on the nature of acoustic medium, constituents, and the region under investigation. Numerical Results of ARF Calculation and Discussion As an example of the developed theory application, investigation of the acoustic radiation force influencing a system consisting of the two liquid spherical drops placed in the liquid medium in the acoustic field is carried out.Calculations are performed numerically, making use of Equation ( 15) for pressure and (26) for the ARF itself.It is worth mentioning here that ARF is of primary interest due to the technological applications mentioned in the Introduction section.With this idea in mind, several types of bearing outer liquid (liquid medium) and liquids filling the drops were considered.The parameters of the incident sound wave were chosen to correspond to moderate radiated power.Dimen-sionless amplitude A in Expression (4) is chosen to be equal to 0.918 × 10 -4 for the case under consideration. The calculation algorithm is organized as a two-stage procedure.At the first stage, the determination of the velocity field potential according to Expression ( 5) is carried out.For this aim, calculation of the coefficients A (s) n and A (s) n , s = 1, 2, from the truncated system of the algebraic Equation ( 12) with account of the Expression (10), is carried out with the preset accuracy of 10 -6 , which is secured by comparison of the evaluation results for the sequentially increasing number of the equations in the truncated system.As soon as the n and A (s) n are found, the absolute value of the linear hydrodynamic acoustic pressure is calculated with making use of the expressions ( 8), (9), and (10). In the second stage, the ARF itself is calculated, making use of (26). As mentioned in the Introduction, estimation of the ARF is of interest for the technological applications listed above.The approach proposed and expressions derived allow one to characterize the ARF without direct determination of the acoustic pressure field, which is not of primary interest in the present paper. In the present study, water is considered as a bearing liquid (liquid medium) for the system of two drops.The following characteristics of the water are used: sound velocity is a 0 = 1500 m/sand density is ρ 0 = 1000 kg/m 3 .The properties of the drop liquids and geometrical parameters of the system will be individually indicated for each particular case. To study the influence of the drop radii on the ARF behavior, dependencies of the ARF on the incident wave frequency are shown in Figure 2 for two carbon tetrachloride drops placed in water on the distance = 50 mm between drop centers.The sound velocity and density of the carbon tetrachloride are chosen to be equal to a = 926 m/s and ρ = 1594 kg/m 3 , respectively.The radius of the first drop, R 1 , is always equal to 5 mm, while the radius of the second drop, R 2 , varies in Figure 2 (values are shown at the corresponding curve).The ARF acting upon the first drop, F , is shown with dashed line, while the same quantity for the second drop, F (2) z , is depicted with a solid line elsewhere in the paper.Let us reiterate here that we agreed that the acoustic wave propagates in the direction from the first drop to the second one. Axioms 2023, 12, 940 11 To study the influence of the drop radii on the ARF behavior, dependencies o ARF on the incident wave frequency are shown in Figure 2 , is depicted with a solid elsewhere in the paper.Let us reiterate here that we agreed that the acoustic propagates in the direction from the first drop to the second one. Figure 2a shows that ARF acting upon the first drop is independent of the rad the second drop (the dashed lines coincide to the drawing accuracy).It always rem negative for the case studies within the radius range of interest.This means that th rection of (1) z F is always opposite to the direction of incident wave propag Therefore, ARF causes the start of drop motion towards the direction of sound propagation.These results corresponds to the data obtained in [5,20].Figure 2a shows that ARF acting upon the first drop is independent of the radius of the second drop (the dashed lines coincide to the drawing accuracy).It always remains negative for the case studies within the radius range of interest.This means that the direction of F (1) z is always opposite to the direction of incident wave propagation.Therefore, ARF causes the start of drop motion towards the direction of sound wave propagation.These results corresponds to the data obtained in [5,20]. The values and direction of F (2) z vary depending on the incident wave frequency.Details are shown by the solid lines in Figure 2b.Here again, the radii of the second drop are written at the corresponding lines.It is obvious there are frequency intervals where the direction of the F (2) z is opposite to the F (1) z leading to the divergence of drops (they drift out of each other).In the other intervals where the directions are the same, the drops start to drift unidirectionally.Figure 2 illustrates one more remarkable situation.There are frequency values at which the ARF acting upon the second drop is equal to zero, meaning that it remains immobile. In contrast to Figure 2, where R 1 ≥ R 2 , Figure 3 contains curves that illustrate the same dependencies but for the system where the second drop radius is larger than the one of the first drop, i.e., R 1 ≤ R 2 .The behavior of the curves for the ARF acting upon the first drop, F , is quite different from the one depicted in Figure 2. The next geometrical parameter that significantly influences the ARF response distance, , between the drop centers.In Figure 4, the dependencies of (1) z F (2) z F on the  are shown for several different frequencies of the incident wave fo same system of two carbon tetrachloride drops of equal radii 1 2 R R = = 5 mm plac water.The wave frequency is written by the correspondent curve.Analysis of the f evidently shows that the closer drops are located, the higher valuesof ARF observe increase in the distance  leads to a decrease in the interaction between drops. The following features of the ARF behavior can be noted: at all frequencies in gated, the value of ARF acting upon the first drop remains negative, causing the st the drop shift in the direction which is opposite to the direction of incident wave p The value of this ARF decreases as the drop radius becomes smaller.It can be explained by the general reduction in the drop surface and, therefore, integration over the smaller surface leads to the lower ARF value.Nevertheless F , is qualitatively the same as in Figure 2. Therefore, one can again control the drop's movement by varying the frequency of the incident wave. The next geometrical parameter that significantly influences the ARF response is the distance, , between the drop centers.In Figure 4, the dependencies of F Figure 5 illustrates the influence of the physical characteristics of the liquid volved (both the outer liquid medium and the liquid inside the drops) on the aco radiation forces.Here, typical results for the system under investigation are shown outer liquid is always water, while drops are considered to be ones of the same li (carbon tetrachloride, benzene, or mercury) possessing the range of densities and sp of sound in them.The following features of the ARF behavior can be noted: at all frequencies investigated, the value of ARF acting upon the first drop remains negative, causing the start of the drop shift in the direction which is opposite to the direction of incident wave propagation (see Figure 4a).The direction of the F (2) z varies depending on the distance between the drop centers.Once again, it can be directed either towards the first drop or away from it, and there are specific values of the frequency providing zero values of the F (2) z .This means that the second drop remains immobile and doesn0t start to move under the conditions mentioned. Figure 5 illustrates the influence of the physical characteristics of the liquids involved (both the outer liquid medium and the liquid inside the drops) on the acoustic radiation forces.Here, typical results for the system under investigation are shown.The outer liquid is always water, while drops are considered to be ones of the same liquid (carbon tetrachloride, benzene, or mercury) possessing the range of densities and speeds of sound in them.Figure 5 illustrates the influence of the physical characteristics of the liquid volved (both the outer liquid medium and the liquid inside the drops) on the aco radiation forces.Here, typical results for the system under investigation are shown outer liquid is always water, while drops are considered to be ones of the same l (carbon tetrachloride, benzene, or mercury) possessing the range of densities and s of sound in them.In Figure 5a, dependencies of the ARF on the incident wave frequency are shown for two drops of carbon tetrachloride, benzene (a = 1298 m/s and ρ = 870 kg/m 3 ), and mercury (a = 1450 m/s and ρ = 13593 kg/m 3 ).The drops are of the same radii, i.e., R 1 = R 2 = 5 mm.The distance between centers is = 30 mm.The liquid of the drops is labeled by the corresponding curve. It can be seen that the density of the inner liquid affects ARF, acting on the first drop significantly.A denser liquid (mercury) causes changes even in the ARF direction.This corresponds to the case of a rigid particle placed in an ideal compressible liquid [5].The F (2) z force changes its sign in comparison to the cases of benzene and carbon tetrachloride as well.The ARF acting on the second drop is of variable direction depending on the incident wave frequency, as was mentioned for Figure 2. The frequencies providing zero value F (2) z = 0 occurred as well.The same tendency for more dense liquids is preserved for the ARF dependencies on the distance between drop centers presented in Figure 5b for the same system as in Figure 5a for incident wave frequency 40 kHz.The values of the ARF are of the opposite sign and are much higher than for the less dense liquids.In all the cases, the behavior is also influenced by the speed of sound in the liquid.The variations in the direction of the ARF acting upon the second drop are also observed. Conclusions The case of two immobile spherical liquid drops placed in an ideal compressible fluid in the field of an acoustic wave propagating along the line passing through the drop centers is under investigation.The approach to estimate the acoustic radiation force induced due to acoustic pressure imposed onto the drops and the interaction between them is elaborated. The calculation algorithm is organized as a two-stage procedure.At the first stage, the determination of the velocity field potential is carried out.The problem is solved by the variable separation method.To satisfy the boundary conditions on spherical surfaces, the expansion of the incident and reflected wave potentials over the spherical wave functions are used.Required constants in the solution are calculated from an infinite system of algebraic equations, which is solved by a truncation method. At the second stage, encompassing the determination of hydrodynamic forces acting on the liquid spheres with their subsequent averaging over the suitable time interval, the ARF itself is calculated making use of the expression derived. The main regularities of the interaction between liquid spherical particles as well as ARF induced by the acoustic wave irradiation of the system of the drops are studied with the application of the proposed approach.It was found the behavior of the ARF depends significantly on the geometrical parameters of the system, properties of the liquids involved, and frequency of the incident wave.The possibility of controlling the drops' movement by varying the frequency of the incident wave for a particular setup of the system is defined. Figure 1 . Figure 1.Twoliquid drops placed in an ideal compressible liquid and affected by the acoustic field.The plane acoustic wave propagates along the line passing through the centers of the spherical drops in the direction 1 2 O O . Figure 1 . Figure 1.Two liquid drops placed in an ideal compressible liquid and affected by the acoustic field.The plane acoustic wave propagates along the line passing through the centers of the spherical drops in the direction O 1 O 2 . n , s = 1, 2 of the generalized Fourier series expansions for the potentials Φ 1, 2, . . .that can be found by the truncation method.Calculation of the coefficients A 1 R for two carbon tetrachl drops placed in water on the distance =  50 mm between drop centers.The soun locity and density of the carbon tetrachloride are chosen to be equal to a = 926 m/ ρ= 1594 kg/m 3 , respectively.The radius of the first drop, , is always equal to 5 while the radius of the second drop, 2 R , varies in Figure 2 (values are shown a corresponding curve).The ARF acting upon the first drop, (1) z F , is shown with d line, while the same quantity for the second drop, Figure 2 .FFFigure 2 . Figure 2. Dependencies of the ARF on the incident wave frequency for the system of two c tetrachloride drops placed in water on the distance =  50 mm.(a) Graphs of the ARF for t dius of the first drop 1 R = 5 mm and radius of the second drop 2 R = 5, 4, 3, 2, 1 mm; (b) d of the (2) z F .Figure 3 . Figure 3. Dependencies of the ARF on the incident wave frequency for the system of two c tetrachloride drops placed in water on the distance =  50 mm.Radius of the second drop R mm.(a) Graphs of the ARF for radius of the first drop 1 R = 5, 4, and 3 mm; (b) graphs of th for radius of the first drop 1 R = 2 and 1 mm.The value of this ARF decreases as the drop radius becomes smaller.It can b plained by the general reduction in the drop surface and, therefore, integration ove smaller surface leads to the lower ARF value.Nevertheless Figure 3 . Figure 3. Dependencies of the ARF on the incident wave frequency for the system of two carbon tetrachloride drops placed in water on the distance = 50 mm.Radius of the second drop R 2 = 5 mm.(a) Graphs of the ARF for radius of the first drop R 1 = 5, 4, and 3 mm; (b) graphs of the ARF for radius of the first drop R 1 = 2 and 1 mm. all values of geometrical parameters and frequencies investigated.The behavior of the acoustic radiation force acting upon the second drop, F(2) z Figure 4 . Figure 4. Dependencies of the ARF on the distance, , between the drop centers.(a) Curves fo incident wave frequencies of 40 and 50 kHz; (b) details of the ARF acting on the second dro frequencies of 10, 20, 30, 40, and 50 kHz. Figure 5 . Figure 5. Example of the liquid properties' effect on the ARF behavior for two drops of carbo rachloride, benzene, and mercury placed in water.(a) Dependencies of the ARF on the inc wave frequency; (b) ARF dependencies on the distance between drop centers.In Figure5a, dependencies of the ARF on the incident wave frequency are show two drops of carbon tetrachloride, benzene ( a = 1298 m/sec and ρ= 870 kg/m 3 ), mercury ( a = 1450 m/s and ρ= 13593 kg/m 3 ).The drops are of the same radii, 1 2 R R = = 5 mm.The distance between centers is =  30 mm.The liquid of the dro labeled by the corresponding curve. Figure 4 . Figure 4. Dependencies of the ARF on the distance, , between the drop centers.(a) Curves for the incident wave frequencies of 40 and 50 kHz; (b) details of the ARF acting on the second drop for frequencies of 10, 20, 30, 40, and 50 kHz. Figure 4 . Figure 4. Dependencies of the ARF on the distance, , between the drop centers.(a) Curves f incident wave frequencies of 40 and 50 kHz; (b) details of the ARF acting on the second dr frequencies of 10, 20, 30, 40, and 50 kHz. Figure 5 .Figure 5 . Figure 5. Example of the liquid properties' effect on the ARF behavior for two drops of carb rachloride, benzene, and mercury placed in water.(a) Dependencies of the ARF on the in wave frequency; (b) ARF dependencies on the distance between drop centers.In Figure 5a, dependencies of the ARF on the incident wave frequency are show two drops of carbon tetrachloride, benzene ( a = 1298 m/sec and ρ= 870 kg/m 3 ) mercury ( a = 1450 m/s and ρ= 13593 kg/m 3 ).The drops are of the same radi 1 2 R R = = 5 mm.The distance between centers is =  30 mm.The liquid of the dr	2023-10-05T15:19:04.980Z	2023-09-29T00:00:00.000	{ "year": 2023, "sha1": "dbf5d48f8c26b1ebad3463bb7e077d4fa72f3849", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2075-1680/12/10/940/pdf?version=1695995382", "oa_status": "GOLD", "pdf_src": "ScienceParsePlus", "pdf_hash": "a67655206f56536948a4ec63e35b0eaf3b1180cb", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Computer Science" ] }
259095712	pes2o/s2orc	v3-fos-license	The magnetic dual chiral density wave phase in a rotating cold quark matter The effect of rotation on the formation of the magnetic dual chiral density wave (\MD) in a dense and magnetized cold quark matter is studied. This phase is supposed to exist in the extreme conditions prevailing, e.g., in a neutron star. These conditions are, apart from high densities and strong magnetic fields, a relatively large angular velocity. To answer the question of whether the rotation enhances or suppresses the formation of this phase, we first determine the effect of rotation on the energy dispersion relation of a fermionic system in the presence of a constant magnetic field and then focus on the thermodynamic potential of the model at low temperature $T$ and finite chemical potential $\mu$. The thermodynamic potential consists, in particular, of an anomalous part leading to certain topological effects. We show that in comparison with the nonrotating case, a term proportional to the angular velocity appears in this anomalous potential. We then solve the corresponding gap equations to the chiral and spatial modulation condensates, and study the dependence of these dynamical variables on the chemical potential ($\mu$), magnetic field ($eB$), and angular velocity ($\Omega$). It turns out that the interplay between these parameters suppresses the formation of the \MD~phase in relevant regimes for cold neutron stars. This is interpreted as the manifestation of the inverse magnetorotational catalysis, which is also reflected in the phase portraits $eB$-$\mu$, $eB$-$\Omega R$, and $\mu$-$\Omega R$, explored in this work. I. INTRODUCTION Exploring the phase diagram of quark matter under extreme conditions is one of the important subjects in nuclear physics.These conditions are, among others, high temperature, large baryon chemical potential, and the presence of uniform electromagnetic fields.Big questions related to these subjects are discussed in [1].Open problems and possible past, present, and future theoretical and experimental strategies to answer these questions are discussed recently in various reports and reviews [2][3][4][5].Apart from standard simulation methods in lattice Quantum Chromodynamics (QCD) [3], new computational tools, such as machine learning [6], are developed and applied to either analyze the experimental data or to overcome the deficiencies of standard methods in working with QCD under extreme conditions. Neutron stars are the natural playground to study quark matter under extreme conditions.These stars, which are produced by gravitational collapse of very massive stars, are the densest objects in the Universe.Their inner density is several times larger than the nuclear saturation mass density ρ n ∼ 2.5 × 10 17 kg/m 3 (see e.g.[7]).There have been several attempts to determine the equation of states of neutron stars [8].The latter is a necessary input for the Toleman-Oppenheimer-Volkoff equation, which together with the mass continuity equation yields the mass-to-radius ratio of these stars [9].Using effective models, like the Nambu-Jona-Lasinio (NJL) model, it was found that the matter in the interior of neutron stars is in the color superconductivity (CS) phase [10].In particular, the three-flavor color-flavor locked (CFL) phase seemed to be the most favored phase [11]. The CFL phase, however, does not pass certain astrophysical tests [12], and is thus ruled out. Apart from large densities, neutron stars exhibit very strong magnetic fields.The strength of the magnetic fields of magnetars is estimated to be ∼ 10 18 G for nuclear matter and ∼ 10 20 G for quark matter (see [12] and references therein).The effect of constant magnetic fields on CS phases is studied, e.g. in [13].In [14], it is shown that the two-flavor CS phase is also favorable in a magnetized quark matter at intermediate chemical potentials.By increasing the baryon density from low values to densities a few times higher than the nuclear saturation density, another phase can be formed at low temperatures.This phase, originally introduced in [15] and studied in several follow-up papers, e.g.[16], is characterized by quark-hole pairs having a finite total momentum and leading to standing waves.In [17], in analogy to the static spin density waves, known from condensed matter physics, a density wave is introduced in quark matter at moderate densities which is referred to as the "dual chiral density wave".It is represented by a dual standing wave in scalar and pseudoscalar condensates.The effect of uniform magnetic fields on the formation of this phase is studied for the first time in [18].The corresponding phase is dubbed the MDCDW phase.It is characterized by two dynamical variables, chiral and spatial modulation condensates, which are determined by solving the corresponding gap equations.It is further shown that because of a certain asymmetry appearing in the energy dispersion relations corresponding to the lowest Landau level (LLL), the thermodynamic potential consists of an anomalous term.As it is argued in [19,20], the anomalous term leads to certain topological effects that include, among others, an anomalous nondissipative Hall current and an anomalous electric charge.Moreover, this phase is characterized by the formation of a hybridized propagating mode known as an axion-polariton, which has interesting astrophysical consequences discussed exten-sively in [21].Recently, in [22], the phase diagram of the MDCDW phase was explored at finite temperatures and in the presence of uniform magnetic fields.It is shown that the MDCDW phase is favored at magnetic fields and temperatures compatible with neutron stars.In particular, at intermediate densities, where a remnant mass is formed, the spatial modulation increases.This opens the possibility for this phase to be a favorable candidate for the quark matter in neutron stars. In addition to high densities and large magnetic fields, neutron stars are also characterized by relatively large angular velocities of about Ω max ∼ 10 3 Hz.This leads to a linear velocity ∼ 10 −2 c − 10 −1 c, where c is the light velocity.It is the purpose of this paper, to study the effect of rotation on the formation of the MDCDW phase.We assume that the magnetic field and angular velocity are uniform.This is only possible for cold neutron stars which are expected to rotate uniformly [23].This assumption justifies another assumption concerning the temperature in this paper: Here, in contrast to [22], we neglect the effect of temperature and its possible interplay with high densities, magnetic fields, and rotations on the formation/suppression of the MDCDW phase in a cold quark matter. As it is argued in [24,25], a certain interplay between the effect of rotation and the magnetic field destroys chiral condensates.An effect which is referred to as "rotational magnetic inhibition" [24] or "inverse magnetorotational catalysis" (IMRC) [25].For the MDCDW phase, the consequence of this effect would be a vanishing of the spatial modulation condensate, as a result of a certain correlation between two condensates in this phase.In this paper, we show that this indeed happens.A fact that may rule out this phase to be favorable for quark matter in rotating, dense, magnetized, and cold neutron stars.We emphasize that the present work is a natural extension of [22], where it is argued that the MDCDW phase is "viable candidate for the matter of neutron stars" [22].Particularly in this context, we show that the inclusion of rotation to the condition imposed on the model used in [22] suppresses the formation of the MDCDW phase.It is not clear how robust the conclusions presented in [22] as well as those in the present paper, are in a more realistic model of a neutron star, where certain conditions, like β-equilibrium, isospin symmetry, and charge neutrality have to be imposed [26]. The organization of this paper is as follows: In Sec.II, we start with the Lagrangian density of a two-flavor NJL model with U (1) L × U (1) R × SO(2) × R 3 symmetry in the presence of a background magnetic field and introduce the rotation by implementing an angular velocity parallel to the magnetic field in the Hamiltonian of the model.Then, defining two inhomogeneous condensates, we introduce the mass and spatial modulation condensates.We then solve the energy eigenvalue equation and determine, in particular, the energy eigenvalues and eigenfunctions in a cylindrical coordinate system.The solution to the Dirac equation in a rotating fermionic system with an without boundary conditions, in the absence and/or presence of uniform magnetic fields is studied in [27][28][29][30], apart from [24,25].In the present paper, we neglect the effect of boundary conditions.In Sec.III, we determine the thermodynamic (one-loop effective) potential at finite temperature and density, and in the presence of constant magnetic fields.In particular, we focus on the low temperature limit as in [18] and explore the effect of rotation on the anomalous part of the effective potential.Having the thermodynamic potential at hand, it can be minimized with respect to two condensates.In Sec.IV A, we present our numerical results for the dependence of mass and spatial modulation condensates on the chemical potential µ, magnetic field eB and linear velocity ΩR, with Ω the angular velocity and R the radius.The results confirm the fact that rotation destroys both condensates as a consequence of the IMRC effect.Moreover, it is shown that for µ = 0, the role of µ is played by a pure rotation in interplay with a background magnetic field.In Sec.IV B, we then explore the eB-µ, eB-ΩR, and µ-ΩR phase diagrams to study the impact of the IMRC effect on various phases appearing in the model.We conclude our results in Sec.V. II. THE MODEL To study the MDCDW phase in a rotating and dense medium, we start with the Lagrangian density of a twoflavor gauged NJL model, with e > 0. The Dirac γ matrices in (II.2) and (II.3) are defined in the Weyl representation with the Pauli matrices τ = (τ 1 , τ 2 , τ 3 ).Moreover, the total angular momentum in the z direction is given by As is described in [12], the MDCDW phase at finite baryon density is characterized by two nonlocal condensates with ∆ = const and q • x ≡ q µ x µ .For simplicity, we assume that the modulation vector is aligned along the magnetic field direction z, q µ = (0, 0, 0, 2b) [12,18].Expanding ψψ 2 and ψiγ 5 τ 3 ψ 2 in the interacting part of the Lagrangian (II.1) around their mean fields (II.5), we arrive first at the semi-bosonized Lagrangian where the dynamical mass m ≡ −2G∆, with ∆ given in (II.5) and θ ≡ bz.Performing at this stage a local chiral transformation as in [12,18] ψ → e +iγ 5 τ3θ ψ, ψ → ψe +iγ 5 τ3θ , (II.7) we arrive at (II.8) Here, we used / ∂θ = γ 3 b.Plugging τ 3 into (II.8),we obtain the corresponding Lagrangian density for each flavor L f , Here, / Π f is defined in (II.3) and s f ≡ sign(q f ) is given by (s u , s d ) = (+1, −1) for up and down quarks, respectively.To determine the energy spectrum of this model, we solve the energy eigenfunction equation, with the Hamiltonian for each flavor f , Here, γ • Π f includes the spatial part of Π µ f , defined in (II.3).Following the method presented in App.A, the corresponding energy eigenvalues to the lowest and higher Landau levels (LLL and HLL) read: -For LLL (n = 0), we obtain with ϵ = ±1. -For HLL (n > 0), we obtain In what follows, we determine the thermodynamic potential of this model. III. THE THERMODYNAMIC POTENTIAL To determine the thermodynamic potential of the MDCDW model at finite temperature T and chemical potential µ, let us start with where V is the four-dimensional space-time volume, and the partition function Z is given by with the mean-field Lagrangian that, comparing to L MF from (II.8), we have introduced µ and neglect the current mass m 0 .Using then the definition of the Hamiltonian H f from (II.12), V eff is first given by where the trace is to be built over the spin (s), flavor (f), and color (c) degrees of freedom.To do this, we consider, as in [24], a cylindrical volume with L z , the length of the cylinder, and R, its radius.Assuming as in [24] that the maximum of ψ(ρ, φ, z) is inside the cylinder, it is possible to determine an upper bound for the summation over ℓ (see the Appendix of [24]).This leads to the following phase space for the positively and negatively charged quarks in a rotating medium Here, n and ℓ are the quantum numbers corresponding to Landau levels and rotation, respectively.Moreover, N f , the Landau degeneracy factor for each flavor f = {u, d} is defined by Here, S = πR 2 .Using the replacement p 0 → iω k ≡ iπT (2k + 1), and and performing the summation over the Matsubara frequencies ω k by making use of with β ≡ T −1 , we arrive at the thermodynamic potential at finite T and µ, where and N c is the number of colors, and E n f for n = 0 and n > 0 are given in (II.13) and (II.14), respectively.Similar to [18], we separate V (1) eff into the zero T part, V T=0 , the part including µ and Ω, V µ,Ω , and the finite T part, (III.12) In what follows, we consider only the T → 0 case.Using lim T →0 T ln 1 + e −βx = −xθ(−x), the V T ̸ =0 vanishes.We thus focus on V T =0 and V µΩ in (III.12).Similar to the nonrotating case, V T =0 has to be appropriately regularized.According to the definition of E n f from (II.13) and (II.14),E n f + Ωj = E Ω=0 f is independent of j = ℓ+1/2 (or equivalently ℓ).A summation over ℓ is therefore possible, and leads to a factor N f , defined in (III.6).The zero T part of the potential is thus given by Apart from the factor N f /S, this is the same integral that appears also in [18,22].Using the same proper-time regularization as in [18,22], it reads Here, Λ is a cutoff regulator.It is possible to perform the summation over Landau levels n and ζ as well as ϵ.Using the final expression for V T =0 , is given by coth (s\|q f eB\|) . (III.16) Let us now consider V µ,Ω from (II.12).To regularize it, we use as in [18], a cutoff regularization by introducing a Heaviside θ as a function of the regulator To evaluate V µ,Ω , it is necessary to separate it into two parts, corresponding to LLL and HLLs, n = 0 and n > 0, Let us first consider the LLL contribution to the effective potential, V n=0 µ,Ω .Following the procedure described in [18], we arrive first at For numerical purposes, it is necessary to sum over ϵ and perform the integration over p z in the first term of V n=0 µ,Ω .To do this, we use the method described in [18], and arrive after some work at2 with a 1 ≡ b − Ωj − µ, a 2 ≡ b, P ai = a 2 i − m 2 , i = 1, 2, and I(Λ 1 , Λ 2 ) defined in (B.5).The second term in (III.20), is the well-known anomalous potential, V anomal [12,18], Having in mind that j = ℓ + 1/2, and performing the summation over f and ℓ according to (III.5), we have The anomalous potential is thus given by Let us notice that for Ω = 0, the result for V anomal coincides with the anomalous part of the effective potential presented in [12].According to the above results, after an appropriate cutoff regularization, the LLL contribution to the effective potential V µ,Ω is thus given by where V f,ℓ is given in (III.21) and V anomal in (III.24). Let us now consider the HLL contribution to the effective potential, V n>0 µ,Ω from (III.19).We simplify this expression step by step for our numerical purposes.Using the definition of Plugging this expression into V n>0 µ,Ω from (III.19), it is given by (III.28) with which results from θ (Λ ′ − E + ), and Here, µ f for f = {u, d} are defined by They arise by an appropriate redefinition of the summation over ℓ for two different flavors in (III.5).After performing the summation over k in (III.28),we arrive finally at where K f,ϵ,n (p z ) is a lengthy conditional expression, presented in Appendix B 2. To summarize, V eff at zero temperature and finite density is thus given by In what follows, we use this effective potential to study the effect of rotation on a dense and magnetized quark matter in the MDCDW phase. IV. NUMERICAL RESULTS As it is shown in the previous section, the effective potential of a MDCDW medium at zero temperature, finite density, and in the presence of rotation is given by V The regularized expression for V T =0 , presented in (III.14),does not depend on µ and Ω, but depends through E n,Ω=0 f on m and b.These are taken as two dynamical variables in this context.The part of the potential including µ and Ω is separated into two parts: The LLL contribution V n=0 µ,Ω , presented in (III.25)includes, in particular, the anomalous potential V anomal from (III.24).This part, together with the HLL contribution V n>0 µ,Ω , presented in (III.32)depend explicitly on m and b.To study the effect of µ, eB and Ω on the formation of these two dynamical variables, we have to determine the global minima of V eff with respect to m and b.To do this, we performed a numerical computation in the regime µ ∈ [0, 0.8] GeV, √ eB ∈ [0, 0.8] GeV, corresponding to eB ∈ [0, 0.64] GeV 2 , and the linear velocity ΩR ∈ [0, 0.1].To convert the values of Ω and eB to Hertz (Hz) and Gauss (G), we use 1 GeV= 1.52 × 10 24 Hz and eB = 1 GeV 2 corresponding to B ∼ 1.7 × 10 20 G. The main purpose of the present paper, is to answer the question of whether the MDCDW phase survives the extreme conditions of cold neutron stars.In particular, what is the effect of rotation on the formation/suppression of this phase in these extreme conditions?In this regard, let us notice that the maximum radius of a neutron star R max ∼ 10 km, its maximum angular velocity Ω max ∼ 10 3 Hz [24].These lead to a maximum linear velocity (ΩR) max ∼ 3 × 10 −2 c.As concerns the magnetic field of a neutron star, its maximum value is B max ∼ 10 17 G [31], which corresponds to √ eB ∼ 0.02 GeV.The chosen intervals for µ, eB, and ΩR are thus quite relevant to study the effect of rotation on the MD-CDW phase in cold, magnetized and rotating neutron stars.In what follows, we work with dimensionless quantities.To do this, all dimensionful variables are rescaled with Λ = 1 GeV.We work, as in [24], with R = 10 3 Λ −1 , and set, whenever necessary GΛ 2 = 6 [18], and Λ ′ = 10Λ [18]. In Sec.IV A, we present our numerical results for the µ, √ eB, and ΩR dependencies of m and b for different fixed values of ΩR, √ eB, and µ.In Sec.IV B, we then present the corresponding phase portraits to √ eBµ, √ eB-ΩR, and µ-ΩR. A. The effects of µ, eB and Ω on the chiral condensate m and spatial modulation b The µ dependence In Fig. 1, the µ dependence of the chiral condensate m (solid orange) and spatial modulation b (dashed blue) is plotted for fixed values of ΩR and √ eB.In each row ΩR is constant and √ eB varies.In contrast, in each column √ eB is constant and ΩR varies.The first to fourth rows correspond to ΩR = 0, 10 −3 , 5 × 10 −3 and 10 −2 , and the first to third columns correspond to √ eB = 0.15Λ, 0.25Λ and 0.45Λ.Let us first compare the plots in the first row.The result from Fig. 1(a) is comparable with the result presented in [22].The four different regions described in [22] can also be identified in this plot.In the first region µ ≲ 0.45Λ, the chiral symmetry is broken, whereas the system is spatially symmetric (b = 0).In the second region, 0.45Λ ≲ µ ≲ 0.6Λ, m decreases, whereas b increases.The remnant mass is visible in a third region, 0.6Λ ≲ µ ≲ 0.75Λ.The spatial modulation b is nonvanishing and rather large in this regime.A fourth region is visible in µ ≳ 0.75Λ.In this regime m and b both increase with increasing µ. 3 These four regimes 1(d), 1(g), and 1(j)], where √ eB = 0.15Λ is constant, and ΩR increases.Here, ΩR merely affects the production of m.It decreases (vanishes) by increasing ΩR.As it turns out, the remnant mass appearing in the third region, 0.6Λ ≲ µ ≲ 0.75Λ, disappears.Consequently, b also vanishes in this regime.This specific feature of ΩR in destroying the dynamical mass is expected from [24,25].As concerns the fourth regime, however, it turns out that the system reenters from a chirally symmetric homogeneous regime into a chiral symmetry broken phase for µ ≳ 0.75Λ, where the spatial symmetry is also broken by a nonvanishing and rather large b. By increasing √ eB and for small values of ΩR, the remnant mass in the third region increases.This is expected from the magnetic catalysis, and is visible by comparing the plots in the second column [Figs.1(b), 1(e), 1(h)], corresponding to √ eB = 0.25Λ with the ones in the third column [Figs.1(c), 1(f), 1(i)], corresponding to √ eB = 0.45Λ.Here, as it turns out, b increases also in the first region µ ≲ 0.4Λ.However, once ΩR is large enough, m vanishes in the third region because of the IMRC [25].Larger ΩR destroys the dynamical mass even in the fourth regime µ ≳ 0.75Λ [see Fig. 1(ℓ)] with ΩR = 10 −2 and √ eB = 0.45Λ.Hence, large enough ΩR does not allow any reentrance in the symmetry broken phase in the fourth regime µ ≳ 0.75Λ.In IV B, we study the effect of eB and ΩR on the critical values of µ. 2(c)].We conclude therefore that even for µ = 0, a certain interplay between eB and ΩR leads to a finite b.This is indeed expected because of the similarity between µ and ΩR discussed in [24]. For larger values of ΩR, the rotation completely destroys m, and for vanishing m, the spatial modulation b also vanishes [Figs.2(d To study the effect of µ on the above scenario, the eB dependence of the chiral condensate m (solid orange) and spatial modulation b (dashed blue) is plotted for fixed values of ΩR and nonvanishing µ in Fig. 3.In each row ΩR is constant and µ varies.In contrast, in each column µ is constant and ΩR varies.The first to fourth rows correspond to ΩR = 0, 10 −3 , 5 × 10 −3 and 10 −2 , and the first to third columns correspond to µ = 0.2Λ, 0.4Λ, and 0.6Λ.It is possible to compare the results presented in Figs.3(a)-3(c) with those from Fig. 2 of [18], where ΩR = 0. To do this, one shall, however, be very cau-tious because, according to [22] and also our results from Fig. 1, the small remnant mass at intermediate µ is not considered in [18].Hence, by comparing, e.g.Fig. 3(c not considered in Fig. 2(d) of [18] in the same regime of √ eB ≤ 0.2Λ.Let us emphasize that the result from Fig. 3(c) is consistent with Fig. 1(a) for ΩR = 0, which is by itself in complete agreement with the result presented in [22]. The plots presented in the first column of Fig. 3 [Figs.For moderate values of µ = 0.4Λ and ΩR = 0, 10 −3 , the scenario is similar to the case of µ = 0.2Λ with the same ΩR's [compare Figs.3(b) and 3(e) with 3(a) and 3(d)].For ΩR = 5 × 10 −3 and µ = 0.4Λ, however, m decreases with increasing eB, whereas b increases rapidly with increasing eB [see Fig. 3(h)].The fact that m decreases for strong magnetic fields, moderate µ, and relatively large ΩR is related to the IMRC effect [25].This effect becomes more apparent for larger values of ΩR = 0.01, which is large enough to destroy m, and consequently b, in the regime √ eB ≳ 0.4Λ [see Fig. 3(k)].This scenario completely changes for larger values of µ (see the plots in the third column of Fig. 3).Here, for µ = 0.6Λ, b is larger than m, as long as ΩR ≲ 0.005 [see 3(c), 3(f), and 3(i)].For large value of ΩR = 0.01, both m and b vanish in the whole interval √ eB ∈ [0.15Λ, 0.8Λ] [see Fig. 3(ℓ)].This result coincides with the results presented in Figs.1(j)-1(ℓ).Comparing the plots from Figs. 3(j)-3(ℓ), it turns out that for large Ω = 0.01, the critical value of eB decreases with increasing µ.This is expected from the IMRC effect [25], and is completely visualized in these plots. The ΩR dependence In Fig. 4, the ΩR dependence of the chiral condensate m (solid orange) and spatial modulation b (dashed blue) is plotted for fixed values of µ and √ eB.In each row µ is constant and √ eB varies.In contrast, in each column √ eB is constant and µ varies.The first to fourth rows correspond to µ = 0, 0.2Λ, 0.4Λ, 0.6Λ, and the first to third columns correspond to √ eB = 0.15Λ, 0.25Λ and 0.45Λ.We notice that in contrast to previous figures, the ranges of the horizontal axes are not the same in all the plots from Fig. 4. Let us first compare the plots in each row: By comparing the plots from the first to third row [Figs.4(a)-4(c), 4(d)-4(f), 4(g)-4(i)] together, it turns out that the interplay between the IMRC effect arising from a simultaneous increase of eB and ΩR leads to decreasing the critical value of ΩR with increasing eB.As concerns Figs.4(j)-4(ℓ) in the fourth row, however, large value of µ = 0.6Λ leads to b > m, in contrast to all the other plots.In Fig. 4(j), a remnant mass (m ̸ = 0) in the regime ΩR ≲ 0.002, similar to that which appeared in the third region of Fig. 1 In this section, we present the phase portraits √ eB-µ for fixed ΩR, √ eB-ΩR for fixed µ, and µ-ΩR for fixed √ eB, and demonstrate the impact of the IMRC effect on the phase diagrams of this model, as well as similarities/differences between µ and ΩR in creating the MD-CDW phase. In Fig. 5, the phase portrait √ eB/Λ-µ/Λ for fixed ΩR = 0, 10 −3 , 5 × 10 −3 , 10 −2 , 5 × 10 −2 is plotted.The red solid lines separate two different regimes µ < m (denoted by "1") and µ > m (denoted by "2") in the MD-CDW phase, orange solid and green dashed lines separate the MDCDW phase with µ < m and µ > m from the symmetry restored phase with m = 0 (this normal phase is denoted by "0").They correspond to first-and secondorder phase transition lines, respectively.Figure 5(a) is comparable with the results presented in [18], where only two regimes µ < m and µ > m appear in the whole pa-rameter space.For ΩR = 10 −3 , the parameter space is slightly different: A very small region of the normal phase appears in the interval µ∈[∼ 0.63Λ, ∼ 0.67Λ] for small √ eB ≲ 0.152Λ [see the sub-figure inserted into Fig.5(b)].The phase transition from the µ > m regime of the MDCDW phase to the normal phase is of secondorder.This symmetry restored phase corresponds to the third region, which is demonstrated in Fig. 1(d).The √ eB/Λ-µ/Λ phase diagram for ΩR = 0.005 in Fig. 5(c) becomes more complex.Apart from two regimes µ < m and µ > m that are indicated by the red solid line, and the orange solid and green dashed lines that separate the MDCDW phase with µ < m and µ > m from the normal phase, there is another region in the upper right corner of the phase portrait which is separated by a magenta dotted-dashed line from the µ > m regime of the MDCDW phase.In this region, b is larger than Λ = 1 GeV, the cutoff of the model.This regime shall thus be completely excluded from the phase portrait.The black circle at (µ, √ eB) ∼ (0.115Λ, 0.795Λ) demonstrates the position of a critical point at the end of a second-order transition line. 4n Fig. 5(d) for ΩR = 0.01, the phase portrait includes two regimes µ < m and µ > m in the MD-CDW phase.Because of the small µ > m regime in the lower right corner of the phase portrait, for √ eB ≲ 0.3Λ, a reentrance from the normal phase into the MD-CDW phase occurs for µ ≳ 0.75Λ.For larger values of ΩRs, the normal phase becomes larger and the MD-CDW phase becomes suppressed.As it is demonstrated in Fig. 5(e), for ΩR = 0.05 the MDCDW phase remains only in the left corner of the parameter space, in the interval µ ≲ 0.4Λ and √ eB ≲ 0.25Λ.In both cases, the critical eB decreases with increasing µ.All these effects are the manifestation of the effect induced by an interplay between ΩR and eB.Apart from this aspect, a comparison between Figs. 5(d) and 5(e), the µ > m regime in the MDCDW phase becomes also very narrow (see the region between the solid red and dashed green lines in these figures), and a critical point appears in (µ, √ eB) ∼ (0.03Λ, 0.534Λ) for ΩR = 0.01 in Fig. 5(d), which is then shifted to (µ, √ eB) ∼ (0.069Λ, 0.243Λ) in Fig. 5(e) for ΩR = 0.05.Black circles in Figs.5(d) and 5(e) demonstrate the position of these critical points. In Fig. 6, the phase portrait √ eB/Λ-ΩR for fixed µ/Λ = 0, 0.2, 0.4, 0.6 is plotted.In Fig. 6(a), for µ = 0 only the m > µ regime of the MDCDW phase appears in the left corner of the parameter space.As it is shown, the critical value of eB decreases with increasing ΩR.This is an indication of the IMRC effect.The transition from the µ < m regime of the MDCDW phase to the normal phase is of the first order.This confirms the results from Fig. 2, where, in particular, a first order phase transition occurs for ΩR ≳ 0.005 in Figs.2(c)-2(e).By increasing µ, an extremely narrow region with µ > m appears in the phase diagram.The red solid line in Fig. 6(b) separates two different regimes µ < m and µ > m in the MDCDW phase, and the green dashed line separates the MDCDW phase with µ > m from the normal phase.In Fig. 6(c), by increasing µ to µ = 0.4Λ, the MDCDW phase becomes further suppressed, so that it completely disappears for ΩR ≳ 0.06.For µ = 0.6Λ, the µ < m regime of MDCDW phase completely disappears from the parameter space, and its µ > m regime is shifted to the left corner, where the critical eB does not significantly changes in terms of ΩR [see Fig. 6(d)].This shows the interplay between eB, ΩR and µ in destroying the MDCDW phase through the IMRC effect Finally, in Fig. 7, the phase portrait µ/Λ-ΩR for fixed √ eB/Λ = 0.15, 0.25, 0.45 is plotted.The red solid lines separate two different regimes µ < m (denoted by "1") and µ > m (denoted by "2") in the MDCDW phase, and the green dashed lines separate the MDCDW phase with µ > m from the normal phase (denoted by "0").As it is shown in Fig. 7(a), for small values of √ eB = 0.25Λ, the MDCDW phase exists even for large values of ΩR ∼ 0.1.The critical µ decreases with increasing ΩR [green dashed line in Fig. 7(a)].For larger values of √ eB = 0.45Λ, the MDCDW phase is completely shifted to the left corner of the parameter space.The µ > m regime in this phase is very narrow.A transition to the normal phase is only possible through this regime.The critical µ does not depend significantly on ΩR.Let us remind that all green dashed lines correspond to the second-order phase transitions. V. CONCLUDING REMARKS Starting with the Lagrangian density of a two-flavor gauged NJL model, and assuming a rigid rotation about a certain axis with a constant angular velocity Ω and a magnetic field eB aligned in the same direction, we defined two inhomogeneous condensates in order to introduce the MDCDW phase in a rotating, dense, and magnetized cold quark matter.The question was whether this phase survives the extreme conditions prevailing, e.g., in a neutron star.In contrast to [22], the temperature and its possible interplay with high densities, constant magnetic fields and angular velocities does not play any role in the present paper.We consider a low temperature limit and showed that the temperature dependent thermodynamic potential vanishes in this limit.Taking this limit is justified by the assumption that the neutron star rotates with a constant angular velocity [23]. We first determined the energy eigenvalues of the model, and showed that in comparison with the nonrotating case, they are shifted by a term Ωj, where j = ℓ+1/2 is the corresponding quantum number to J z = L z + Σ z /2 (see Sec. II for more details).We then determined in Sec.III, the corresponding thermodynamic potential to this model at finite temperature T , chemical potential µ, and magnetic field eB.We showed that it consists of three parts, a vacuum part V T =0 , an µ, Ω dependent part V µ,Ω , and a temperature dependent part V T ̸ =0 [see (III.12)].The latter vanishes by taking the limit T → 0. We thus focused on V T =0 and V µ,Ω .As it is argued, V T =0 is independent of Ω.To regularize it, we used the same proper-time regularization scheme as in [18,22].This gives us the possibility to compare our numerical results for Ω = 0 with the results presented in these two papers.The final expression for V T =0 is given in (III.16), where a summation over Landau levels is already performed.As concerns V µ,Ω from (III.17), we first separated it into two parts: The LLL and HLL parts, that are given in (III.18) and (III.19).To derive the LLL part, we followed step by step the same regularization method as is described in the Appendix of [18].We showed, in particular, that the LLL part of the potential includes an anomalous term proportional to bΩ, apart from a term proportional to bµ, which appears originally in [18] (see (III.24)).This term appears as a consequence of the asymmetry appearing in the energy dispersion relations of the lowest and higher Landau levels and, according to [12], leads to important topological effects (see below for further discussions).Another difference comparing to the nonrotating case from [18], appears in the HLL part of the effective potential, whose final result which is given in (III.32)depends explicitly on Ω. Minimizing the thermodynamic potential with respect to two dynamical variables m and b, the chiral and spatial modulation condensates, we studied in Sec.IV, the µ, eB, and ΩR dependence of these two gaps, separately.We showed, that at µ = 0, nonvanishing ΩR acts as an additional chemical potential. 5For small ΩR up to ΩR ∼ 0.005, the rotation enhances the production of b and thus the MDCDW phase, even for vanishing µ.Once ΩR and eB become larger, b vanishes because of the IMRC effect and the resulting suppression of m (see Fig. 2).For nonvanishing µ, the situation is worse.Here, the IMRC effect is amplified because of the similar roles played by ΩR and µ.They act as inverse magnetic catalyzers (magnetic inhibitors).According to the results from Fig. 4, we conclude that, because of the IMRC effect, the MDCDW phase is suppressed in the regimes of eB, µ, and ΩR that are relevant for cold neutron stars. In the second part of Sec.IV, we further explored the eB-µ, eB-ΩR, and µ-ΩR phase portraits.We showed that, as expected from [18], the MDCDW phase consists of two regimes: µ < m and µ > m.The transition from the first regime µ < m to the normal (chiral symmetry restored) phase is of the first order, whereas the transition from the second regime µ > m to the normal phase is of the second-order.The results from this part confirm the conclusion concerning the role of IMRC effect on destroying the MDCDW phase in regimes that are relevant for the cold neutron stars.The fact that the critical magnetic field decreases with increasing ΩR (µ) for fixed µ (ΩR) is a direct consequence of the IMRC effect. Because of the aforementioned similarity between µ and ΩR, the MDCDW phase is created even at µ = 0 for ΩR > 0. This is in particular demonstrated in the plots of Fig. 2 and the phase portrait in Fig. 6(a).According to these results, the critical value of eB decreases with increasing ΩR for µ ≤ 0.4Λ.This opens the possibility for the MDCDW phase to be created during the early stages of heavy ion collisions, where µ ∼ 0, √ eB ∼ 0.1 − 0.5 GeV (eB ∼ 5−15m 2 π with the pion mass m π ∼ 140 MeV) [32], and ΩR ∼ 0.1 [33].Here, however, it is necessary to consider the effect of finite temperature and its interplay with the magnetic field and rotation.First results in this direction is recently presented in [34], where the effect of magnetic fields is neglected.In [35], the effects of rotation on confining properties of compact electrodynamics in two spatial dimensions is studied.It is, in particular, shown that at finite temperature the phase diagram of a uniformly rotating system possesses, in addition to a confining phase at low temperature and a deconfining phase at high temperature, a mixed inhomogenous phase at intermediate temperatures.The latter has a confining region at the core and a deconfining region at the edge of the rotating system.It would be interesting to extend the present work in this direction. As aforementioned, in this paper, we focused on the MDCDW phase in rotating cold quark matter.This phase is characterized by a certain asymmetry in its LLL spectrum.As it is shown in [19], this asymmetry which occurs also in the nonrotating case, arises by the interplay between the inhomogeneous condensate and the magnetic field.It is also shown to be the origin of the nontrivial topological properties of this model, leading, in particular, to the creation of an anomalous, nondissipative electric Hall current in the corresponding Maxwell equations.As it is argued in [19], in the nonrotating case, the anomalous baryon (quark) number density ρ B an associated with this anomalous four-current is related to the regularized Atiyah-Parodi-Singer (APS) index η B [36] via Here, b is the spatial modulation in the MDCDW phase and N c the number of colors.According to [19], another way to determine ρ B an is by appropriately regularizing the anomalous part of the thermodynamic potential corresponding to the MDCDW phase in a nonrotating phase [18].According to [18], an µ, with baryonic charge density given in (V.1).As concerns the rotating case, we determined in Sec.III, the anomalous part of the thermodynamic potential of the MDCDW model in a rotating medium.It is given in (III.24) and leads to where ρ B an is given in (V.1), and ρ Ω an reads with R the radius of the rotating cylinder and the magnetic length L f,B ≡ \|q f eB\| −1/2 .Let us notice that ρ Ω an arises from the summation over a flavor dependent ℓ in (III.22)[for the limits of ℓ, see (III.5)],where Ωj played the role of a chemical potential associated with rotation [24].In analogy to ρ B an , ρ Ω an from (V.2) leads immediately to anomalous electric charge density J 0 Ω ≡ f eq f ρ Ω an,f associated to ρ Ω an,f for each flavor, which may appear in the corresponding Maxwell equation as its baryonic counterpart J 0 B ≡ f eq f ρ B an,f in [19].It would be interesting to find the relation between ρ Ω an and the APS index η via an appropriate regularization of η ≡ lim We postpone this computation to our future works.From (A.5), we obtain and As it turns out, the above matrix N f commutes with the diagonal matrix with (A.12) This help us to determine the eigenfunction ψ f in (A.8). Hence, the problem is reduced to determining the eigenfunction ψ f and the eigenvalue λ f in Applying N f from (A.11) to ψ f from (A.3), we arrive at two differential equations for f i (ρ), i = 1, • • • , 4, is a linear combination of a hypergeometric function of the first and second kind 1 F 1 (a; b; z) and U (a; b; z). 6Requiring that g ∓ (x) are regular at x → 0, g ∓ (x) is given by g ∓ (x) = A ∓1 F 1 (−κ ∓ ; \|ℓ ∓ \| + 1, x). (A.20) 6 Similar results are also found in [24,25,30]. Plugging this result into (A.15),we thus arrive at Assuming, at this stage, that the fermionic system has no spatial boundary condition, the hypergeometric function can be replaced by the associated Laguerre polynomial, .24).At this stage, we shall determine κ ∓ .To do this, we use the fact that the lower index κ ± in the Laguerre polynomial shall be positive.According to (A.17), depending on ℓ, this leads for the two cases of positive and negative q f , or alternatively positive and negative s f , to different allowed values for κ ∓ .The results for ℓ ≤ −1 and ℓ ≥ 0 are summarized in Tables I and II (see also [25] for more details). TABLE I.The allowed values for κs, s ≡ ± for ℓ ≤ −1, ℓ ≥ 0, and positive and negative charges, according to (A.17 2. The final expression for K f,ϵ,n (pz) in (III.32) In this section, we present the final expression for K f,ϵ,n (p z ) defined by where J f,ϵ,n,k (p z ) is defined in (III.30).We perform the summation over k in (B.7) using an appropriate Mathematica program and arrive at the following conditional expression: with and the conditions In the above expressions, E ±,0 are defined by Here, E + and µ f for f = {u, d} are defined in (III.26) and (III.31). ) and 2(e)].For ΩR ≳ 0.005, the transition from the MDCDW phase with m, b ̸ = 0 to the symmetry restored phase with m = 0 is of first order [see Fig. 6(a)].By comparing the plots from Figs. 2(c)-2(e), it turns out that the critical value of eB, for which m and b vanish, decreases with increasing ΩR.This result coincides with the results presented in the phase diagram eB-ΩR in Fig. 6. and 2 ( d)].The critical value of eB, for which m and b vanish, is, however, smaller for nonvanishing µ. FIG. 6. color online.The phase portrait √ eB/Λ vs. ΩR for fixed µ/Λ = 0, 0.2, 0.4, 0.6.The red solid lines separate two different regimes µ < m (denoted by "1") and µ > m (denoted by "2") in the MDCDW phase, and the orange solid and green dashed lines separate the MDCDW phase with µ < m and µ > m from the symmetry restored phase with m = 0 (denoted by "0").The orange solid line in panel (a) and the green dashed lines in panels (b)-(d) correspond to first-and second-order phase transitions.The regime denoted by "Sat", which is separated from the µ > m regime of the MDCDW phase by a dotted-dashed magenta line, shall be excluded from the phase diagram (see the description in text).	2023-06-08T01:16:01.498Z	2023-06-07T00:00:00.000	{ "year": 2023, "sha1": "590a5959703923618ad7838795bb8c54666b7951", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "590a5959703923618ad7838795bb8c54666b7951", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
234084182	pes2o/s2orc	v3-fos-license	Application of Case based Teaching Method in Computer Aided Design Teaching of Art and Design Case teaching method (hereinafter referred to as TM) is an effective TM to cultivate students’ ability and improve their quality. The 21st century is the information age, computer-aided art design has become the latest means of artistic design expression. Through computer-aided technology, designers can express themselves through a new form of artistic expression, which brings profound changes to the aesthetics and thinking of art design. Therefore, we should integrate science and technology with artistic connotation, which will promote the sustainable development of computer-aided art design. Case TM is based on the needs of teaching objectives and content to select a typical example of a TM, which can better carry out art design teaching. Introduction With the rapid development of computer aided design (hereinafter referred to as CAD) software, its function is more and more powerful, which has been widely used in various fields, such as industrial design, web design, environmental art design, advertising design, film and television design, etc. In modern art design, CAD is an important part of modern design art, which is an important means of visual communication [1] . CAD provides a new carrier for art creation and art design, which provides a new form and space for artistic expression. Case TM is a typical example selected according to the needs of teaching objectives and contents. Through analysis and summary, we can adapt and process into teaching cases, which will guide students to participate in thinking, analysis, discussion, expression and other activities [2] . Through case TM, students can cultivate their comprehensive ability in specific problem situations. Case TM can increase students' interest in learning, which helps to cultivate students' comprehensive ability. By integrating theory with practice, we can reflect the practicality of the subject, which is conducive to promoting students to learn to communicate and cooperate [3] . CAD software reflects the progress and diversification of art design means, which is the basic skill to learn art design. CAD is a kind of technology, which is a method and means to make the design scheme as a result [4] . Commonly used CAD software mainly includes: photo0shop, flash, CorelDRAW, AutoCAD, 3dsmax, Dream weaver, fireworks, etc., which is suitable for various art design majors. In practical learning, many students can master the application method of software, but they do not understand how to use software to complete various high-quality design tasks. Case TM is more suitable for software teaching, which can make students have a certain understanding of the comprehensive use of software in a short time. Through case teaching, we cultivate students' design consciousness [5] . Characteristics of heuristic teaching Compared with the traditional classroom teaching, case teaching breaks the traditional classroom teaching mode of theoretical knowledge education through typical case analysis, which is conducive to the change of students' innovative thinking and the improvement of innovative skills. For a long time, the theoretical knowledge system of art and Design Majors in China is relatively perfect, and the skills and level of theoretical level are higher. However, in the actual operation process, most students lack practical ability. Through the effective use of case TM in art design teaching, schools can change the roles of traditional classroom teachers and students, which can help students improve their enlightening thinking consciousness and design ability. Through heuristic teaching, the school can achieve the goal of training practical operation skills [6] . Characteristics of Participatory Teaching In the process of art design teaching, we not only need to improve students' theoretical knowledge system, but also need to cultivate students' practical operation ability. Therefore, in classroom teaching, teachers must combine theoretical knowledge with practical operation closely. Through case TM, we can improve students' ability of theoretical connection and innovation level. Through the analysis of typical cases, students can use theoretical knowledge to participate in case analysis, which can improve the enthusiasm of students to participate. Through the consolidation of theoretical knowledge, we can finally solve practical problems. Characteristics of democratic teaching Different from the traditional theory teaching classroom, the case TM establishes the position of teacher leading and student-centered in art design teaching, which can also promote the close cooperation between students and teachers. In art design teaching, students are divided into groups to analyze typical cases. Through the case study, we can clarify their roles and discuss freely. By putting forward opinions and suggestions, students sort out ideas and consolidate theoretical knowledge, which can strengthen practical skills and realize democratic teaching. Realization of interactive art works The creation method is the product of history, which has the profound brand of the times and society. The interactive art works contain the common rules in artistic creation, which may be adopted by artists of different times, different societies and different nationalities. Art works are different from products, which will lead to the complexity of work design and weaken the creativity and originality of artists. The interaction design process model is the Royce waterfall model, as shown in Figure 1. Desktop Virtual Reality Art Design In terms of software, Superscape VRT is a representative product of desktop virtual reality art design launched by dimension international. It contains three main editors, namely shape editor, world editor and visualizer. Among them, shape editor provides interactive tools for constructing polygon objects in virtual environment, world editor is used to connect these objects with their motion or behavior in virtual environment, and visualizer provides performance stage of virtual environment. Through Superscape VRT software, we can carry out interior decoration design, as shown in Figure 2. Network Virtual Reality Art Design At present, the famous network virtual reality art design software has Cult3D, Anark and so on. Cult3 D is a product of Cycore Company, which is the establishment of interactive 3D objects on the network. Cult3D can display the interaction of different events and functions in a visual way, which can control the rotation, movement, zooming and shrinking of 3D products on the Internet. The latest version of Cult3D can also add sound and operation wizard to objects, which can increase the richness of the presentation. The biggest advantage of Cult3D is the small amount of files, which can retain the beautiful texture and texture. With Cult3D software, we can create virtual space, as shown in Figure 3. Case design and selection The key to implement case teaching is to select typical cases. Therefore, we need to use some representative cases, which needs to choose the best case from the perspective of teaching. In addition, the design and selection of cases must be difficult, which can stimulate the enthusiasm. In the selection of cases, the selected cases must be typical, authentic, targeted, interesting, educational and epochal. For example, in the late comprehensive design of "web design and production", students need to give a practical topic, such as making personal job search website. Teaching case design can stimulate interest. Therefore, more practical topics can enable students to quickly enter the role of development. Through the combination of web design and production knowledge, we can improve the interest in theoretical knowledge. In order to let the students master the knowledge better, we should explain the relevant knowledge before the case study. According to the students' actual level and acceptance ability, teachers can gradually and inductively exercise students' ability. Case analysis and discussion Case analysis and discussion is the most important step in case teaching, which is to organize students to analyze and discuss cases. Case discussion is usually conducted in groups. After the discussion, representatives are elected to speak and other students supplement. When students discuss, teachers should be able to listen and be good at guiding, which requires full affirmation of correct understanding and original opinions. At the same time, teachers should guide and correct the vague wrong understanding, which will guide students step by step towards the established goal of case analysis. In the course of Photoshop graphic design, the making of static logo is an important part. When introducing the characteristics of static logo images, teachers can prepare and display some typical static logo pictures in advance. By observing the characteristics of the pictures, students can discuss in groups and choose representatives to speak. Through observation and discussion, students can summarize the characteristics of these pictures. After understanding the characteristics of the static logo image, students need to submit the specific scheme of logo image production. In the process of analysis and design, teachers should constantly prompt and guide, which can make each student produce pictures with both characteristics and requirements. Summarize the case After the case discussion, the teacher should summarize the case first. Teachers should not simply draw conclusions on the results of the case discussion, but should evaluate the whole discussion. By pointing out the theoretical problems of the case and the lack of discussion, teachers can further guide them to think deeply. If necessary, teachers should make more detailed and in-depth analysis and discussion of cases, which will help students expand their thinking and broaden their horizons. Through the deep understanding of the case, students find out their strengths and weaknesses by combining their own discussion and teachers' summary. Case TM is an advanced TM. However, Case TM can't completely replace other TMs. On the one hand, the case TM can't completely replace the traditional theory TM. The advantages of case TM in cultivating students' ability are obvious, but the traditional theory TM also has its own advantages. The case TM is not antagonistic to audio-visual teaching and on-the-spot observation, and each of them has its own strong points to achieve the training goal. Therefore, in the actual teaching process, we should optimize the classroom TMs. According to the syllabus, teaching content and the actual situation of students, teachers should choose and use different TMs, which can really improve the effect of classroom teaching. By improving the teaching quality, the school can cultivate new high-quality application-oriented talents in line with the development of the times. Conclusion CAD course provides systematic and professional education for art and design industry. Through the case TM, we can improve the students' ability of autonomous learning and exploration by using computers, which will help teachers change their educational views. Case TM reverses the roles of teachers and students in the classroom, which emphasizes that students are the main body of teaching activities. Therefore, teachers must mobilize the enthusiasm of students, which is the embodiment of the transformation from traditional teaching to modern teaching.	2021-05-10T00:03:29.323Z	2021-02-01T00:00:00.000	{ "year": 2021, "sha1": "3e89f550e86a43e37e98b68da59a15787c6e297f", "oa_license": null, "oa_url": "https://doi.org/10.1088/1742-6596/1744/3/032095", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "b2552c7ddcaa954f0154d78af9b89b05b5dbf911", "s2fieldsofstudy": [ "Art" ], "extfieldsofstudy": [ "Physics", "Computer Science" ] }
256172593	pes2o/s2orc	v3-fos-license	Cannabidiol Modulates Alterations in PFC microRNAs in a Rat Model of Depression Cannabidiol (CBD) is a potential antidepressant agent. We examined the association between the antidepressant effects of CBD and alterations in brain microRNAs in the unpredictable chronic mild stress (UCMS) model for depression. UCMS male rats were injected with vehicle or CBD (10 mg/kg) and tested for immobility time in the forced swim test. Alterations in miRNAs (miR16, miR124, miR135a) and genes that encode for the 5HT1a receptor, the serotonergic transporter SERT, β-catenin, and CB1 were examined. UCMS increased immobility time in a forced swim test (i.e., depressive-like behavior) and altered the expression of miRNAs and mRNA in the ventromedial prefrontal cortex (vmPFC), raphe nucleus, and nucleus accumbens. Importantly, CBD restored UCMS-induced upregulation in miR-16 and miR-135 in the vmPFC as well as the increase in immobility time. CBD also restored the UCMS-induced decrease in htr1a, the gene that encodes for the serotonergic 5HT1a receptor; using a pharmacological approach, we found that the 5HT1a receptor antagonist WAY100135 blocked the antidepressant-like effect of CBD on immobility time. Our findings suggest that the antidepressant effects of CBD in a rat model for depression are associated with alterations in miR-16 and miR-135 in the vmPFC and are mediated by the 5HT1a receptor. Introduction Depression is one of the most common psychiatric disorders worldwide [1]. Antidepressants are the current recommended standard of treatment for depression; however, their effectiveness is only slightly efficacious compared to placebos [2]. There has been growing evidence that cannabidiol (CBD) may have therapeutic effects on depressive symptoms. Given its safety profile, CBD is a promising treatment for mood disorders; however, the exact molecular mechanisms underlying its potential antidepressant effects are largely unknown. Pre-clinical studies demonstrated the antidepressant effects of CBD, expressed as lower immobility time in the forced swimming test (FST) (i.e., less despair) [3][4][5] and higher saccharine consumption in the saccharine preference test (i.e., lower anhedonia) [3,6]. In mice that underwent olfactory bulbectomy (OBX), a model for depression, CBD improved depressive-like symptoms and elevated serotonin and glutamate levels in the prefrontal cortex (PFC) [7]. Human studies suggest that CBD has ameliorating effects in several disorders, which are in high comorbidity with depression, such as insomnia, borderline personality disorder, and social anxiety [8,9]. CBD has a very low affinity for both cannabinoid CB1 and CB2 receptors [10], but it modulates endocannabinoid function through its ability to inhibit the hydrolysis of anandamide and to act as a transient receptor potential vanilloid 1 agonist. Another major mediating pathway for CBD-anti-depressive effects may be through the serotonergic 5-HT1a receptor [4,[11][12][13]. CBD administered into the medial PFC (mPFC) was found to induce antidepressant-like effects in the FST through indirect activation of CB1r and 5-HT1a [13]. Repeated administration of CBD was found to prevent long-lasting anxiogenic effects promoted by a single predator exposure; pretreatment with the 5-HT1a antagonist WAY100635 attenuated the CBD effects, suggesting the involvement of 5-HT1a in the mediation of those effects [11]. CBD activates the extracellular signal-regulated kinase (ERK) pathway through the 5-HT1a receptor [14], resulting in β-catenin accumulation in the cytosol and therefore in the cell nucleus [15]. β-catenin is a multi-functional protein that plays an important role in the mature central nervous system; its dysfunction has been implicated in several neuropsychiatric disorders, including depression [16]. We have recently found, that downregulating β-catenin levels in the nucleus accumbens (NAc), blocked the therapeutic-like effects of the fatty acid amide hydrolase (FAAH) inhibitor URB597 on anxiety-and depression-like behaviors in rats exposed to a rat model of post-traumatic stress disorder (PTSD) [17]. It has been shown that β-catenin is a critical regulator in the development of behavioral resilience, activating a network that includes downstream microRNAs (miRNAs, miRs) [18]. miRNAs are small non-coding RNA molecules comprising 19-25 nucleotides. miRNAs are implicated in a range of psychiatric disorders including anxiety and depression [19][20][21][22][23]. MiR-16 was found to be less abundant in the cerebrospinal fluid of patients with major depression [31] and lower levels of miR-16 in the NAc were associated with susceptibility to stress in mice [18]. Moreover, MiR-16 modulates the expression of the serotonin transporter (SERT), a major target for SSRIs [31,32]. In raphe cells, elevated levels of miR-16 induced a decrease in the expression of SERT [31,32]. We have recently found that in adult males and females, exposed to early life stress (ELS), the FAAH inhibitor URB597 restored an ELS-induced decrease in mPFC miR-135a in females and miR-16 in males and the associated depressive-like phenotype in both sexes [33]. In the PFC, early adolescent stress downregulated miR-135a expression [34], while upregulating in the hippocampus [26,34], suggesting a brain-region-dependent effect. MiR-135a levels were significantly lower in the blood and brain of depressed human patients. MiR-135-knockdown also prompted an increase in 5-HT1a and SERT levels in the raphe nucleus [27]. A similar effect of miR-135a downregulation on 5-HT1a overexpression was seen in the PFC [34]. The aim of the present study was to examine in the unpredictable chronic mild stress (UCMS) model for depression, whether the antidepressant properties of CBD are associated with alterations in miRNAs implicated in depression (miR-16, miR-124, and miR-135) and with important target genes of these miRNAs and CBD. We examined the expression of genes that encode for the serotonergic 5HT1a receptor and SERT, and the expression of genes that encode for β-catenin and CB1. This was examined in the ventromedial PFC (vmPFC), NAc, and raphe nucleus, brain regions highly associated with the etiology of depression [39][40][41]. The Effects of Chronic CBD Administration during UCMS on Behavior We examined the effects of chronic CBD administration (10 mg/kg, i.p.) during the last 3 weeks of a 6-week UCMS model on immobility in the FST and on motoric and anxiety-like behavior in the open field test (OFT). All analyses were conducted using two-way ANOVA [stress × drug (2 × 2)]. Figure 1. The effects of CBD treatment on behavior in rats exposed to UCMS. (a) FST: UCM treated with a vehicle spent more time immobile than the No UCMS groups and UCMS rats t with CBD. (b) OFT total distance: No UCMS rats treated with the vehicle covered less distance pared to all groups. (c) OFT time in the center: no differences between groups were observed. forced swim test; OFT-open field test; UCMS: unpredictable chronic mild stress; CBD: cannab , p < 0.05, , p < 0.01, **, p < 0.001. The Effects of Chronic CBD Administration during UCMS on miRNA Expression Following the behavioral tests, we assessed the expression of miR-16, miR-124 miR-135 in the vmPFC, NAc, and raphe nucleus. All analyses were conducted using way ANOVA [stress × drug (2 × 2)]. The Effects of Chronic CBD Administration during UCMS on miRNA Expression Following the behavioral tests, we assessed the expression of miR-16, miR-124, and miR-135 in the vmPFC, NAc, and raphe nucleus. All analyses were conducted using two-way ANOVA [stress × drug (2 × 2)]. Pearson bivariate correlations tests (Table 1) were conducted between miRNA expression in the different brain regions and the behavioral measures to explore the association between the depressive-like behavior of the rats and their miR expression. For immobility, the most robust effect was observed with vmPFC miR-135 levels (r = 0.428, p < 0.05), suggesting that increased immobility was associated with vmPFC miR-135 upregulation. A negative correlation was observed with NAc miR-135 levels (r = −0.339, p < 0.05). For total distance in the OFT, significant correlations were observed with NAc miR-124 (r = 0.455, p < 0.01), raphe miR-135 (r = 0.474, p < 0.05), and raphe miR-16 (r = −0.473, p < 0.01) levels. These suggest that increased locomotion behavior was associated with increased NAc miR-124 and raphe miR-135 and decreased raphe miR-16. The Effects of Chronic CBD Administration during UCMS on Possible Target Genes Previous findings suggested that the effects of CBD on depressive-like behavior may be mediated via serotonergic mechanisms and CB1r activation [4,[11][12][13]. We have recently shown that the stress-preventing effects of FAAH inhibition are mediated by β-catenin [17]. Hence, we next examined alterations in the expression of several target genes in UCMS rats treated with CBD. We examined the expression of htr1a and slc6a4 genes that encode for the serotonergic 5HT1a receptor and SERT, respectively, and the expression of ctnnb1 and cnr1, the genes that encode for β-catenin and CB1, respectively. All analyses were conducted using twoway ANOVA [stress × drug (2 × 2)]. Pearson bivariate correlations tests were conducted between miRNA expression and genes in the various brain regions. In the NAc (Table 3), the most robust correlations were observed between miR-16 and ctnnb1 (r = 0.479, p < 0.01) and between miR-124 and ctnnb1 (r = 0.463, p < 0.01), suggesting that levels of these microRNAs were associated with the β-catenin gene. Table 2. Pearson correlation coefficients between miRNA levels and genes in the vmPFC in rats exposed to UCMS and CBD. Table 3. Pearson correlation coefficients between miRNA levels and genes in the NAc in rats exposed to UCMS and CBD. In the raphe nucleus (Table 4), the most robust correlations were observed between miR-124 and slc6a4 (r = 0.408, p < 0.05) suggesting that levels of this microRNA were associated with SERT gene. Table 4. Pearson correlation coefficients between miRNA levels and genes in the raphe nucleus in rats exposed to UCMS and CBD. Does the 5-HT1a Antagonist WAY100635 Block the Effects of Chronic CBD Administration during UCMS on Behavior As CBD was found to restore UCMS-induced downregulation of the htr1a gene that encodes for the serotonergic 5HT1a receptor, we pharmacologically examined whether the effects of CBD on behavior are mediated by the activation of the 5HT1a receptor. To that end, in a different set of animals, we assessed whether the antidepressant effects of CBD (see Figure 1) are mediated by the serotonergic 5HT1a receptor by administering a 5HT1a receptor antagonist. We administered the 5HT1a-antagonist WAY100635 (0.1 mg/kg) along with CBD every day during the last 3 weeks of UCMS. Other groups were injected with CBD or WAY or vehicle for comparison. All analyses were conducted using two-way ANOVA [stress × drug (2 × 2)]. Discussion In this study, we show for the first time that CBD can restore UCMS-induced upregulation of miR-16 and miR-135 in the vmPFC as well as the associated despair-like behavior. UCMS also downregulated the 5-HT1a gene htr1a in the vmPFC; using a pharmacological approach with the 5-HT1a receptor antagonist WAY, we found that the antidepressant-like effects of CBD are mediated by the 5HT1a receptor. The antidepressant effects of CBD on despair-like behavior in the FST corroborates with previous findings [3][4][5][42][43][44]. In the open field, CBD did not restore the UCMSinduced increase in locomotion activity and UCMS had no significant effect on anxiety-like behavior measured as time spent in the center of the open field. The expression of miR-16, miR-124, and miR-135 was significantly affected by UCMS, corroborating with previous studies demonstrating that these miRNAs are correlated with anxiety-and depressive-like phenotypes and are significantly downregulated or upregulated following stress exposure, depending upon the brain region studied and the type of stressor [18,23,[26][27][28][29]45,46]. Specifically, we found that UCMS decreased the expression of miR-124 in the NAc and raphe and increased miR-16 in the NAc and the raphe and miR-135 in the raphe. However, these effects were not normalized by CBD treatment. Similarly, UCMS affected the target genes, decreasing slc6a4 and cnr1 expression in the vmPFC (genes coding SERT and CB1), and decreasing ctnnb1 (β-catenin) in the PFC and NAc, with no effect for CBD treatment. Alterations in miRNAs In general, higher serum levels of miR-16 are associated with a resilient phenotype to stress [47], while patients with major depression exhibit lower CSF expression of the same microRNA [30,48]. Nevertheless, these findings are rarely associated with specific brain regions. In mice, social defeat stress was associated with susceptibility to stress and lower accumbal miR-16 [18], while in rats, ELS-induced stress was associated with lower miR-16 in the vmPFC [32]. It has been shown that maternally deprived rats, but not rats exposed to chronic unpredictable stress, showed higher hippocampal miR-16 expression than control rats [26]. Taken together these findings suggest that different stressors differentially affect the expression of miRs in a brain-region-dependent manner. miR-124 In the vmPFC, UCMS rats that were treated with CBD had higher levels of miR-124 only compared to control rats who were treated with CBD. UCMS decreased miR-124 in the NAc and in the raphe nucleus with no effect of CBD treatment, suggesting that its antidepressant effects are mediated by a different mechanism. Previous studies showed that rats that were treated chronically with corticosterone as a model of depression presented higher miR-124 levels in the PFC [49], and vmPFC suppression of miR-124 via lentiviral vector decreased depressive symptoms [50]. Other studies have shown a decrease in miR-124 in the hippocampus following UCMS and that using an agomir to increase miR-124 had an antidepressant-like effect [51]. Similarly, addictive behavior to cocaine was associated with lower NAc levels of this microRNA [52][53][54], thus indicating that miR-124 may play an important role in the reward system. However, in another study, UCMS increased hippocampal miR-124 expression, and downregulation of miR-124 using an antagomir decreased depressive-like behavior [55]. miR-135 Previous findings showed decreased miR-135 in the PFC and raphe of mice exposed to chronic stress [27,33] and decreased PFC miR-135 in rats exposed to early life stress [32,33]. Downregulation of raphe miR-135 was observed following exposure to the chronic social defeat stress model in mice [27]. Alterations in Serotonergic Targets, β-catenin, and CB1 3.2.1. htr1a (5HT1a Gene) We found that the UCMS-induced decrease in the 5HT1a gene in the vmPFC was reversed by CBD. This corroborates with the expectation for lower levels of the 5HT1a gene in regions where miR-135 is high, and vice versa [27]. Importantly, we found that the antidepressant-like effects of CBD were mediated by the 5HT1a receptor, as co-administration of CBD and the 5HT1a antagonist, blocked the therapeutic-like effects of CBD in the FST in UCMS rats. This corroborates with a previous study in which CBD was microinjected into the vmPFC in rats exposed to the FST and the OFT [13]. In the NAc, both UCMS and CBD led to a decrease in the 5HT1a gene, as control rats that were treated with vehicle had higher levels of this gene than all the other groups. In the raphe, UCMS or CBD had no effect on 5HT1a gene expression. slc6a4 (SERT Gene) SERT modulation is the main mechanism on which SSRIs are based [56]. Total SERT knockout results in depressive and stressed behavior [57], but variations in its expression in different brain regions can lead to a more complex effect. Therefore, lower expression of the SERT gene was expected to be observed in UCMS rats; we found a decrease in slc6a4 in the PFC, with no effect of CBD treatment. In the NAc, no differences were observed between the groups; in the raphe nucleus, UCMS-vehicle rats demonstrated increased levels compared to controls who were treated with CBD. As a target of miR-16, levels of the SERT gene were expected to be lower in regions where miR-16 is overexpressed, and vice versa [31,58]. We found that UCMS decreased SERT gene expression in the vmPFC while elevating miR-16. However, even though CBD reversed the effect of UCMS on miR-16 expression in the vmPFC, it did not affect the SERT gene, suggesting other mechanisms which are involved in the regulation of this gene (e.g., miR-15a) [58]. ctnnb1 (β-catenin Gene) We found that UCMS decreased β-catenin gene levels in the vmPFC and NAc, with no effect in the raphe. This is in line with other studies showing decreased expression of β-catenin in the brain and specifically the PFC and NAc [59,60]. CBD did not restore UCMS-induced downregulation of β-catenin, but in the vmPFC, UCMS rats treated with CBD were not different from any of the other groups. Hence, β-catenin is altered following UCMS exposure, and it regulates microRNA expression [18], but our findings did not demonstrate a robust effect of CBD on β-catenin mRNA in UCMS rats. In general, increased β-catenin expression correlates with resilience to stress and depression [18,61], and CBD was shown to target the Wnt/β-catenin pathway [62]. CBD functions as a negative allosteric modulator of CB1, and it has been shown to prevent CB1 internalization [70,71]. In our study, CBD did not restore the UCMS-induced decrease in cnr1 in the vmPFC. Subjects Male Sprague Dawley rats (60 days old) were group-housed at 22 ± 2 • C under 12 h light/dark cycles (lights turned on at 07:00). Rats were allowed water and laboratory rodent chow ad lib, except when the UCMS procedure required deprivation. The experiments were approved by the University of Haifa Ethics and Animal Care Committee, and adequate measures were taken to minimize pain and discomfort (696/20). UCMS Protocol Rats were subjected to a random sequence of mild stressors [72,73] for 6 weeks. These included cage soiling with 300 mL water, group housing, water and/or food deprivation, reversal of light/dark cycle, cage tilting to 45 • , and physical restraint (see supplementary information (SI); Table S1). No UCMS rats were handled and were not subjected to the stress protocol. Pharmacological Agents No UCMS and UCMS exposed rats received daily injections (i.p.) of vehicle, CBD (10 mg/kg), or the 5-HT1a-antagonist WAY100635 (WAY; 0.1 mg/kg; Sigma, St. Louis, MO, USA) during the last 3 weeks of the 6-week UCMS model. Drugs were freshly prepared and administered in 1 mL/kg of vehicle. Rats were injected between 15:00 pm and 17:00 pm, irrespective of the stress schedule. Drugs were dissolved in 2% Tween-80 and 98% saline. Doses were based on previous work [74,75]. Behavioral Tests All tests occurred in a dim light (15-20 lx) and took place between 1300 and 1600 h. Forced Swim Test (FST) Conducted in a cylindrical water container (62 cm diameter, 40 cm height, filled with water at a temperature of 22 • C). The water level was such that the rat could only touch the bottom with the tip of its tail. Rats were exposed to the swim tank for 15 min habituation on the first day and 5 min on the second day. Video films of the second day of each FST session were analyzed for passive coping (immobility). An immobility index was calculated: time spent immobile divided by the total time spent in the arena. Quantitative Real-Time PCR (qRT-PCR) Rats were sacrificed and brain tissues of the vmPFC, NAc, and raphe nucleus were harvested for molecular analysis (see SI, Figure S1). RNA extraction, cDNA preparation, and qRT-PCR were performed as previously described [76] to detect the expression of miRNAs (miR-16, miR-135, miR-124) and mRNAs (htr1a, slc6a4, ctnnb1 and cnr1; genes coding to 5HT1a, SERT, β-catenin, and CB1r, respectively). For miRNA, 500ng of total RNA was reverse transcribed cDNA using qScript microRNA cDNA Synthesis Kit (Quanta Biosciences, Gaithersburg, MD, USA). For mRNA, 1000 ng of total RNA was converted into cDNA using qScript cDNA Synthesis Kit (Quanta Biosciences, Gaithersburg, MD, USA). This was followed by Real-Time SYBR Green qRT-PCR amplification using specific primers (Quanta Biosciences, Gaithersburg, MD, USA) according to the manufacturer's instructions. RT reactions were carried out by a Step One real-time PCR system (Applied Biosystems, Waltham, MA, USA). Fold-change values were calculated using the ddCt method relative to the housekeeping gene hypoxanthine phosphoribosyl transferase RNU6 (miRNA) or HPRT (mRNA). Primers for both miRNAs (miR-16-5p, miR-124-5p, and miR-135a-5p) and mRNAs (see Table 5) were designed and synthesized by Agentek (Tel Aviv, Israel). Primer suitability was determined using standard curve analysis, melting curve analysis, and linearity and threshold. Statistical Analysis The results are expressed as means ± SEM. For statistical analysis, one-way ANOVA, two-way ANOVA, and Pearson bivariate correlation test were used as indicated. All post hoc comparisons were made using Tukey's range test. Significance was set at p ≤ 0.05. Data were analyzed using SPSS 27 (IBM, Chicago, IL, USA). Normality assumption was examined using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Conclusions We show for the first time that CBD can prevent UCMS-induced increases in vmPFC miR-16 and miR-135. The antidepressant effects of CBD in rats exposed to the UCMS model for depression were mediated by the 5HT1a receptor. CBD seems to have positive effects of diminishing depressive-like behaviors with the advantage of not being addictive or having many side effects [77]. However, the mechanisms underlying its therapeutic effects are still not entirely clear and involve multiple targets.	2023-01-24T16:30:56.053Z	2023-01-20T00:00:00.000	{ "year": 2023, "sha1": "b478ef77bfbadf885be96b299e21db71c56ef8da", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1422-0067/24/3/2052/pdf?version=1674876555", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "4720fd87eeb8bacd5d5663181ea90c8e01e72635", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
97121596	pes2o/s2orc	v3-fos-license	Characterization of the high-pressure superconductivity in the Pnma phase of calcium The thermodynamic parameters of the superconducting state in calcium under the pressure at 161 GPa have been calculated within the framework of the Eliashberg approach. It has been shown that the value of the Coulomb pseudopotential is high (0.24) and the critical temperature (25 K) should be determined from the modified Allen-Dynes formula. In addition, it has been found that the basic dimensionless ratios of the thermodynamic parameters significantly diverge from the BCS predictions, and take the following values: (i) The zero temperature energy gap to the critical temperature(R1) is equal to 4.01. (ii) The ratio R2 equals 2.17. (iii) The quantity R3=0.158. Finally, it has been proven that the electron effective mass is large and takes the maximum of 2.32me at TC. I. INTRODUCTION Under the influence of the high pressure, the elemental calcium undergoes a series of structural phase transitions.In particular, one can distinguish seven phases in the range of pressure (p) from 0 to 241 GPa [1], [2], [3] (please see Figure 1 (A) for the details). The two first phases, namely Ca-I and Ca-II, have been classified as a fcc and bcc structures, respectively [1], [2].The third phase (Ca-III) has been primarily linked with the sc structure, however the recent reports suggest other assignments.On the basis of the theoretical studies, Teweldeberhan et al. proposed the Cmmm structure [4].Nakamoto et al. also vote in favor of the Cmmm structure [5].On the other hand, Mao et al. have predicted the transition from the sc-like structure to the monoclinic phase at 30 K and p ≃ 40 GPa [6].It needs to be underlined that the stability of the structure sc in the area of the existence of the phase Ca-III is being confirmed by the results achieved by Errea et al. and Yao et al., at least for the temperature of 300 K [7], [8]. The existence of the phases Ca-IV and Ca-V has been experimentally examined in the papers [2] and [5].Fujihisa et al. have proposed for them the following assignment: the structure Ca-IV should be characterized by P 4 1 2 1 2 and Ca-V by Cmca space groups, respectively [9]. Electronic address: d.szczesniak@ajd.czest.pl(B) The dependence of the critical temperature on the pressure: stars -Okada et al. [14], squares -Yabuuchi et al. [15], circles -Sakata et al. [3]. In the year 2010, Nakamoto et al. discovered the new Ca-VI phase with the Pnma structure [10].Further, in the year 2011, Sakata et al. have reported the existence of the host-guest phase Ca-VII [3].We can notice that the high-pressure phase of the host-guest character had been previously predicted by Arapan et al. and then by Ishikawa et al. [11], [12]. The first mention of the existence of the pressureinduced superconducting state in calcium was provided by Dunn and Bundy in 1981 [13].Fifteen years later, Okada et al. determined the dependence of the critical temperature (T C ) on the pressure up to the 150 GPa [14] (please see Figure 1 (B) for details).In the year 2006, Yabuuchi et al. have repeated the experimental studies of Okada [15].It has been found that the values of the critical temperature increase much faster together with the increase of p in comparison to the results achieved by Okada.The last notable experimental results have been obtained by Sakata et al. [3].On the basis of Figure 1 (B), it can be easily noticed that for p = 216 GPa, the critical temperature takes the value equal to 29 K (the highest observed T C among all elements).However, this result has been challenged by Andersson [16]. In the presented paper, we have determined all relevant thermodynamic parameters of the superconducting state that is induced in calcium under the pressure at 161 GPa.We draw the readers' attention to the fact that the pressure of 161 GPa represents the highest value of p considered by Yabuuchi et al. [15].Additionally, the high value of the critical temperature at p = 161 GPa, which is equal to ∼ 25 K, has been recently confirmed by the results obtained by Sakata et al. [3]. For the purpose of this paper, we have assumed that the phase Ca-VI is being characterized by the P nma crystal structure.To support this assumption we quote the results presented in: [10], [17] and [18]. The complicated form of the electron-phonon pairing kernel is represented by the expression: where Ω max is the maximum phonon frequency (Ω max = 71.37 meV), and α 2 F (Ω) indicates the Eliashberg function, which models the shape of the electron-phonon interaction in a detailed way.In the presented paper, the form of the α 2 F (Ω) function has been taken from Yin et al. [17].The depairing interaction between electrons is described with the use of the Coulomb pseudopotential (µ * ).The symbol θ denotes the Heaviside function and ω c is the phonon cut-off frequency: ω c = 3Ω max . In the presented paper the Eliashberg equations have been solved for 2201 Matsubara frequencies (M = 1100).In this case, the obtained solutions are stable for the temperatures greater than or equal to T 0 = 5 K.A detailed discussion of the numerical method has been presented in [20]- [24]. III. THE COULOMB PSEUDOPOTENTIAL The physical value of the Coulomb pseudopotential (µ * C ) can be defined by using the condition: [∆ m=1 ] T =TC = 0, where the critical temperature is equal to the experimental value (T C = 25 K) [15].In Figure 2 (A) we have presented the dependence of the order parameter on ω m for selected values of µ * .One can notice that together with the increase of the Coulomb pseudopotential, the largest value of the order parameter (∆ m=1 ) decreases.Additionally, in Figure 2 (B) we have outlined the complete form of the function ∆ m=1 (µ * ).On the basis of the obtained results, we have found that the physical value of the Coulomb pseudopotential is equal to 0.24. The above result means that the depairing electron correlations in calcium are relatively strong (the classical low-temperature superconductors µ * C is about 0.1 [25]).It can be noted that similar non-standard value of µ * C has been obtained for lithium and CaLi 2 [26], [27], [28], [29], [30].For example, the properties of the superconducting state in the fcc phase of lithium for the pressure values 22.3 GPa (T C = 7.27 K) and 29.7 GPa (T C = 13.93K) have been specified in the paper [29].It has been shown that the physical value of the Coulomb pseudopotential increases with p from 0.22 to 0.36.In the case of CaLi 2 , the parameter µ * C is equal to 0.23 (p = 45 GPa and T C = 12.9 K) [30]. The high values of µ * C in Ca, Li, and CaLi 2 are difficult to explain in the framework of the classical Morel-Anderson (MA) model [31], where: The symbol µ is defined by: µ ≡ ρ (0) U C , where ρ (0) is the electronic density of states at the Fermi level and U C is the Coulomb potential; ω e and ω ph denote the characteristic electron and phonon frequency, respectively.Since ω e ≫ ω ph , the MA pseudopotential is of the order 0.1 and µ * C ≪ µ.It can be noted that the MA model corresponds to treating the irreducible vertex to the first order in U C .Recently, Bauer, Han, and Gunnarsson have extended the MA theory to the second order in U C .The main result is that the retardation effects lead to the reduction µ → µ * C also in the higher order calculation, but not as efficiently as in the first order [32].The model presented in the paper [32] is probably the most advanced attempt to explain the high values of µ * C in Ca, Li, and CaLi 2 .In particular, Bauer et al. have given the following expression for the physical value of the Coulomb pseudopotential: where a = 1.38 and α ≃ 0.10.On the basis of the equation ( 4) one can easily estimate the value of U C for the real materials.In the paper, we assume the following: ω e = W (W is the half-band width), Ω ln (Ω) ), and ρ (0) = 1/2W (the constant DOS).In the literature, the values of all the important parameters are provided only for CaLi 2 (p = 45 GPa).In particular: µ * C = 0.23, W = 1991 meV, and ω ln = 17.02 meV [30], [33].The result is the following: U C = 2803 meV. Next, we address an important issue, namely, how big is the error bar of the calculated physical value of the Coulomb pseudopotential (∆µ * C ).First of all, we can notice that in the framework of the presented analysis, the value of µ * C depends on the shape of the Eliashberg function and the accuracy of the experimental value of T C .For calcium the appropriate Eliashberg function, taken from [17], has been calculated by using the linearresponse method (full-potential LMTART code [34]).In that paper, the Eliashberg function error bar has been omitted ([∆µ * C ] α 2 F (Ω) = 0).On the other hand, the value of the critical temperature has been measured with the accuracy about ±1 K [15].On the basis of these facts, we have obtained: [∆µ * C ] TC = ±0.02. IV. THE CRITICAL TEMPERATURE In the framework of the presented formalism, the exact value of the critical temperature should be obtained on the basis of Eliashberg equations.However, in the case of the data interpretation, it is far more convenient to use the simple formula that explicitly reproduces the results of the advanced numerical calculations.In the branch literature there are known two basic formulas that serve for the determination of the critical temperature's value.The first one has been introduced by McMillan [35]; the second one is the Allen-Dynes expression [36].Unfortunately, in the case of calcium, both formulas considerably underestimate the critical temperature.According to the above, the Allen-Dynes formula has been modified in such a way, that it allows us to reproduce the numerical results correctly.Particularly, in order to achieve the proper values of the fitting parameters, the dependence of the critical temperature on the Coulomb pseudopotential has been analyzed on the level of the Eliashberg equations (only α 2 F (Ω) has been considered as the physical input parameter).Next, the least-squares method was applied.The obtained result is presented below: where the functions f 1 (µ * ) and f 2 (µ * ) are expressed by the formulas: and The parameters, that depend on the Eliashberg function, can be determined on the basis of the expressions: For calcium under the pressure at 161 GPa, we have respectively: √ ω 2 = 34.36meV and λ = 1.27.The fitting functions Λ 1 (µ * ) and Λ 2 (µ * ) are presented in the following forms: and In Figure 3 we have presented the numerical solutions obtained with the use of the Eliashberg equations and the modified Allen-Dynes formula.Additionally, for the comparison purposes, we have depicted the results based on the classical formulas derived by Allen-Dynes and McMillan.On the basis of Figure 3, one can observe, that the expression (4) perfectly reproduces the exact Eliashberg numerical predictions. The constants in the equation ( 4) deviate notably from the original parameterization.This situation is connected with the fact that the analysis based on the real-axis Eliashberg equations suggests only the semiphenomenological form of the T C -formula: λ−µ ⋆ (1+cλ) (see the detailed discussion in [25], p. 1051-1052).In the case of calcium under the pressure at 161 GPa, the values of the Allen-Dynes parameters are inappropriate.Thus, the constants (a ∼ c) should be fit to the data taken from the exact solutions of the Eliashberg equations.We can notice that the change of the coefficient in the expression (4) under ω ln from 1.2 to 1.45 slightly lowers the phonon frequency; the two remaining parameters (1.03 and 0.06), in comparison to the classical parameterization, increase the value of the effective electron-phonon coupling constant. Moreover, the parameterization of the strong-coupling correction function (f 1 (µ * )) and the shape correction function (f 2 (µ * )) also deviates from the original form.The achieved result indicates that for the high-pressure superconducting state in calcium the shape function has greater significance than in the classical superconductors. The value of the critical temperature for p = 160 GPa has been also calculated in the paper [17].By using the Allen-Dynes formula the Authors have qualitatively reconstructed the experimental value of T C .However, in the examined case the physical value of the Coulomb pseudopotential has been strongly lowered (µ * C ∼ 0.15).In the last step, we boldly underline that in the literature exist other calculations of λ than has been presented in the paper [17].In particular, Lei et al. have suggested a very large value of the electron-phonon coupling constant (λ = 3.75 for p = 155 GPa and the sc structure) [37].On the other hand, Aftabuzzaman and Islam have predicted λ = 0.903 for p = 161 GPa and the Pnma structure [18].The last result is similar to the result obtained by Yin et al. [17]. V. THE CHARACTERISTICS OF THE SOLUTIONS ON THE IMAGINARY AXIS The form of the order parameter on the imaginary axis for selected values of the temperature has been presented in Figure 4 (A).It has been shown that together with the increase of ω m the absolute values of ∆ m are decreasing and are subjected to saturation.It should be underlined that taking the negative values by the order parameter function is connected with the non-zero value of the Coulomb pseudopotential. When analyzing the temperature's dependence of the order parameter, we found that absolute values of the function ∆ m decrease together with the temperature's growth.The above result means that together with the growth of the temperature, the less number of Matsubara frequencies give significant contribution to the Eliashberg equations. The full dependence of the maximum value of the order parameter (∆ m=1 ) on the temperature has been plotted in Figure 4 (B).We can observe that the values 2∆ m=1 (T ) with a good approximation reproduce the temperature dependence of the energy gap at the Fermi level.In Figure 5 (A) we have presented the form of the wave function renormalization factor on the imaginary axis.Similarly as for the order parameter, the increase of ω m causes the decrease of the successive values of the function Z m .In the case of the high values of ω m , the function Z m is subjected to the saturation and takes the value equal to one. Further, Figure 5 (B) presents the full dependence of the maximum value of the wave function renormalization factor on the temperature.It can be noted that the presented function with a good approximation determines the temperature dependence of the electron effective mass.Moreover, from the obtained results, we can conclude that the electron effective mass takes a high value in the entire area of the existence of the superconducting state. VI. THE PHYSICAL VALUE OF THE ORDER PARAMETER In order to determine the physical value of the order parameter for the chosen temperature, the solutions of the Eliashberg equations on the imaginary axis (iω n ) should be analytically continued on the real axis (ω).In the presented paper we have used the method introduced by Beach et al. [38].The form of the order parameter on the real axis is being reproduced by using the function: where p ∆j and q ∆j denote the number coefficients, and r = 550.The dependence of the real and imaginary part of the order parameter on the frequency for selected values of the temperature has been presented in Figure 6.Additionally, the rescaled Eliashberg function has been specified.On the basis of the presented results, one can observe that in the range of the low frequencies (from 0 to about 20 meV), only the real part of the order parameter takes the non-zero values.From the physical point of view, the obtained result defines the lack of the damping effects.For the higher values of the frequency (from about 20 meV to about 40 meV), the real part of the order parameter takes relatively high values, which are clearly induced by the characteristic peaks in the Eliashberg function.Furthermore, we can notice that in the discussed range of the energy, the imaginary part of the order parameter becomes non-zero and strongly increases together with the increase of frequency.For the higher frequencies (above 40 meV), the real part of the order parameter begins to vanish.This fact is related to the extinction of the Eliashberg function itself.The physical value of the order parameter for the chosen temperature should be determined on the basis of the expression [19], [25]: In the case of the superconductors the most interesting is the value of the order parameter for the temperature of zero Kelvin (∆ (0) ≃ ∆ (T 0 )).On the basis of the simple calculations we have made the following estimation: ∆ (0) = 4.32 meV. Let us mention that familiarity with the value of ∆ (0) and T C allows to calculate the dimensionless ratio: R 1 ≡ 2∆ (0) /k B T C .In the case of calcium we have obtained the following: The above result indicates that R 1 considerably exceeds the value predicted by the BCS theory: [R 1 ] BCS = 3.53 [39]. VII. THE ELECTRON EFFECTIVE MASS The influence of the electron-phonon interaction on the electron effective mass (m * e ) can be determined on the basis of the expression: m * e = Re [Z (0)] m e , where the symbol Z (0) denotes the value of the wave function renormalization factor on the real axis and m e is the bare electron mass.The form of the wave function renormalization factor on the real axis has been calculated with the use of the analytical continuation method: where p Zj and q Zj are the number coefficients, and r = 550. In Figure 7 we have presented the shape of the function Re[Z (ω)] and Im[Z (ω)] for the critical temperature.Similarly to the situation which took place in the case of the order parameter, for the low frequencies the nonzero is only the real part of the wave function renormalization factor.In the energy range around 20 meV we can observe characteristic but not so strong amplification of Re[Z (ω)], which is clearly correlated with the peaks of the Eliashberg function.Additionally, the function Im[Z (ω)] is non-zero.In the range of the high frequencies, Re[Z (ω)] decreases together with the increase of ω.Next, the dependence of the ratio m * e /m e on the temperature has been determined.The results have been presented in the inset in Figure 7.We have found that the electron effective mass is large in the entire range, in which the superconducting state exists, and reaches its maximum equal to 2.32 for T = T C . We can notice that for T = T C the value of the ratio m * e /m e can be calculated with a great accuracy by using the simple formula: m * e /m e ≃ 1 + λ = 2.27.The consistency between the exact numerical result and the analytical approach is the measure of the presented analysis. From the physical point of view, the result presented above is particularly important, since it can be verified in a simple way if the measurement of the Sommerfeld coefficient is to be made in the future. VIII. THE THERMODYNAMIC CRITICAL FIELD AND THE SPECIFIC HEAT The thermodynamic critical field (H C ) and the difference between the specific heat in the superconducting and state (∆C ≡ S − C N ) can be calculated on the basis of the free energy difference (∆F ≡ F S − F N ): The dependence of the free energy difference on the temperature has been presented in Figure 8.We can see that in the whole range of the existence of the superconducting phase, the value of the ratio ∆F/ρ (0) is negative.From the physical point of view, it means that the superconducting state is thermodynamically stable. The thermodynamic critical field should be determined on the basis of the expression: The influence of the temperature on the value of the ratio H C / ρ (0) has been presented in the Figure 8.The difference of the specific heat has been determined on the basis of the formula: Additionally, the values of the specific heat in the normal state have been also determined: where γ ≡ 2 3 π 2 (1 + λ).In Figure 9, we have plotted the dependence of the specific heat in the superconducting state and the normal state on the temperature.The characteristic "jump", which appears at the critical temperature, can be easily noticed.On the basis of the specified thermodynamic functions, we have calculated the values of the dimensionless ratios: R 2 ≡ ∆C (T C ) /C N (T C ) and R 3 ≡ T C C N (T C ) /H 2 C (0).We have obtained the following: R 2 = 2.17, (18) and Taking into account the results above, we can state that the values of the considered ratios significantly diverge from the values predicted by the classical BCS theory.In particular: [R 2 ] BCS = 1.43 and [R 3 ] BCS = 0.168. IX. SUMMARY In the paper, we have determined all relevant thermodynamic parameters of the superconducting state in calcium under the pressure at 161 GPa.We have conducted all numerical calculations in the framework of the Eliashberg formalism, where the electron-phonon spectral function α 2 F (Ω) has been taken form the paper [17].On the basis of the exact numerical results, we can state that the depairing electron correlations in calcium are relatively strong (µ * C = 0.24).In the next step, the values of the parameters in the Allen-Dynes formula have been calculated.It has been shown that the critical temperature is properly determined by the modified Allen-Dynes expression. Furthermore, we have proven that the thermodynamic properties of the superconducting state significantly diverge from the predictions based on the simple BCS theory.In particular, the following values of the thermodynamic ratios have been obtained: R 1 = 4.01, R 2 = 2.17, and R 3 = 0.158. In the last step, we have shown that the electron effective mass is large in the entire area of the existence of the superconducting state, and [m * e ] max = 2.32m e at T = T C . FIG. 1 : FIG. 1: (A) The sequence of the structural phase transitions in calcium determined on the basis of the experimental data.(B) The dependence of the critical temperature on the pressure: stars -Okada et al. [14], squares -Yabuuchi et al. [15], circles -Sakata et al. [3]. 24 FIG. 2 : FIG. 2: (A)The dependence of the order parameter on ωm for selected values of the Coulomb pseudopotential (T = TC).(B) The maximum value of the order parameter as a function of the Coulomb pseudopotential. FIG. 3 : FIG. 3: The dependence of the critical temperature on the Coulomb pseudopotential.The filled circles mean the results obtained by using the Eliashberg equations; the arrow indicates the experimental value of the critical temperature (µ * C = 0.24).The solid line denotes the results obtained with the help of the modified Allen-Dynes formula.Finally, the dashed and dotted lines have been generated based on the classical Allen-Dynes formula and the McMillan expression, respectively. FIG. 4 : FIG. 4: (A)The dependence of the order parameter on ωm for selected values of the temperature.(B) The maximum value of the order parameter as a function of the temperature. FIG. 5 : FIG. 5: (A) The dependence of the wave function renormalization factor on ωm for selected values of the temperature.(B) The maximum value of the wave function renormalization factor as a function of the temperature. FIG. 6 : FIG.6:The real and imaginary part of the order parameter on the real axis for selected values of the temperature.The rescaled Eliashberg function has been also presented. FIG. 7 : FIG. 7: The real and imaginary part of the wave function renormalization factor on the real axis.Additionally, the rescaled Eliashberg function has been outlined.The inset represents the dependence of m * e /me on the temperature. FIG. 9 : FIG.9:The specific heat in the superconducting state and in the normal state as a function of the temperature.	2019-04-06T13:11:45.266Z	2011-10-13T00:00:00.000	{ "year": 2011, "sha1": "1b0d0023f830fb1a786cfc0d908e8f3f0b6891b6", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1110.2870", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "1b0d0023f830fb1a786cfc0d908e8f3f0b6891b6", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics", "Materials Science" ] }
118856466	pes2o/s2orc	v3-fos-license	Analytical solutions of the coupled Gross-Pitaevskii equations for the three-species Bose-Einstein condensates The coupled Gross-Pitaevskii equations for the g.s. of the three-species condensates (3-BEC) have been solved analytically under the Thomas-Fermi approximation. Six types of spatial configurations in miscible phase are found. The whole parameter-space has been divided into zones each supports a specific configuration (miscible or immiscible). The borders of the zones are described by analytical formulae. Due to the division, the variation of the spatial configuration against the parameters can be visualized, and the effects of the parameters can be thereby understood. There are regions in the parameter-space where the configuration is highly sensitive to the parameters. These regions are tunable and valuable for the determination of the parameters. I. INTRODUCTION In recent years there are a number of literatures dedicated to the theoretical [1][2][3][4][5][6][7][8][9][10][11]. and experimental [12][13][14][15][16][17] study of the two-species Bose-Einstein condensates (2-BEC) (also refer to the references in [17]). The 2-BEC provides an important tool to clarify the inter-species and intra-species interactions. The miscible and immiscible phases of the ground state (g.s.) have been predicted and have been experimentally confirmed [14]. On the other hand, the study on the condensates with more than two species is very scarce. [18,19] Since the multi-species BEC is in principle experimentally achievable [20], a primary theoretical attempt might be worthy to see whether interesting physics is involved and whether this new field deserves a further study. This paper is dedicated to this purpose, namely, a primary theoretical study on the three-species Bose-Einstein condensates . The spatial configurations of the 3-BEC, as in 2-BEC, are expected to have also three phases: miscible, immiscible, and asymmetric phases (as shown below). For the first phase the atoms of each kind of species are compactly distributed surrounding the center of the trap, the distribution of some species is broader and some narrower. For the second either at least one species leaves completely from the center or at least one species is distributed in more than one disconnected spatial domains. For the first and second, the distribution keeps the symmetry as the trap. For the third, the distribution does not keep the symmetry of the trap. This paper is dedicated to the g.s. in miscible phase. The emphasis is placed on the qualitative aspect. The trap is assume to be isotropic. The spin-degrees of freedom are frozen. By introducing the Thomas-Fermi approximation (TFA, * Corresponding author: C.G.Bao, stsbcg@mail.sysu.edu.cn in which the kinetic energy is neglected), the coupled Gross-Pitaevskii equations (CGP) for the g.s. are solved analytically. This enable us to carry on the analysis in an analytical way. According to the relative distributions of the three species, the miscible phase has been further classified into six types. The analytical formalism derived in the follows enable us to divide the whole parameterspace into zones, each supports a specific type. Based on the division, the variety of the spatial configurations and their variation against the parameters (the intra-and inter-species interactions, the particle numbers, masses, and those for the trap) can be visualized, and thereby the effect of these parameters can be clarified. II. HAMILTONIAN AND THE COUPLED GROSS-PITAEVSKII EQUATIONS We consider three kinds of atoms N A A-atoms with mass m A and interacting via V A = c A Σ i<i ′ δ(r i −r i ′ ), N B B-atoms with m B , V B , and c B , and N C C-atoms with m C , V C , and c C . The particle numbers are assumed to be huge (say, larger than 10000). The interspecies interactions are V AB = c AB Σ i<j δ(r i −r j ) with the strength c AB , V BC with c BC , and V CA with c CA . These atoms are confined by the harmonic traps 1 2 m s ω 2 s r 2 ( s = A, B or C). We introduce a mass m o and a frequency ω. Then, ω and λ ≡ /(m o ω) are used as units for energy and length in this paper. The total Hamiltonian is where We consider the g.s. in which no spatial excitations are involved. Thus, each kind of atoms are fully condensed into a state which is most advantageous for binding (otherwise, the energy would be higher). Accordingly, the total wave function of the g.s. can be written as where u 1 , u 2 , and u 3 are for the A-, B-, and C-atoms, respectively. From minimizing the total energy, we obtain the set of CGP. One of them is where ε A is the chemical potential. Via cyclic permutations of the three indexes (A, B, C) and the three (u 1 , u 2 , u 3 ), from eq.(3) we obtain the other two equations. It is emphasized that the three equations of normalization u 2 l dr = 1 (l=1, 2, and 3) should hold. III. FORMAL SOLUTIONS UNDER THE THOMAS-FERMI APPROXIMATION Since N A , N B and N C are considered to be large, the approximation TFA has been adopted. A recent numerical evaluation of this approximation is referred to the papers [11,21]. Under the TFA, the CGP become , they are called the weighted strengths (W-strengths). they are the weighted energies for a single particle. In this paper all the interactions are considered as repulsive. Accordingly, all the W-strengths are positive. Furthermore, it is safe to assume that all the u l /r are always non-negative. Recall that there are originally 15 parameters (N s , m s , ω s , c s , c ss ′ ). From eq.(4) we know that their combined effects are fully represented by the nine α ll ′ . Among them, only eight are independent because they are related as α 12 α 23 α 31 = α 21 α 32 α 13 . Thus, based on the W-strengths, related analysis could be simpler. We define a matrix M with its element (M) ll ′ = α ll ′ . The determinant of M is denoted by D. The algebraic cominor of α ll ′ is denoted as d ll ′ . Obviously, the element of the inverse matrix (M −1 ) ll ′ = d l ′ l /D. The set of equations (4) has four forms of formal solution, each holds in a specific domain of r: (i) Form III: When all the three wave functions are nonzero in a domain, they must have the form as where D X l is a determinant obtained by changing the l column of D from (α 1l , α 2l , α 3l ) to (ε 1 , ε 2 , ε 3 ). D Y l is also a determinant obtained by changing the l column of D to (1/2, 1/2, 1/2). Once all the parameters are given, the three Y l are known because they depend only on α ll ′ . However, the three X l have not yet been known because they depend also on ε 1 to ε 3 . When Y l is positive (negative), u l /r goes down (up) with r. Obviously, once Y l is positive, X l must be large enough to prevent u l /r to be negative. (ii) Form II: Let (l, m, n) be a cyclic permutation of (1,2,3), the same in the follows. When one and only one of the wave functions is zero inside the domain (say, u n /r = 0), the other two must have the form as Once the parameters are given, the six Y n ′ must be positive and large enough. (iii) Form I: When one and only one of the wave functions is nonzero in a domain (say, u l /r = 0), it must have the form as Obviously, u l /r in this form must descend with r. This form could emerge only if ε l is positive and sufficiently large. (iv) Form 0: In this form all the three wave functions are zero. If u l /r is nonzero in a domain but becomes zero when r = r o , then a downward form-transition (say, from Form III to II) will occur at r o . Whereas if u l /r is zero in a domain but becomes nonzero when r = r o , then a upward form-transition (say, from Form II to III) will occur at r o . In this way the formal solutions will link up continuously to form an entire solution. They are continuous at the transition points because the wave functions satisfy exactly the same set of nonlinear equations at r o . However, their derivatives are in general not continuous at r o . When all the W-strengths are given, however, there are three unknowns ε l contained in the entire solution. Once they are known all the X l and X (l) l ′ can also be known. Due to the requirement of normalization, we have three additional equations. They are sufficient to determine the three ε l as shown below. IV. THREE LEMMAS There are three lemmas related to the linking of formal solutions. Lemma I: The three Y l can not all be negative. Let us define a vector are a set of orthogonal unit vectors, and all the α ll ′ are assumed to be positive as mentioned. Therefore → Ω l is situated inside the first octant. It can be rewritten as . The three → q l , → q m , and → q n should be linearly independent (otherwise, the determinant D is zero and the Form III does not exist). Then, → n can be expanded as Thus, the sign of Y l is determined by n l . Since all the three → q l to → q n are inside the first octant, if all the three n l , n m , and n n were negative, − → n would be in the first octant. This is in contradiction with the definition of → n. Thus the three Y l can not all be negative, and the lemma is proved. This lemma implies that Form III must transform to Form II somewhere because at least one of the Y l is positive, and therefore at least one the wave functions is descending and eventually arrives at zero. n can not both be negative. When l = 3, we define three 2-dimensional vectors . All of them are situated in the first quadrant. Then, Y . Y n . This lemma implies that Form II will transform to Form I somewhere because at least one of the wave functions is descending. Otherwise, it will transform to Form III if the missing wave function emerges. This lemma implies that Form II will either transform to Form I somewhere because at least one of the wave functions (say, u n /r) is descending, or transform to Form III if the missing wave function emerges earlier than the vanish of u n /r. Lemma III: In a domain (or at a point) where all the three u l /rare zero, no wave function can emerge and becomes nonzero in this domain (at the point). If u l /r emerges singly, then it must have the form eq.(10), therefore u l /r must descend with r and the emergence fails. If u l /r and u m /r emerge in pair at the same place, then both Y (n) l and Y (n) m should be negative to assure the uprising. This fails due to Lemma II. If all the three u l /r emerge together at the same place, then all the three Y l should be negative to assure the uprising. This fails due to Lemma I. Thus, the Lemma III is proved. Due to Lemma III, once the unique nonzero wave function in Form I arrives at zero, say, u l /r = 0 when r = r out , then r out will be the outmost border for all kinds of atoms. V. LINKING THE FORMAL SOLUTIONS TO FORM AN ENTIRE SOLUTION IN MISCIBLE PHASE With the three lemmas, we are going to link up the formal solutions to form an entire solution. To this aim, we will first make some presumptions so that the formal solutions can be linked up in a specific way. Then, we find out a subspace in the whole parameter-space. When the parameters are chosen inside this subspace, all the presumptions can be recovered so that the entire solution stands. In this way the whole space is divided into zones each supports a specific spatial configuration of the g.s.. Based on the division, we are able to obtain various types of phase-diagrams to demonstrate the variation of the g.s. against the parameters. For the miscible phase, the first domain (starting from r = 0) must have Form III. Therefore, the three X l > 0 should be presumed. Due to Lemma I, there is at least a positive Y l . Without loss of generality, it is assigned that X l /Y l is the smallest positive ratio among the three ratios. Accordingly, among the three wave functions, u l /r will arrive at zero first (refer to eq.(5)). Thus, the first domain ends at r a ≡ X l /Y l , where a downward form-transition occurs. For miscible phase u l /r is not allowed to emerge again because it is not allowed to distribute in disconnected region. Therefore u l /r is distributed only in (0, r a ). From the normalization ra 0 u 2 l dr = ra 0 (X l r 2 − Y l r 4 )dr = 1, we have X l = (15/2) 2/5 Y 3/5 l and r 2 a = ( This equation implies that the W-strengths should be so chosen that Y l ≥ Y m and Y l ≥ Y n hold. This choice assures that u l /r will arrive at zero first and the presumption X l > 0 can be recovered. The second domain will have the Form II and starts from r a . Since at least one of the two wave functions must descend with r (Lemma II ), we can assign the one that arrives at zero first with the index m, and we define m (refer to eq.(8)). Then, the equation of normalization for u m is Making use of the continuity at r a , namely, m r 2 a , we obtain and X It is clear that, in order to have u l /r descending in the second domain, Y (l) m > 0 is necessary to be presumed. Together with the previously mentioned condition Y l ≥ Y m , r b is well defined from eq.(12) and r b ≥ r a holds. Furthermore, once X (l) m is known, X m can be known from the continuity at r a as Recall that X m > 0 has been presumed. In order to recover this presumption, the W-strengths should be so chosen to ensure u n /r is distributed in three domains. In the first domain (0, r a ) where all the wave functions are nonzero, it must have the form u 2 n /r 2 = X n − Y n r 2 . In the second domain (r a , r b ), u 2 n /r 2 = X (l) n r 2 . While in the third domain (r b , r c ≡ √ 2ε n ) only u n /r is nonzero and appears as u 2 n /r 2 = 1 αnn (ε n − r 2 /2). When r = r c , u n /r arrives also at zero. Due to Lemma III, r c is the outmost border for all the atoms. Making use of the continuity at r a and r b , X n , X (l) n and ε n are related as X n = X . Inserting these two relations into the normalization rc 0 u 2 n dr = 1, we have Thus, ε n can be known when all the W-strengths are given. After ε n is known, X (1) n and X n can be known from the continuity as shown above. Thus u n /r is obtained. Furthermore, making use of eq.(6) we have ε l = Σ l ′ α ll ′ X l ′ . Thus, when all X l are known, all ε l can also be known. In the above form of u n /r, ε n > r 2 b /2 is required to assure un r \| r=r b > 0. Thus, from eqs. (14,12) the Wstrengths should be so chosen that is satisfied. In order to have un r \| r=ra > 0 (i.e., X should be satisfied. In order to have un r \| r=0 > 0 (i.e., X n > 0) should be satisfied. Thus, the miscible phase with all the three u l /r compactly surrounding the center and with the ranges r a ≤ r b ≤ r c will appear when the W-strengths are so given that the conditions (i) Y l ≥ Y m and Y l ≥ Y n . (ii) Y (l) m > 0, and (iii) eqs. (13,15,16,17) are satisfied. This specific miscible phase is denoted as {l,m,n} to demonstrate that u l /r has a narrowest distribution and u n /r has a broadest distribution. VI. DIVISION OF THE PARAMETER-SPACE Obviously, the above inequalities together define a specific zone in the multi-dimensional space of parameters. The borders of the zone are given by the surfaces defined by the equalities arising from changing each of the above inequality to equality. Note that the labels (l,m,n) can be any permutation of (1,2,3). Therefore, there are six types of miscible states and, correspondingly, six types of zones. Let the zones associated with {l,m,n}={1,2,3}, as Zone I to Zone VI, respectively. The zone not belonging to the above six is for the immiscible phase and is denoted as Zone 0. Once the whole space has been divided into zones, one can select any subset of parameters as variables while the others are fixed. This leads to various types of phase-diagrams that demonstrate the variation of the spatial configuration against the selected parameters. An example is given in Fig.1. Note that it was found in 2-BEC that the g.s. might be asymmetric when the interspecies interaction is sufficiently repulsive. This happens when α 2 lm ≥ α ll α mm (or c ss ′ ≥ √ c s c s ′ ), and is expected to happen also in 3-BEC. Therefore, in Fig.1, c ss ′ ≤ 3 4 √ c s c s ′ are chosen to avoid the possible appearance of the asymmetric configurations. Due to the choice of the parameters, a number of symmetries are involved in Fig.1. (i) Let c AB / √ c A c B and c CA / √ c C c A be denoted as x and y. A reflection with respect to the axis x = y is equivalent to the B-and C-atoms interchanging their names. Therefore, the pattern is invariant against the reflection together with an interchange of the indexes 2 and 3 (say, the zone {1,2,3} is changed to {1,3,2}), and therefore Zone I is changed to II. Similarly, III↔V and IV↔VI. (ii) When x = 1/2, due to the specific choice of the parameters, α 12 = α 21 = α 23 = α 32 . In this case the symmetry inherent in the CGP assures Y 1 = Y 3 , and u 1 = u 3 . Thus, for the two labels 1,3,2 and 3,1,2 (they are related to each other by interchanging 1 and 3), the g.s. can be denoted by either one of them at the axis x = 1/2. Accordingly, once Zone II appears in one side of the axis, Zone V will also appear in the other side as its partner. Similarly, III and IV are partners. In general, the axis x = 1/2 is replaced by a surface Y 1 = Y 3 in the parameter-space. On the surface u 1 and u 3 overlap. (iii) Similarly, when the axis y = 1/2 be the common border of two neighboring zones, the labels for these two zones are related to each other by an interchange of 1 and 2 (say, {1,2,3} and {2,1,3}). Thus, I and III are partners. V and VI also. As before, one can prove that, u 1 and u 2 overlap at the horizontal line y = 1/2 (or, in general, on the surface Y 1 = Y 2 ). (iv) The point O is the intersection of the x = 1/2 and y = 1/2 axes (in general, the intersection of the two surfaces Y 1 = Y 2 and Y 2 = Y 3 ), where all the α ll ′ are equal, and the three wave functions u l overlap. Accordingly, in the neighborhood of O all the six types {l,m,n} have an equal probability to appear as shown in the figure. Since the solutions have been obtained in an analytical way, it is straight forward to plot the wave functions. Examples are shown in Fig.2. In 2a the wave functions are associated with the point P 2 marked by a cross in Fig.1, where the g.s. is in the {1,3,2} phase. The pattern associated with P 1 is identical with that of 2a but u 2 and u 3 interchange. The pattern associated with P 3 is close to 2a (not exactly the same) but u 1 and u 3 interchange. All the three points are not far away from the point O. Therefore the three wave functions are not remarkably different from each other. Otherwise, they might be very different. 2b is associated with P 4 , where the g.s. is also in the {1,3,2} phase. However, due to P 4 is very close to the zone of immiscible phase, the B-atoms tend to leave completely from the center and tend to form a shell as shown by the dash curve. It was found that in the neighborhood of the border separating the miscible and immiscible phases, the con- figuration is very sensitive to the variation of parameters. For an example, P 4 (marked in Fig.1) has x = 0.71, and accordingly (u 2 /r) r=0 = 0.048 (it implies that the Batoms are very few at the center). When P 4 shifts a little away from the border so that x becomes 0.70, (u 2 /r) r=0 becomes 0.207. Thus, the neighborhood of the above border is a region of sensitivity. In this region a tiny change in the parameters) might cause an explicit change in the configuration. The existence of regions of sensitivity in the parameter-space is a notable phenomenon. One more example is given in Fig.3 to demonstrate the effect of particle numbers. In this figure the number of A-atoms is smaller. The symmetry with respect to the x = y axis (i.e., an interchange of 2 and 3) remains, while the other symmetries appear no more. Note that the Zone I and II are dominant in Fig.3 implying that the A-atoms are closer to the center. Thus, when all the three c s are close to each other and the three γ s also, the kind of atoms with least particle number will be closer to the center. VII. FINAL REMARKS We have succeeded to derive the analytical solutions of the CGP for the 3-BEC under the TFA . Thereby the parameter-space has been divided into zones each supports a specific configuration. Based on the division, various types of phase diagrams can be plotted, and the variation of the spatial configurations against the parameters can be visualized. From the experience of 2-BEC, when the particle numbers are large and when both kinds of atoms are distributed surrounding the center (i.e., u l are nonzero when r = 0), the wave functions obtained under TFA and beyond TFA overlap nearly completely (refer to Fig.1a and 1b of [11], where a detailed discussion on the accuracy of the TFA is made). Since this paper concerns only this case , the TFA is believed to be applicable. Nonetheless, this remains to be further clarified. Obviously, this paper is far from a complete description of the 3-BEC. Note that, when the inter-and intraspecies interactions are close in strengths or the former is stronger than the latter, Symmetric immiscible states and asymmetric states may emerge. The details and the classification of these states remain to be studied. The variety of the spatial configurations of 3-BEC implies that rich physics is involved. In particular, just as in 2-BEC, regions of sensitivity have been found. When a realistic parameter falls in a region of sensitivity, it can be more accurately determined. Obviously, these regions of 2-BEC and 3-BEC are different. Thus, in addition to the 2-BEC, the 3-BEC will be helpful in the determination of parameters. Recall that the BEC are a valuable tool because they are tunable. One can consider that the addition of the third kind of atoms into a 2-BEC is an one more way to tune the system. Incidentally, the above analytical approach is quite general and can be generalized to deal with the condensates with more than three species.	2016-11-19T12:51:39.000Z	2016-11-19T00:00:00.000	{ "year": 2016, "sha1": "5e427e15856a1e6ddc0235df00481e46dae4194c", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1611.06358", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "5e427e15856a1e6ddc0235df00481e46dae4194c", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
201619818	pes2o/s2orc	v3-fos-license	Multisystem impairment in South African adolescents with Perinatally acquired HIV on antiretroviral therapy (ART) Abstract Introduction Adolescents with perinatally acquired HIV (PHIV) are at risk of chronic disease due to long‐standing immune suppression, HIV disease and antiretroviral therapy (ART) exposure. However, there are few data on multisystem disease in this population. We investigated the overlapping burden of neurocognitive, cardiovascular, respiratory and/or renal impairment among PHIV positive (PHIV+) adolescents. Methods In this cross‐sectional analysis, participants aged 9 to 14 years on ART for >6 months were recruited from seven sites across Cape Town from July 2013 through March 2015, together with age‐matched HIV‐negative (HIV‐) adolescents. Impairment at enrolment was assessed across neurocognitive functioning (using the youth‐International HIV Dementia Scale); cardiac function (echocardiogram abnormality); respiratory function (abnormal spirometry) and renal function (abnormal glomerular filtration rate). Results and Discussion Overall, 384 PHIV+ and 95 HIV‐ adolescents were included (mean age, 11.9 years; 49% female). Median age of ART initiation was 4.2 years (IQR: 1.7 to 7.6) and median CD4 count was 709 (IQR: 556 to 944) with 302 (79%) of PHIV+ adolescents virologically suppressed. Abacavir and Zidovudine were the most commonly used nucleoside reverse transcriptase inhibitors (NRTIs) with 60% of adolescents on non‐nucleoside reverse transcriptase inhibitors (NNRTI) and 38% on a protease inhibitor (PI). Among PHIV+ adolescents, 167 (43.5%) had single system impairment only, 110 (28.6%) had two systems involved, and 39 (10.2%) had three or four systems involved. PHIV+ participants had more 2‐system and 3‐system impairment than HIV‐, 110 (28.6%) versus 17 (17.9%), p = 0.03 and 39 (10.2%) versus 3 (4.3%), p = 0.03. PHIV+ participants who had failed a year of school (73.8% vs. 46.4%, p = 0.00) and with a viral load >1000 copies/mL at enrolment (16.8% vs. 8.1%, p = 0.03) were more likely to have dual or multisystem impairment. Of those with cardiac impairment, 86.7% had an additional system impaired. Similarly, in those with neurocognitive impairment, almost 60% had additional systems impaired and of those with respiratory impairment, 74% had additional systems impaired. Conclusions Despite relatively early ART initiation, there is a substantial burden of multisystem chronic impairment among PHIV+ adolescents. This phenomenon needs to be further explored as this population ages and begins to engage in adult lifestyle factors that may compound these impairments. \| INTRODUCTION There are 2.1 million adolescents aged 10 to 19 years living with HIV, the majority of whom live in sub-Saharan Africa and have perinatally acquired infection [1,2]. This population is growing as most children on antiretroviral therapy (ART) are surviving into adolescence, due to increased access and earlier initiation of ART [3]. Single system morbidity in perinatally HIV-positive (PHIV+) adolescents is well described but there is limited data on multimorbidity in this population. Cardiac, respiratory, neurocognitive or chronic renal impairment are common long-term sequelae of perinatal HIV infection [4][5][6][7]. PHIV+ adolescents in the United States (USA) and Europe have high rates of chronic lung disease, kidney disease and neuropsychiatric problems [8]. In sub-Saharan Africa, however, diagnosis and initiation of ART occur much later than in the USA, with the median ages at first visit and at ART initiation of 7.1 and 7.9 years compared to 0.7 and 0.9 years respectively [9]. HIV+ children in sub-Saharan Africa are therefore at greater risk of chronic morbidity due to untreated HIV in childhood, long-standing immune suppression and associated infection, suboptimal ART formulations and regimens or lack of access to care [10]. The interplay between these factors along with chronic inflammation result in PHIV+ adolescents being at risk for multisystem impairment [11]. Single system morbidity has been described in resource limited settings, mostly focusing on cardiac or chronic lung disease [12,13]. A recent study from Asia showed that infectious HIV-related morbidity was more common in younger adolescence (10 to 14 years of age) with a trend toward noninfectious and treatment-related morbidity in later adolescence, however, this study did not specifically explore chronic or multisystem morbidity [14]. Despite the high prevalence of single system morbidity, there are surprisingly few data on prevalence of and risk factors for multisystem involvement in PHIV+ adolescents on ART, especially in Africa. Multisystem morbidity may be associated with worse clinical outcomes, increased healthcare utilization and more difficulty in adhering to ART due to a high pill burden. Many studies of single organ impairment are limited by small sample sizes, with no comparison group of HIVadolescents and many are from the pre-ART era. As growing numbers of PHIV+ adolescents present to overburdened health systems in resource limited countries, optimizing strategies to best care for them is crucial. The aim of this study was to investigate the prevalence of and risk factors for overlapping multisystem (neurocognitive, cardiovascular, respiratory and renal) impairment in PHIV+ adolescents in the Cape Town Antiretroviral Cohort (CTAAC). \| Study population This was a cross-sectional study of PHIV+ children and adolescents enrolled in CTAAC, a longitudinal cohort study in Cape Town, South Africa. Children between 9 and 14 years on ART for more than six months were enrolled from seven sites in the Western Cape Province, South Africa with age matched HIV-youth of similar ancestry from July 2013 to March 2015. Children and adolescents between the ages of nine to fourteen years were considered to be perinatally infected [15]. Ethical approval was given by the Faculty of Health Sciences, University of Cape Town and Stellenbosch University, Human Research Ethics Committee (051/2013). Parents gave informed consent and assent was obtained from all adolescents. All participants knew their HIV status as a pre-requisite to study enrolment. For this analysis, we included only those participants with complete respiratory, cardiac, neurological and renal assessments from the enrolment visit ( Figure 1). \| Sociodemographic data and other health information Routine sociodemographic data were collected at enrolment and each participant's clinical record was reviewed at their primary treatment facility. Participants were screened for cardiac and respiratory symptoms such as wheeze, cough, shortness of breath and a validated respiratory questionnaire derived from the International Study of Asthma and Allergies in Childhood study was performed at enrolment [16]. A physical examination including Tanner staging, WHO HIV staging, blood pressure (BP) and anthropometry was performed at enrolment. Body Mass Index (BMI) was calculated as weight in kilogrammes divided by height in metres squared (kg/m 2 ). BMI was classified according to WHO reference standards [17]. BP was measured using an electronic sphygmomanometer (Spot Vital Signs, Welch Allyn, New York, NY, USA). All anthropometric measures were performed by one of two trained study nurses to ensure standardization of measures. Laboratory \| Cardiac measures Echocardiograms were performed by two trained research echocardiographers using either a Philips iE33 or CX50 (Phillips, Eindhoven, The Netherlands) using standardized techniques. All echocardiograms were interpreted by a single paediatric cardiologist and a random subset was read by a second blinded paediatric cardiologist. Left ventricular shortening fraction was measured by M-mode and ejection fraction was derived using standard methods. The ejection fraction was also measured using the modified Simpson's method [20]. Left ventricular diastolic function was measured using Doppler assessment of mitral inflow. Tissue Doppler techniques were used to measure mitral annular velocity. Tricuspid annular plane excursion was measured using M-mode [20]. Pulmonary artery pressures (systolic and diastolic) were estimated using standard continuous and pulse wave Doppler methods. Cardiac dimensions were assessed either using direct measurement of 2-D images or M-mode recordings. Body surface area (BSA) was estimated using the Mosteller formula [21]. Echocardiographic structural parameters were expressed as raw means as well as a deviation from the BSAcorrected mean (z-scores), based on normal values [22]. \| Respiratory measures Spirometry was done using the NDD EasyOne Pro LAB â (NDD, Switzerland). Testing adhered to the American Thoracic Society/European Respiratory Society (ATS/ERS) guidelines [28][29][30]. Lower limit of normal (LLN) for spirometry outcome variables was calculated using the African-American reference cohort in global lung initiative (GLI) software, À1.64 standard deviations (SD) below the mean [31]. Lung function testing was deferred if the participant had an acute respiratory illness. \| Neurocognitive measures The youth-International HIV Dementia Scale (y-IHDS), a sensitive screening test for neurocognitive disorders, was used to screen for cognitive impairment [33]. The y-IHDS is a 3-part test that includes timed finger tapping, a time alternating hand sequence test and a two-minute delayed recall of four words [34]. Each participant is asked about a history of repeating a grade/grades at school, with one point subtracted from the total score for positive response. The test was conducted in the participant's home language by trained study doctors. Neurocognitive impairment was defined as a y-IHDS less than or equal to 10 [33]. \| Renal measures Enrolment blood was taken to assess creatinine and standing height was measured using a stadiometer with a moveable headboard in centimetres. Serum creatinine was measured in lmol per litre by the enzymatic method. The modified Schwarz formula was used to estimate glomerular filtration rate (GFR) [35]. Renal impairment was defined as glomerular filtration rate (GFR) below 90 mL/min/1.73 m 2 . \| Primary outcomes The primary outcomes were the number of participants with single, dual and multisystem impairment. Single, dual and multisystem impairment were defined as having impairment of only one (single), only two (dual) or three or more (multisystem) of the following systems: cardiac, respiratory, neurological or renal. Baseline variables and outcomes for PHIV + and HIVÀ adolescents were compared using t-tests, Wilcoxon and chisquare tests as appropriate. Among PHIV+ participants, logistic regression was performed to evaluate factors associated with having dual or multisystem impairment. Covariates considered for associations with multisystem impairment included anthropometry, a history of having had Tuberculosis (TB), HIV laboratory parameters, and duration and type of ART. Statistical analysis was performed using Stata version 14.1. StataCorpInc. College Station, TX, USA. \| RESULTS Four hundred and seventy-nine participants (384 PHIV+ and 95 HIVÀ) had complete respiratory, cardiac, neurological and renal assessments from the enrolment visit. \| PHIV+ participants Neurocognitive impairment was the most common type of single system impairment. Overall, a low TAPSE and FAC, indicative of right ventricular dysfunction, accounted for the majority of echocardiogram abnormalities with 104 (27.1%) and 42 (10.9%) of those PHIV+ adolescents that had cardiac impairment having an abnormal TAPSE and FAC respectively. Left heart dysfunction was rare with only one PHIV+ adolescent having a decreased LVSF and 32 (8.3%) having evidence of left ventricular diastolic dysfunction. No participant had dilated cardiomyopathy and 2 (0.5%) PHIV+ adolescents had raised pulmonary artery pressure. Ninety -seven (25.3%) PHIV+ participants had an abnormal FEV1 and 35 (9.1%) had a FEV 1/FVC<LLN. The most common patterns of dual and multisystem impairment were "cardiac and neurocognitive" and "cardiac, neurocognitive and respiratory" impairment. Only 2 (0.5%) PHIV+ participants had impairment of all four systems. In participants with any impairment, the majority had additional system impairment: Of those found to have cardiac impairment 126/ 177 (86.7%) had an additional system impaired. Similarly, in those with neurocognitive impairment almost 60% had additional systems impaired and of those with respiratory impairment 74% had additional systems impaired. In univariate analysis having failed a grade or having a viral load >1000 copies/mL at enrolment, were more common in those with dual or multisystem impairment compared to those with no impairment or single system impairment, 73.8% versus 46.4%, p = 0.00 and 16.8% versus 8.1%, p = 0.03 (Table 3). Participants with dual and multisystem disease were also older 12.2 years versus 11.8 years, p = 0.02. In multivariable analysis failing a grade was associated with dual or multisystem disease (OR = 3.2, CI 2.1 to 5.1, p ≤ 0.01). \| DISCUSSION This is the first study to report on multisystem impairment in African PHIV+ children and young adolescents on ART. The study found the prevalence of any cardiac, respiratory or neurocognitive impairment in PHIV+ participants was significantly higher than in HIVÀ participants. Similar findings of individual system involvement have been reported in other studies of adolescents with perinatally acquired HIV with a high prevalence of cardiac, respiratory, neurocognitive or less commonly renal impairment [5,36,37]. Neurocognitive impairment occurring most commonly, is perhaps the most concerning morbidity, impacting on PHIV+ adolescents as it influences all spheres of health including treatment adherence and school performance. The measure used, the y-IHDS, is only a screening test and cannot definitively diagnose the type or extent of neurocognitive impairment. The score has been validated previously in a subset of this cohort and shown to be sensitive for neurocognitive disorder screening [33]. In addition, it is quick and easily performed in busy clinic settings. We found a prevalence of neurocognitive impairment of 56% using y-IHDS, similar to the 45% of HIV+ youth that met criteria for neurocognitive disorder diagnosis through an extensive battery of neurocognitive tests in the same setting [38]. However, in other African settings, the adult IHDS score overestimated the burden of neurocognitive impairment [39]. Follow-up data from our adolescent cohort will be valuable to assess whether the y-IHDS screening tool correlates with confirmed impairment. Cardiac impairment was the second most commonly affected system. This was based on echocardiogram parameters reflecting subtle right ventricular dysfunction. The majority of these participants were asymptomatic with no difference between PHIV+ and HIVÀ participants. These results are consistent with a Spanish study showing subtle cardiac abnormalities found on echocardiogram and no difference in the control population [4]. but the authors did not report on right heart dysfunction. Lower TAPSE has been reported in HIV+ young adults but not in HIVÀ controls [40]. Follow-up of all participants in our study may indicate if these findings are clinically relevant as both TAPSE and FAC are more useful in longitudinal studies [41,42]. Respiratory single system impairment was reflected by subtle reduction in lung function that may impact on adult lung health. Our findings are consistent with the prevalence of abnormal spirometry, reported as between 24% and 38% in various African adolescent cohorts from Zimbabwe, Malawi or South Africa [32,43,44]. Renal impairment was surprisingly rare given the genetic predisposition to HIV nephropathy in Black Africans but may reflect survivor bias as those with more severe kidney disease may have died or already being followed at specialized renal clinics. The prevalence of proteinuria and microalbuminuria, risk factors for chronic kidney disease has been shown to be high in HIV+ adolescents, however, we previously reported no difference in these measurements between PHIV+ and HIVÀ adolescents in this cohort [45]. Dual system impairment affected about a quarter of participants but multisystem involvement was relatively uncommon. This may reflect that the cohort were ambulatory and relatively well, were on ART for several years and were adherent to therapy. Despite this, there remains a small but clinically significant proportion of adolescents that will need complex clinical services to ensure optimum care. In addition, a significant proportion of those with impairment in one system had another system involved. Multivariable analysis found that failing a grade at school and age at enrolment was significantly associated with dual or multisystem disease. School failure may be due to neurocognitive impairment or chronic or prolonged illness and hospitalization resulting in missing significant periods of the school year [46]. As this was a cross-sectional study we were unable to infer that multisystem disease had a causal relationship with school failure. Checking for school failure There was no evidence for the association of duration or for the relatively later start of ART and dual or multisystem impairment. A possible explanation for later start of ART not being associated with multisystem impairment is that historically children that were started on early ART prior to the guidelines recommending early start for all were severely ill and this illness may have resulted in multisystem impairment despite early access to ART. In addition, treatment histories were often not available or difficult to interpret. Reasons for switching ART regimens were poorly documented and it was not possible to accurately assess viral suppression prior to study enrolment. The study is limited by the cross-sectional analysis that limited the ability to detect longitudinal changes. Only four systems were included, and a more comprehensive assessment of system involvement including hearing, dermatological complications of HIV and musculoskeletal abnormalities may be useful. An additional limitation is that the measures of impairment that we chose for each system differ in their ability to assess severity of impairment with the y-IHDS score being a relatively crude estimate of neurocognitive impairment compared to detailed assessment of lung function that is obtained with spirometry. Severity of impairment across different systems thus cannot be directly compared. \| CONCLUSIONS In those PHIV+ adolescents that had one system impaired a significant proportion had another system involved. Although multisystem impairment is relatively rare, a small minority of youth will require clinical attention for complex multisystem issues. Young adolescents with system impairment will need close observation as they transition to adulthood and are increasingly at risk for engaging in adult lifestyle factors such as smoking or recreational drug use. Adult onset diabetes and hypertension may also compound these impairments. Longitudinal follow-up is needed to ascertain whether system impairment may impact long-term morbidity. LJF contributed to the initial concept of the paper, did the statistical analysis and wrote the manuscript. KB did statistical analysis. SM contributed towards data management and did statistical analysis. LG, DG performed and interpreted spirometry and were involved in the initial pulmonology concept of CTAAC, LZ read echocardiograms and contributed to the initial cardiology concept of CTAAC, PN gave input on renal measures, DJS and JH gave input on neurocognitive concepts, MFC, LM, HJZ were involved in the initial concept of the paper and obtained funding for CTAAC. All authors have read and approved the final manuscript. A C K N O W L E D G E M E N T S The authors acknowledge the CTAAC co-investigators: Helena Rabie, James Nuttall, Brian Eley, Linda Gail-Bekker, Paul Roux and the CTAAC study staff, caregivers and adolescents.	2019-08-24T13:05:10.017Z	2019-08-01T00:00:00.000	{ "year": 2019, "sha1": "0cadfad2de7e92f045020db0ade666d3a1110fe4", "oa_license": "CCBY", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/jia2.25386", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "676d132ea37c74162df54002e949a345ff97cb5e", "s2fieldsofstudy": [ "Medicine", "Psychology", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
55124716	pes2o/s2orc	v3-fos-license	China ’ s “ Ant Tribe ” Present Social Survival Situation and Personal Financial Advice The new term “ant tribe”, is referring to a social group of low-income college graduates that are in search of work and livelihood in the big cities (Lian Si, 2009). It is estimated this social group has already surpassed more than 3 million people, with a rapidly increasing rate of 0.2-0.3 million people a year. The investigation carried out on the issues of “ant tribe”, involved 327 people including 192 in the “ant tribe” group. The survey put forwards that this social group has three typical characteristics and four basic social features. The psychological state has been analyzed from three aspects and the author from the financial professional viewpoint proposes some advices for their personal finance. Keyword: ant tribe, investigation, typical characteristics, social features, survival situation, personal finance For further understanding of current China's "ant tribe" survival situation, an in-depth investigation on "ant tribe" issues was carried out by my group for a month time from January to February 2012, focusing on the "ant tribe" groups in Beijing and Zhengzhou city.The survey involved 327 participants, including 192 of the "ant tribe" community.As we and the "ant tribe" are peers, some of them were my high school classmates and friends who recently graduated from the university and started their "ant tribe" life.So, despite a very heavy workload in this project, and complicated situation, we managed to complete it successfully and smoothly.Through in-depth interviews, questionnaires and living together with the "ant tribe" for more than 10 days, we obtained a lot of information on the "ant tribe" group's living conditions, occupations, income levels and their mental state.After the fact-finding, I organized the retail field operators of my investigation team to discuss two times and we determined that every team member -from different point of view and personal interest -writes his or her own investigation reports.According to the survey records, through data analysis and with related literature references, I'm delighted to report the following: What Is the "Ant Tribe"? The new term of "ant tribe" originated from Dr. Lian, a Chinese young scholar in 2009.It refers to those who are low-income college graduates and inhabited by the group in search of work and livelihood in the big cities.The young graduates who live in "humble abode", the squalid conditions and low-cost housing located in the urban villages or in the rural-urban fringe zones of the big cities.This huge social group has appeared quietly in the big cities of China since 2005 and few people paid attention to them before 2009.In September, 2009 Dr. Lian published a book "The University Graduates Inhabited Villages--'Ant Tribe' Survey Report", which has caused the Chinese government and all sectors of society to pay wide attention.According to "The Chinese Talent Blue Book (2010)" (Pan C., 2010) disclosed the data, at this time, only in the Beijing area a conservative estimation of the "ant tribe" had more than 100,000 people.In addition, in recent two years this phenomenon of "ant tribe" not only exists in the first-tier Chinese cities like Beijing and Shanghai, but has also spilled into the second-tier Chinese cities such as Wuhan, Guangzhou, Xian, Chongqing, Taiyuan, Zhengzhou and the economically developed third-tier cities.It is estimated this social group has already surpassed 3 million, with a rapidly increasing rate of 0.2-0.3 million people a year.Dr. Lian had originally given a description of this group: "They are 'weak strong' just like ants and they are known to a large community by having three typical characteristics: university graduates, low income, inhabited by group".While, the author finds that the "ant tribe" also shall have four other basic social features: 1) All with degrees of higher education.They are the products of universities enrollment expansion in recent years and they are between 22 to 29 years old, and most of them are the generation born in the 80s or the 90s.This generation accounts for 95.3% in the ant tribe group, and it's been only 6-7 years since they graduated.They are mainly singles, some claiming to be the "migrant worker", some claiming to be the "ant tribe" members, but all most call themselves the "diaosi 1 " and they call their home "woju 2 ". "The Chinese Talent Blue Book ( 2011)" (Pan C., 2011) showed that the "ant tribe" bunch who graduated from the 211 project universities accounts for 28.9%, the ratio of relative data compared to 2009 has increased nearly twofold, and the proportion of postgraduates with a master degree has increased from 1.6% in 2009 to 7.2% in 2010, soaring 3.3times.The China Internet "Forum of Postgraduates" commented: "Even the postgraduate degree holders will descended to the 'ant tribe' group, and the proportion is rising, this explains the high degree is no longer a high income, high treatment marks". 2) Low income, unstable job.The "ant tribe" members are mainly working in private enterprises and individual enterprises.The nature of work is chiefly in retail of technology related products, such as computer and mobile phone stores, IT services industry, product marketing, some have occupation in restaurants and supermarkets, etc.Most of them have no insurance and the labor contract is short with low and unstable income, they job-hop frequently, many of them being unemployed or underemployed.Some need to support their family by sending funds back home; while others depend on their family's financial support; and some "ants" appear malnourished, hardly looking forward to work, chronic fatigue syndrome disorder occurring from time to time.The results of this survey show that: the "ant tribe" in Beijing has the average monthly income of 2200 Yuan RMB.Zhengzhou's "ant tribe" has an average monthly income of 1000-1500 Yuan accounting for 33%, 1500-1800 Yuan accounting 31%, 1800-2000 Yuan accounting for 24%, and there are 12% of the "ant tribe" respondents below 1000 Yuan. Figure 1. Zhengzhou ants -average monthly income The survey data also shows that there are 67.7% of the "ant tribe" respondents have not the habit of eating breakfast and 88.1% of them control their meal cost to 10 Yuan max, and the majority of people would consider a meal of over 5 Yuan as too expensive.All this stems from the fact of "unsteady income, low earnings."Xiao Zhao and Liu Xiang, working at a Zhengzhou tourism company as tour guides, told us, "When we bring tourists in the tourism seasons to the south areas of China and to Tibet, we can earn 7000-8000 Yuan a month, but during the off-seasons we have no income for months, so our average annual income is not high, most of the time we are living a hard life".While, looking at overall conditions, the "ant tribe" living and employment situation in Zhengzhou was slightly better than that in Beijing. 3) Inhabited high-density, a large number.In major cities, many recent college graduates live as "ant tribe" packed tightly in small dormitory rooms with four or more roommates each.The "ant tribe colonies" or "ant tribe villages" are formed in areas inhabited by a large number of people ranging from small-scale ones accommodating 2000-3000 people to large-scale ones of more than 50,000 dwellers.The "ant tribe" inhabited density, in the survey, there are 70% of respondents just have a living area of 10 square meters, 20 square meters or more for one person only accounting for 5.5% and 20% of the "ant tribe" rooms below 5-6 square meters.Apart from the three typical characteristics, this survey finds that the "ant tribe" also has the following four other distinct social characteristics: 4) Come from rural areas or urban low income families.Most of the "ants" come from the rural and economically underdeveloped areas.Those from rural areas account for 60.6% and those from county-level towns account for 27.3% respectively, and the small and medium-sized city kin accounting for 12.1%.They are can be called the "poor second-generation". 5) The first-generation university graduates in the vast majority of families.Many of their parents are farmers or farmer migrant workers, town enterprise worker or small vendors.6) They don't have registered permanent address in the city of their residence.According to Chinese current population administration system, they don't have city accounts, only have a "temporary residence permit" in the city they are working in or seeking work in, and their permanent residence is registered in their home towns.7) Relatively economic independence.They are a group of self-run occupation and their income is low but they live a self-determined life.With the exception of a few who need parental financial support, or a few who need support their large families (aging parents, siblings, grandparents), most of the "ant tribe" members live a life of relative financial independence as their families back home are self-sufficient. In addition, the overwhelming majority is under the age of 30, single or cohabitating.When they get married they will gradually move out of the "ant tribe" villages.Some of the married couples or families with children are still huddled in the "ant tribe village".Those may be the rural migrant workers.Although this social group leads a same life as the "ant tribe", they are not defined into the "ant tribe" class which just refers to the college graduates by Lian et al (2009).I think this social group should be the "ant tribe symbiosis" and studies on the topic of this group shall be carried out in the future. "Ant Tribe" Current Living Status "Above survival, life is following", highly summarized by Dr. Lian (2009) 3.1 Where Are the "Ant Tribe Villages"? The phenomenon of "ant tribe" in China appeared in Beijing, Shanghai, Guangzhou and other big cities first and then began to emerge gradually in the capital cities of provinces.Now that has spread to the medium-sized cities and some economically developed zones.County-level towns and underdeveloped cities have not yet experienced this phenomenon.The "Ant tribe" group lives mainly in urban-rural on the fringes of big cities or urban villages, where the streets are narrow, there is limited living space, and it's generally noisy and the lanes are filled with garbage.The dinghy accommodation is without a central heating system and without hot water.However room rents are cheap, with a low cost of living, food at the street stalls and transport located nearby.There is none of upper class well-to-do folks, civil servants, white-collar workers here, so these places have transformed into the "ant tribe colonies".Like Tang Jialing, an "ant tribe" large-scale ghetto in the north outer of Beijing Wuhuan Road.There was an "ant tribe" of as many as over 50,000 before 2011.In Zhengzhou city, there are many "ant tribe villages" with a populace of over 10,000 such as Zhao Village, Chen Village, Wuli Bao, Liulin Town, etc.The small scale "ant tribe colonies' " accommodate about 2000-3000 and a large number of localities scattered in the city. Two Common Types of "Ant Tribe" Members in Zhengzhou: Meet Little Bai and Little Meng The "ant tribe village" was empty and deserted during daytime; wage earners were away at work.We had contacted Little Bai and Little Meng and met them that evening after work in their "woju".Little Meng's working place was not so far, it takes 20 minutes by bike; Little Bai's working place was far away, commuting to work 1 hour and 20 minutes by bus.We met Little Meng about seven o'clock at the "ant tribe village" called Chen Village in the Zhengzhou city.At this time, most of the "ant tribe" folks had not returned from work.Little Meng led us to their room in an old village building, that belonged to a villager named Chen.There were a lot of similar buildings in the surroundings which belonged to Chen, a man originating from this village.Chen and other villagers had rented the ramshackle buildings out to the "ant tribe", so that the area had transformed into "ant tribe village".The corridor was dim, stairwells piled with sundries and garbage everywhere, exuded an unpleasant smell that couldn't help but made people think that in hot seasons, flies and mosquitoes would swarm around here.Entering a room, we saw the cramped room, narrow bed, dark space, which could be warrant the name: "ant habitat".Between their beds a long table was placed, which was shared by everyone for multiple tasks.There were: a computer, some books, and instant noodles waste boxes, bread bags and disposable chopsticks on the table.It was late at night.The "ant tribe village" was just like 'waking up', the buzzing noise increasing gradually, and that was the "ants" swarming back to their "home".A few of them had had "a rush meal" with their friends or workmates on the way home and most of them prepared a bowl of instant noodles for supper.They enjoyed their leisure while they were sitting in the public living room watching TV or surfing on internet.Most of them were schoolmates at college or friends or workmates, lived here together, led a meager life of economizing on food and clothes."Here we have 70% people who do not have the habit of eating breakfast, and some occasionally eat one meal a day."Those just walked out of the ivory tower who had been the God's favored children, and now many of them are hungry with barely enough money for immediate needs.They construct their love and future dreams on screen and on network.Little Bai got home after eight and he said on his way home, he waited long time for the bus, "It's often like this, there is no way meeting peak time hours." Little Bai and Little Meng were schoolmates at the college and they came from rural areas in Henan province.Little Bai is 24 years old this year.He is the eldest brother and one of the main bread-winners in his family, having to supplement the family's annual income of RMB 5000 per head.The family has two middle school age daughters to support too.Little Bai has a monthly wage of 1600 Yuan, he's got an IT degree from a university in Zhengzhou and is now working in a mobile phone marketing company.He has two years of working experience from three jobs.Even a promise of RMB100 wage increase would make him change jobs.He has to give 500 Yuan monthly in support to his younger sisters thereby reducing the financial burden of his parents."I don't have other ways, only wish to earn more money, and live frugally until my two sisters get to graduate and start to earn a living.I then can make a personal plan for myself," he said.Little Meng, 25 years old, after graduation, he went to Beijing and lived in a small-scale "ant tribe village" where was near the Tang Jialing and wished to find a job in Zhongguancun IT market.As soon as he came into Beijing he realized that finding a job in Beijing was far too difficult.Even a customer sales position in front of Zhongguancun electronic buildings, was hard to get as there could also be a dozen or more candidates competing for just one post.After having lived in Beijing for four month, he spent all the money his parents provided, and did not find a suitable job.Eventually he had to return to the Henan provincial capital city of Zhengzhou and then started living together with Little Bai.In order to make end meet in Zhengzhou, Little Meng rode a bicycle along the street selling milk, to help friends selling computers in Zhengzhou electronic market and a half year later he got his present job of a technician in a network company in Zhengzhou electronic market.He said, "Now, my job is suited to my special training, getting a fixed monthly basic salary of 1000 Yuan, at the end of the year I will have 8000 to 10000 Yuan sales commissions according to performance"."I have got the same living standard as other graduates, I'm very satisfied", he added.He does not approve of the frequent "job-hopping"."When you have not found a suitable new job, if you quit the present one, you might not be able to find a new job for two months, and you might go hungry.Summing it all up, it's better to do original work where annual income is higher"."This is my valued experience I got from Beijing to Zhengzhou along the way", he said.He is the only child in his family.His parents planted vegetables at home and the annual income was more than 20,000 Yuan.He said, "Now, I hope to meet a girlfriend who love me, to earn enough money, to sell my parent's hometown house in the countryside and then I can buy an affordable flat in Zhengzhou city.After I get married I will ask my parents to move to Zhengzhou and whole family will live together.In this way we all will become city dwellers"."My God, when could I earn a room in Zhengzhou, I still have to depend on my parents at home to have vegetables a good harvest, and sell the harvest a good price!" he yelled. Meet the Smart Guy, Li Yan in Zhongguancun, in Beijing We met Li Yan in front of the electronic building in Beijing Zhongguancun Science and Technology Market.The guy was more than 1.8 meters tall, black hair with hair gel, in western attire and shiny leather shoes, white shirt with straight tie, like a movie star.He is 28 years old and comes from Tianjin.Since he graduated from Nan Kai University in Tianjin with a Master degree in Computer Science in 2008, he has come to Zhongguancun Science and Technology Market to work as a computer sales man.His job was to stand in front of the door of electronic edifice devoted to attracting customers.Li Yan had once lived in Beijing's famous "ant tribe village" Tang Jialing and he now has been displaced in the north Tang Jialing of the urban-rural zone newly formed "ant tribe village" by Tang Jialing demolition at the end of 2010.He used to take a crowded bus to and from his workplace every day.He said humorously, "The 'ant tribe' living and working in Beijing are just like looking to find a room in the crowded bus, getting a room will allow you to survive, but the word is : "Squeeze!"Li Yan's monthly basic salary was 1000 Yuan, plus sales commission; the monthly income could reach 1800-2000 Yuan.With two friends shared one bedroom and each person shared a monthly rent of 700 Yuan.After ordinary livelihood expenses for meals and daily expenses there was not much money left.He said, "Not too bad, the company offers a 5-yuan-meal free for lunch.I do not have breakfast and my dinner cost is below 10 Yuan, saving money on food, belly full is ok.The rest of my income goes to buying my clothes".We knew what he means: the clothes were his suits."In high-tech market sales, you can't dress like a beggar to attract customers.While, in fact, we are not richer than the beggars", "if you guys could buy a laptop in our store today, I would accompany with you to do a survey, whether to go to Zhongguancun the 'ant tribe' working place or 'ant tribe village' the 'ant tribe' dormitory, no problem.I don't care about today's sales revenue, but you can't affect my performance in this month!Otherwise, I would not get on the bus."Li Yan's hometown is in the suburbs of Tianjin city, his parents used to be farmers but they have now sold their land due to city's expansion and kept a sum of money.His family has become the citizens of Tianjin City and their income is higher than common worker's family.His family was no longer as poor as before and it would not have been difficult for him to find a job in Tianjin, his hometown that could certainly be better paid than the job in Beijing. With Dreams, the Macroscopic Reasons and Microcosmic Reasons in Arguments "The reasons of we get into trouble not due to ourselves.We stick to difficult position, because we have a dream", that is the "ant tribe" general state of mind.A net citizen called Yuan Xiner said in her blog, "Here, what I want to say is: don't take us the word 'struggling' all the time.Why?Because of I find that most people are accustomed to use an 'individualized thinking' to think about problems.What is the 'individualized thinking?'When the 'ant tribe' problem in front of us, a lot of people subconsciously offer solution to the problem is that, always ask for the 'ant tribe' through their own efforts and struggling.It seems that, the appearance of 'ant tribe' social phenomenon is the result of the university graduates not working-hard enough, lack of the fighting spirit and they always think that as long as we work hard, this kind of social phenomena will disappear".Most of the "ant tribe" considered that their regardless plight not to be a result of personal causes, but are due to the macroscopic reasons of city's high-price housing, university education system and university graduate employment policy. These young people insist on struggling in the cities because there are more and relatively equal opportunities for development.Most of the "ant tribe" deemed that they are the generation not only with more knowledge, technology and culture, but also with ideals and dreams."The aim of going to university is leaving the countryside and become city people; therefore the university graduates should work in the city.We would be the masters of the future city and the development of cities relies on us".In chapter fourth of "2010 China 'Ant Tribe' Living Conditions Survey Report -the Ant Tribe Sense of Social Injustice Research" (Lian Si, 2011) showed that 79.5% of the respondents considered themselves in the "low" and "bottom" segment of the society.There is a great contrast for those university graduates once known as God's favored children.While, for the future, most of them not only have high expectations and ambitions, but are also very self-confident.Nevertheless, whether or not they will become upwardly mobile and get into the upper class of society and become the elites or quasi-elites they have a great deal of doubts.Dr. Lian analyzed, "In such difficult conditions, they can adhere to the big cities to struggle, because they have a dream in their hearts."The Zhongguancun smart guy, Li Yan said, "Working and living in Beijing, my living room is not as good as student's dormitory in my university.I eat junk food every day and plugging away a lifetime would not afford me a house, let alone getting married and raising children.Beijing is not my home.My main reason of sticking here is to get more sales and management experience in Zhongguancun, and in the process accumulate a number of commercial contacts.I will accumulate two years' experience in Beijing and then return to my hometown.I can use my parents' money to open a computer and network company and will be my own boss.Anyway, I should have a room or a flat and get married, you guys know my age."In the discovery of this survey, less than 10% of people thought that the present living predicament is shaped by themselves.The majority of respondents said, "Someone considered that the 'ant tribe' gathered in the cities due to our own problems, which we do not agree.Most of us from rural and small towns, and those farmers and unemployed workers in our hometown have also migrated to the big cities or the southern and eastern well-to-do provinces becoming migrant workers, so we could go back to do what, ah?Although we mind that 'do not get a decent position, no face to see our parents', after all the big cities have more opportunities and also the promise of greater dreams".The survey also finds that, in the past few years the "ant tribe" phenomenon has spread from first-tier cities to the second and third tier cities, perhaps a few years later will penetrate through to the small cities, while the first-tier cities of the "ant tribe" problem will not have been alleviated effectively.Therefore, the "ant tribe" issues still need more research in the social macro policies and a concerted effort by the government and the community to solve the issue.Emphasizing on the "ant tribe" to strengthen personal striving is necessary, but success is just a phenomenon of a few of individuals or small groups that cannot solve the "ant tribe's" overall survival problems.Through hard work and efforts they can survive in the big cities and have made an indelible contribution to the prosperity and development of the cities, that has proved their ability and value in the community.In any case, as they were from small towns and villages with a dream to go the universities in a big city and thus staying in the big cities is only the continuation of their dreams.In fact, each of them has been struggling hard to realize their dreams! Have a Sense of Social Responsibility, Sensitive to the Social Injustice Events Although survey finds that the "ant tribe" has an acute awareness of social unfairness, less than 20% of the "ant tribe" group feels that current society is fair.This survey team by using of correlation analysis methods also shows that there is a high correlation between the sense of social fairness and family economic status and personal monthly income.The higher of family economic status and personal income, the higher the sense of social fairness, and more optimism in judgment of social justice trends.The majority of the "ant tribe" for example, comes from economically less developed regions and among them there was a proportion of 54.7% from rural areas and 20.7% from county-level towns respectively.They are truly the "poor second-generation".They studied hard to go to university and are now saddled with high expectations by their family.Even in university, they were still much more diligent than children from the rich families.They witnessed the classmates -the "rich second-generation" or the "power second-generation" -has open and easy access to good jobs, have the means to purchase big houses and new cars after graduation.All this while, for himself, he lives in an awkward situation, causing his discontent of the gap between the rich and the poor and social injustice.No wonder some people concluded that, "Struggling a decade is not as good as having a good father."(Qiu Qingjian, 2010) During the investigation, when it came to their temporary predicament, many of the "ant tribe" showed a strong sense of dissatisfaction on the "official second-generation" and "rich second-generation". The "ant tribe" attributes to the phenomenon of social inequality rather than personal factors.There ranks in the top three factors are "power", "family background" and "social class".They are extremely dissatisfied with the abuse of power, corruption and all kinds of rent-seeking behavior.They have a strong sense of injustice due to nepotism in the public sector."Now, in the society it can be inherited not only the wealth, but even including power."They generally believe that, if the "rich second-generation" to inherit their father's wealth is reasonable and legitimate.So, now the common existent phenomenon of the "official's second-generation" to inherit their parent's power, that cannot be accepted by the "ant tribe" group (Shi F., Feng L. and Liu Q., 2010).The "official second-generation" and the "rich second-generation" rely on their parental relationship to find a good job and even specially arranged official positions, that makes the "ant tribe" mentally feel lopsided and generate the strong feelings of hostility towards them.Dr. Lian said, "The 'ant tribe' is a young educated-generation, if their discontent accumulates widely, the social instability will be largely gathered.It is very terrible."(2011) The "ant tribe" is a new generation of knowledge, culture, technical workers and they are also the net citizens.They pay close attention to the issues of society and keep in sync with the trend of social development.They are highly attuned to social responsibility and advocate fairness and justice, but their reflection is too sensitive to public events and they are vocal and severe in expressing it.Especially the phenomenon of judicial injustice, the events of the "official second-generation" and the "rich second-generation", their emotional reactions are very strong.Among the disadvantaged groups, it is this particular characteristic that makes it possible for "ant tribe" to really question the unfairness in the society and to monitor government behavior (Sun Y. and Tang Y., 2010).With the help of modern online media they fiercely attack social injustice and sometimes do not even figure out the truth or authenticity of the events they have shown their clear-cut position on.While these views and positions are also very easy to stir up the resonance of the entire group of "ant tribe".Xiao Liu, in the Zaozhuang "ant tribe village" of Zhengzhou city said, "When we gather online support or criticize a public event, the focus point is the event itself, rather than the truth or falsehood of the event.We often express our point of view in the first instance and sometimes we may find out that we were wrong after the clarification of the event.However we don't think we are wrong, while the original information pertaining to the event posted online is inaccurate." He is right, they can just express their claims and demands of social equity and justice, without verifying the facts of the event, or say that they are not obliged to investigate the truth of the events.In any case, in recent years, whether the events of miscarriage of justice exposed by the network or the public opinion toppled the corruption cases of public officials, or called for the return of public morality in the event of "Little Yueyue" in Guangzhou (http://baike.baidu.com/view/4682882.htm), the "ant tribe" was the network public opinion force. The "ant tribe" is a generation with the spirit of great love and mutual help.They can help and care about each other like ants do, like little Bai and Little Meng, with the help of Little Bai, Little Meng could settle down in Zhengzhou and get a job.Now when Little Bai encounters economic difficulties, Little Meng will lend a helping hand.The "ants" have proved their love and resolute spirit by their own actions.The interviewee encountered a lot of people who said that they took part in volunteer teams at 2008 Olympic Games in Beijing.Many of the elder members of the "ant tribe" said they had voluntarily donated to Wenchuan earthquake stricken area in Sichuan Province in 2008.They had donated at least 10 Yuan or more, and some of them have donated more than once to the reconstruction of disaster areas in China.They have showed their love and compassion in many events, which reflects the noble spirit of compassion and understanding and the moral side of this social group. The Older, the Bigger Psychological Pressure and the More Depressed Just out of university, the vast majority of the "ant tribe", although they have met the contradiction between ideals and reality, driven by their aspirations, believe that their own efforts can change their life predicament and look forward to achieve a middle-income lifestyle within the next five to ten years.About 70% of the "ant tribe" pursues ambitions for their future, and are certain of success.There are more than 50% who believe that they will become the elites of society in 5-10 years.However, as the cruel reality unfolds itself before them via fierce labor competition and rising living costs, they realize that only a select few can become the elites, and those reaching white collar class prosperity with housing and a car would not exceed 20%.When they find through several years' struggling, that their own economic disconcerted status has not changed much and as they get on in years, the problems of housing needed to be eligible for marriage and family are becoming more urgent and the pressure as well as anxiety increase.Without holding a decent position, the frustrations of love failure, impulses of sexual deprivation and loneliness are intertwined together, resulting in depressions, sometimes manic, anxiety, irritability, hostility and even despair.Optimism and pessimism coexists, anxiety and expectation mix together.So, here the "ant tribe" members over 30 years old are rarely staying and they are leaving the "colony" for many kinds of reasons and for different options.Some of them may toast to the success, while others may leave with tears of failure.Whether the elder "ant tribe" psychological state is in critical condition, as well as after leaving the post-"ant tribe" living, their mental status and the potential effects, all of this would require a separate research in the future (Deng Haijian, 2010). Improving the "Ant Tribe" Living Conditions by Increasing the Level of Personal Financial Management The "ant tribe" is a group of new generation in modern society and it can be said this is a new class of the contemporary Chinese society.They are striving to integrate into the contemporary Chinese society.To some extent, they are the social group that has not yet firmed their foothold in the community and they are the proletariat.How can they change their living conditions and get rid of their plight in life?This is clearly a complex socio-economic challenge and a political issue.As individuals, how to efficiently use own existing meager financial resources, to avoid unnecessary expenditure and waste and to individually accumulate some entrepreneurial capital, in order to leave the embarrassing "ant tribe" life behind.That is a very practical matter.Here, the author from the perspective of financial management gives advice, expecting -through financial knowledge -to help the "ant tribe" improving their financial management capability with the aim of improving the living situation to a certain extent.First, "ant tribe" shall early grasp the skills of "opening sources, throttling and accounting", the three basic aspects of magic weapon in financial management, and then through basic financial tools, such as savings, bonds, funds and insurance increase income, to improve the economic situation and be better off. What is Financial Management? Financial or property management, refers to the individuals for personal property or the family property management, that is the individuals according to own current actual economic condition, setting to achieve economic goals, in a limited time frame using one or more categories of financial investment tools, through one or more path to reach his economic objectives, planning scheme or solutions. Why Finance? Financial management can help the "ant tribe" to achieve the following purposes: 1) Accelerated path to independence.The "ants" are fresh college graduates and stepping alone into the society, still wet behind the ears.This group includes those living beyond their means sponging off their parents to make ends meet.If you join the team of financial management enthusiasts you could severe the reliance on funds from your family and achieve economic independence quickly. 2) Capital protections.If the "ant tribe" can properly plan managing their own property, it will implement their own limited assets not devalued in any circumstances, even on the basis of capital protection realizes value added. 3) To prepare for the unexpected."It is the unforeseen that always happens".When misfortune occurs unexpectedly, the financial measures taken in advance can hedge "ant tribe" himself to cut losses to the minimum. How to Do Accounting? The objective of accounting is for providing a basis of scientific analysis of one's financial conditions.The "ant tribe" shall develop a good habit of daily bookkeeping and each of individual income and expense shall be recorded in detail to make a monthly summary.Incomes can be divided into three items such as wage income, financial income and other income.Expenditures can be divided into clothing, food, housing, traveling and entertainment -five aspects.If you make a monthly summary, after three months you can approximate your average monthly expenses.Apart from the monthly expenses you shall also need to have funds prepared for emergency that in general should equal six times monthly expenditures. Learn about Several Financial Tools The "ant tribe" may focus on the understanding of financial tools of savings deposit, funds, bonds and insurance.Savings deposit is a financial tool with the aim of setting aside cash -the most liquid asset -and receiving interest.Savings deposit can basically be divided into current deposit and fixed term deposit.Deposit interest calculation formula: Interest = the principal × interest rate × deposit term Savings deposit is one risk-free financial tool, but is one that can gain the minimum return.Savings deposit has liquidity characteristics and the emergency reserves can take as a time deposit to the bank for a possible period of want or need.Investors diversify their funds through the purchase of fund shares or income certificates, bringing together investor's dispersed funds by professional managers to invest in stocks, bonds or other financial assets.The investment income in form of dividends allocated to the investment shared holder is a kind of benefit-sharing and risk-sharing financial product.Funds, in accordance with the investment objectives, can be divided into equity funds, bond funds, mixed funds and money market funds, etc. Equity funds carry high-risk and high-yield characteristics, investing mainly in shares and stocks, the proportion of the investment should not be less than 60%.Bond funds carry low-risk-low-return characteristics.There are various types of bonds that could be main investment target and the proportion of bond investment should be more than 80%.Mixed fund invests mainly in stocks and bonds in the meantime, it is through the configuration of the different asset classes that one can achieve risk and return balance.Money market funds with the characteristics of low-risk, low return and high liquidity, that is an instrument of investment in money market. Funds scheduled investment is referred shortly to a regular fixed investment in a fixed time (such as on the 8th of each month) and a fixed amount (such as 500 Yuan per month) to invest in a specified open funds, such investment can average cost and decentralize risks.Time deposit and funds which yield is better?We assume that the investment period is 10 years and the principal amount is 10,000 Yuan, the deposit has an average interest of 5.5% per annual and the funds has average annual earnings of 10%. The interest of 10-year time deposit = 1 × 10 × 5.5% = 0.55 The income of the Funds = (1 +10%) 10 -1 = 1.59 The income of funds is 2.9 times to the time deposit.However, remember that the higher the return, the greater the risk.We commonly use "100 minus ages" to determine the proportion of personal asset investment risk products.The "ant tribe" mainly is under their 30 years of age, so most of the "ant tribe" can invest their 70 percent of funds in higher risk funds products (Li Xiangfeng, 2011).Although the "ant tribe" has a lower monthly income, the starting point of the scheduled investment funds is very low, just a minimum of 100 Yuan.The scheduled investment funds should be chosen from good funds company in operation.Stock funds and index funds are also fit for fixed investment; those investments after five years shall have an annual return of over 10%. Bond refers to an interest-bearing debt certificate, the issuer is committed to a future date (usually after several years of the issuance of bonds) to repay the principal and interest payable, so the bonds are called "fixed income investment", that usually issued in accordance with certain procedures by the State (including central and local governments), financial institutions, international organizations and commercial enterprises and other institutions.These institutions through issuing bonds raise funds within the community.The advantages of bonds are: a. High-yield.Bond yields are higher than bank interest for the same period in the same sum.b.Low-risk.The yield of Treasury bonds is very stable and the risk of financial bonds slightly higher than the Treasury bonds, but the interest rate is generally higher than the Treasury bonds.Corporate bond yield is generally higher, but the risk is relatively high.c.High-negotiability.Bond is more liquid, you can get timely cash by ways of exchange and pledge in advance. The main risks of bonds are: a.Interest rate risk.When the banks raise interest rates, the price of bond will be reduced.b.Management risk.The issuing unit of debentures manages missteps and makes mistakes in the course of business, that will result in asset value decrease and leaves bond holders with losses.c.Bond default risk.The company issuing the bonds can repay timely interest or principal and the general risks are associated with the company operating in poor conditions or bad reputation.d.Reinvestment risk.Purchasing short-term bonds, but not to purchasing long-term bonds, that is likely to face investment risk.The "ant tribe" should learn how to select the appropriate bond varieties according to their own situation, and implement decentralized investments and diversified investments so as to effectively reduce investment risks. Insurance is a form of contract establishing bilateral economic relations and the insurance fund is established by payment of premiums.Insurance contract within the scope of the provisions of the losses caused by disasters and accidents, for which the sum insured, is paid.Insurance can be divided into three categories: home insurance (also known as property insurance) and personal insurance.The "ant tribe" needs to focus on the personal insurance at present stage, the property insurance you can temporarily not take into account.According to the contract, the personal insurance shall undertake the following four kinds of insurance responsibility: death, disability, illness or reaching age duration of the contract.According to insurance coverage, personal insurance can be divided into life insurance, health insurance, personal accident insurance, etc. Life insurance is on human life as the subject of insurance mark, which is based on a person's life and death (or terminal disability) as a condition for payment of the insurance benefits/claim, it is referred to as life insurance, normally includes live insurance and death insurance and endowment insurance.Accident insurance refers to the policyholder and the insurer through agreement when subjected to accidental injury and disability or death, the insurance company pursuant to an agreement of payment of the insurance benefits to injured persons or beneficiaries of insurance contracts.Health insurance refers to because of disease or childbirth leading to disability or death of the insured person; the insurance company pays the insurance money or compensation for medical expenses of the contract.The disease is the direct cause of the insurance accident, therefore, health insurance, also known as illness insurance. Here I mainly introduce the investment-type insurance and the investment-type insurance is a branch of the life insurance, which is a kind of innovative life insurance that was originally designed by western countries in order to prevent fluctuations in the economy or inflation, on the long-term life insurance losses.Since then it has transformed into a new financial investment tool of risk-sharing and profit-sharing between customers and insurance companies.The current investment-type insurances can be divided into three categories: participating insurance, investment linked insurance and universal insurance.The participating insurance is a kind of insurance that after the insured pays the premium, they can enjoy part of the operating results of the insurance companies.But it is low risk low return which is suited for modest-profit seeking policyholders who give priority to safeguarding their funds.The universal life insurance with some characteristics of the participating insurance, it set a minimum income security and share operating results together by insurance companies and customers.Its premium payment is more flexible thus making it suitable for the low risk tolerance policyholders who are looking to have more options.Investment-linked insurance is a type of insurance which is the insurance linked investment.Insurance in contrast to investment has an emphasis on capital protection.That is suitable for those investors who have rational investment philosophy and pursuit assets that gain high-yield, but it also requires the investors to have a high risk tolerance.In purchase of investment-oriented insurance, you should consider the premium paid level, but you cannot fully experience in the insurance company statistics to evaluate the benefits of insurance, which also need to refer to bank deposits, the return rate of bond market and funds market to take a synthesized consideration. Doing Part Time Jobs and Setting up an Online Shop to Broaden Income Sources The income of the "ant tribe" is low.With limited income, broadening sources of income and reducing expenditure is especially important.By means of a detailed accounting record, you can figure out your higher costs and unnecessary expenses, so as to change your living habits and significantly reduce unnecessary spending in order to achieve the aim of sound financial management.To open sources also need the "ant tribe" to look for suitable source of income beyond their regular jobs and part-time job may be a practical choice.To select a proper part-time job, you should first analyze the characteristics of your own work, such as work time, nature of work and so on.Through the investigation of 192 "ant tribe" member we obtained the conclusion: 80% had working hours from 8 am to 6 pm, evenings and weekends were free time.While, night time is the peak time for activities of young people (Zhang He, 2011).Along with the increasing popularity of Chinese network technology, online shopping, transactions, payment, a new model of e-commerce development is rapid.Run an online shop can achieve low cost and low risk passive income that is suitable part-time occupation for "ant tribe" members.Through the survey, there was more than 60% of the "ant tribe" who run an online-shop on the Taobao, the largest Chinese online shopping website, which can earn a stable monthly income of 3000 Yuan stably after operating for about six months.This revenue to promote the "ant tribe" for early economic independence is of great significance. Clear Focus and on Target, the "Ant Tribe" Financial Case Everyone hopes for a better life that is not only to meet the basic needs for survival.As a result of the "ant tribe" with a limited wage income and difficulty in wealth accumulation, therefore the individual wealth accumulation -to a certain extent -should shift its focus the preservation and appreciation of the assets.Effectively applied wealth to produce investment income makes a successful transition to achieve financial freedom.Now, you have been learning a variety of financial instruments, then how should you plan and design your personal finance?We firstly have to study the following case: Miss Lee, 24 years old of this year, just graduated from the University for two years, now she is working in a joint venture in Beijing.In general, Miss Lee does not have much concept of financial management and her monthly salary is not high.After the expenses of rent and food paid and a portion of money sent to her parents, she is basically with no money left.Other consumer loans 0.00 Funds 2.00 Total 9.00 Total 0.00 The net assets of family 9.00 Miss Lee was alone in struggling and there was not much balance, so she only had a few assets.In the meanwhile, Miss Lee was a novice to financial management and did not know how to improve her financial position.Her monthly salary was fully consumed, so she was muddleheaded over her spending each day.Miss Lee did not know how to buy insurance and she did not know how to select suitable financial products and did not have any plans for her future life either, etc.After working two years, Miss Lee felt deeply the need to strengthen her revenue and expenditure management.Therefore, she urgently needed to make a financial plan for herself, to step-by-step set her own financial situations on the right track. Firstly, you shall learn and master operating financial management.Miss Lee's discretionary income was sent to her parents and such course of action was not desirable.She should be completely on her own to manage the discretionary income.On one hand, you to improve the financial management capability; you must continuously learn and practice.On the other hand, the money was sent to her parents for custody and safekeeping, but the benefit was not too high.For young people, they have a strong ability to bear risks and are suitable for active financial management.It can be appropriate for them to participate in some high-risk investments in order to win higher returns.We often use the "100 minus ages" to identify the proportion of individual stocks, such as the rights and interests of current investment tools.Miss Lee, 24 years old, her assets can be used to invest in the stock market and the stock type funds, in which the investment ratio can reach 76%.Miss Lee can choose some stock type funds to create the fund portfolio plan in order to increase revenue.If one only rely on the savings to accumulate wealth, which not only would be relatively slow, but also could not follow the increment speed of the CPI and inflation. Secondly, the throttling plan is very important.Miss Lee's daily spending without planning, it is also necessary to change that.She should immediately develop a bookkeeping habit and make a list or a summary of her daily expenses to produce detailed statistics within two to three months.Then carry out the adjustment and improvement so that the surplus can be converted to capital available for investment.Then, implementing changes one by one, the end result could give you a big surprise. Thirdly, Miss Lee should make an insurance plan.As her income is not high and as an "ant tribe" there is no family burden, but good health is capital, so you should pay attention to your own health and accident insurance.For example, one of the accident insurance requires an annual insurance premium of 200 Yuan for the sum assured 100,000 Yuan coverage.It is worth to take into consideration.The amount of accident insurance coverage, that with 10 times as your annual income is advisable.In the health insurance options, now there are increasingly occurring serious illnesses at younger age, so it's advisable to buy serious illness insurance as early as possible.Miss Lee can choose pure protective-type insurance of the major illnesses, which annually requires a 400-yuan premium for 100,000 Yuan insurance coverage amount. In short, for the "ant tribe", whether you choose an investment tool or make a protection scheme, you should follow the "clear focus, have a definite objective in view" principle, only through the "grabbing and dropping" skills can effectively utilize the limited economic resources.That is "many a little makes a mickle, prevent trouble before it happens", in order to create a safer and more prosperous future for ourselves.Pan, C. G. (2011) Notes Note 1.The word "diaosi" originated in the Baidu.com'sPost bar (a top Chinese bulletin board system) of soccer player Li Yi.There, fans of Li, who are called Yisi in Chinese, not only talk about soccer but moan about their lives, work and relationships.In just two months, the latest buzzword from the Internet, diaosi, has spread so much that it can be found everywhere-from online forums to micro blogs.Many people even call themselves diaosi.It's not the first time a term has gone viral on the Internet in China.However, the word diaosi is so popular that it has become a cultural phenomenon. The word diaosi was coined first by single, young men who feel they have dead-end lives.Generally, men in this category don't earn enough, are not good looking, and have difficulty winning promotion.Unlike their upper-class contemporaries, they lack influential families, useful social networks for their careers, and most importantly, suitable women to marry. Many young men call themselves diaosi because they feel they are among the lowest echelons of society.They suffer low self-esteem and have stopped trying to improve their lives. Note 2. The term "woju" is a latest network buzzword due to a popular TV series show recently reflecting the contemporary "ant tribe" for the housing difficult.The word woju in Chinese is an old term from 1000 years ago in the Three Kingdoms of Wei and Jin dynasties and it originally referred to snail house but gradually changes into "dwelling narrowness" or "humble abode".The "woju" term in the "dictionary of modern Chinese", either as a noun, means cramped accommodation and it can also be a verb, means living in a small house. Figure 2 . Figure 2. Living area of ant tribe members	2018-12-05T20:04:02.734Z	2013-01-28T00:00:00.000	{ "year": 2013, "sha1": "df0b0bd7c233225a639a3b290eee2c20421c1ee7", "oa_license": "CCBY", "oa_url": "https://ccsenet.org/journal/index.php/ass/article/download/24350/15405", "oa_status": "HYBRID", "pdf_src": "Anansi", "pdf_hash": "df0b0bd7c233225a639a3b290eee2c20421c1ee7", "s2fieldsofstudy": [ "Education" ], "extfieldsofstudy": [ "Sociology" ] }
13913683	pes2o/s2orc	v3-fos-license	Functional assessment of the NMDA receptor variant GluN2A R586K Background: The N-methyl-D-aspartate receptor (NMDAR) is an ionotropic glutamate receptor that has important roles in synaptogenesis, synaptic transmission, and synaptic plasticity. Recently, a large number of rare genetic variants have been found in NMDAR subunits in people with neurodevelopmental disorders, and also in healthy individuals. One such is the GluN2A R586K variant ( GRIN2A G1757A), found in a person with intellectual disability. Identifying the functional consequences, if any, of such variants allows their potential contribution to pathogenesis to be assessed. Here, we assessed the effect of the GluN2A R586K variant on NMDAR pore properties. Methods: We expressed recombinant NMDARs with and without the GluN2A R586K variant in Xenopus laevis oocytes and in primary cultured mouse neurons, and made electrophysiological recordings assessing Mg 2+ block, single-channel conductance, mean open time and current density. Results: The GluN2A R586K variant was not found to influence any of the properties assessed. Conclusions: Our findings suggest it is unlikely that the GluN2A R586K variant contributes to the pathogenesis of neurodevelopmental disorder. Introduction The N-methyl-D-aspartate receptor (NMDAR) is an ionotropic glutamate receptor that has important roles in synaptic transmission and synaptic plasticity, and has been implicated in a range of neurological disorders (reviewed in (Paoletti et al., 2013)).It is remarkable for its high Ca 2+ permeability (MacDermott et al., 1986), requirement for glycine as a co-agonist (Johnson & Ascher, 1987) and block by Mg 2+ , which underlies the NMDAR's voltage-dependence and hence role as a "molecular coincidence detector" (Mayer et al., 1984;Nowak et al., 1984).NMDARs are heterotetramers comprised of two GluN1 subunits and two others, of which GluN2A and GluN2B are the commonest in the mammalian forebrain (Monyer et al., 1994;Watanabe et al., 1992).GluN2A is predominantly expressed post-natally in rodents (Monyer et al., 1994;Watanabe et al., 1992) and humans (Law et al., 2003), where it is thought to frequently associate with GluN2B subunits to form triheteromeric GluN2A/2B NMDARs (reviewed in Wyllie et al., 2013). Recently, advances in sequencing technology have allowed a large number of de novo and inherited GluN2A mutations to be identified in individuals with a range of neurodevelopmental disorders, including epilepsy, intellectual disability, autism and schizophrenia (Burnashev & Szepetowski, 2015).Some of these mutations are gene disrupting, likely resulting in haploinsufficiency, and so far all associated with epilepsy.Other mutations are missense, potentially resulting in a NMDAR with altered function.Identifying the functional consequences, if any, of such mutations may allow insight into key mechanistic pathways underlying the neurodevelopmental disorders with which they are associated.However, it is likely that some mutations are not relevant to the disorders of their carriers, leading either to receptors with no altered function or alterations that are easily compensated for.Distinguishing between mutations that may be pathogenic, and those that are coincidental, aids mechanistic understanding, therapeutics, and genetic counselling. GluN2A R586K was first identified as being maternally inherited in a person with intellectual disability and either epilepsy or an abnormal EEG (Endele et al., 2010).The phenotype of the proband's mother was not reported.GluN2A R586K has undergone no previous functional assessment.However, in silico prediction tools found it to be "possibly damaging" and "non-neutral" (Endele et al., 2010), albeit it is a conservative, within-class amino acid substitution, with both arginine and lysine being positively charged.The substitution alters a residue in the cytosolic loop linking the subunit's M1 and M2 pore domains (Monyer et al., 1992), which has so far not been linked to any particular function.However, the loop's proximity to the pore suggests that any functional consequences could include an impact on ion permeation and/or block of the pore by Mg 2+ ions.In the present study, we performed the first known functional analysis of the GluN2A R586K variant.We made electrophysiological recordings from NMDARs containing mutant subunits expressed in Xenopus laevis oocytes and primary cultured neurons, allowing us to assess the variant's impact on block by Mg 2+ ions, single channel conductance, mean open time and NMDAR current density.We found that the GluN2A R586K variant did not affect any of the parameters assessed. Amendments from Version 1 Introduction: We have added further information on GRIN2A genetic variation in neurodevelopmental disorders, however, we have kept this brief as our aim is not to review the area.We have added further content in line with the referees' suggestions. Results: The concentration of extracellular Mg 2+ used is the most relevant one as it is approximately physiological.Although repeating the experiment with different concentrations of Mg 2+ could strengthen our conclusion of no effect on the potency of Mg 2+ block, it seems unlikely, given the location of the mutation, that voltage-dependent block by extracellular Mg 2+ ions would be affected. We inadvertently included text in the Methods referring to Mg 2+ block at different holding potentials; these experiments were not carried out in the present study and we apologize for our error.We have updated the methods to reflect the fact that Mg 2+ was only studied at one holding potential. Desensitisation: pyramidal cortical neurons are relatively large cells with extensive dendritic arborisations and as such are unfortunately not amenable to studies of desensitisation due to the inability to have rapid solution exchange. Closed time analysis: we did not analyse or report closed times or open probability due to uncertainty in knowing how many channels were present in patches.Open times remain valid to analyse because sections with multiple channels open at once can be discarded. Discussion: We have added further content in line with the referees' suggestions, emphasising that alternative functional impacts may exist and the limited extent of our study. REVISED Two-electrode voltage-clamp recordings cRNA for wild type and mutant subunits was synthesized from linearized plasmid DNA as runoff transcripts using the T7 polymerase mMessage machine RNA synthesis kit (Life Technologies Ltd, Paisley, UK).Each oocyte was injected with 3.7-9 ng of cRNA, comprising a 1:1 molar ratio of GluN1 and GluN2A diluted in RNAse free water. Stage V-VI X. laevis oocytes were obtained from the UK Xenopus centre (Portsmouth, UK) and from Diaclean (CastropRauxel, Germany), collagenased (200 units/ml for 60 minutes), then manually defolliculated prior to injection.After injection, oocytes were placed in separate wells of 24-well plates containing a modified Barth's solution with composition (in mM): 88 NaCl, 1 KCl, 2.4 NaHCO 3 , 0.82 MgCl 2 , 0.44 CaCl 2 , 0.33 Ca(NO 3 ) 2 , 15 Tris-HCl; adjusted to pH 7.35 with NaOH.This solution was supplemented with 50 IU/ml penicillin, 50 mg/ml streptomycin and 50mg/ml tetracycline.Oocytes were placed in an incubator (16-21°C) for 24-48 hours to encourage receptor expression and subsequently stored at 4°C.Recordings were made 48-96 hours post injection.Two-electrode voltage-clamp recordings were made at room temperature (18-21°C) from oocytes that were placed in a solution that contained (in mM): 115 NaCl, 2.5 KCl, 10 HEPES, 1.8 BaCl 2 , 0.01 EDTA; pH 7.35 with NaOH.Recordings were made using a GeneClamp 500B amplifier (Molecular Devices, Union City, CA, USA).Current and voltage electrodes were made from thin-walled borosilicate glass (GC150TF-7.5, Harvard Apparatus, Kent, UK) using a PP-830 electrode puller (Narashige Instruments, Tokyo, Japan).Filling with 3 M KCl gave resistances of between 0.2 and 1.5 MΩ.Application of solutions was determined manually.Data were filtered at 10 Hz and digitized at 100 Hz via a 1401 plus analogue-digital interface (Cambridge Electronic Design, Cambridge, UK) using WinEDR software (version 3.2.7;Strathclyde Electrophysiology Software, Strathclyde University, Glasgow, UK).Oocytes were voltage-clamped at -60 mV.Recordings were rejected if the holding current (in nA) was greater than three times the holding potential (in mV), or if the holding current drifted by more than 10% of the agonist response across the course of the experiment. Single-channel voltage-clamp recordings Single-channel voltage-clamp recordings were made at room temperature from outside-out patches pulled from oocytes that were placed in a solution that contained (in mM): 125 NaCl, 3 KCl, 1.25 NaH 2 PO 4 , 20 HEPES, 0.85 CaCl 2 , 0.01 EDTA; pH 7.35 with NaOH.Steady state recordings were made in the presence of 30 μM glutamate and 30 μM glycine.Recording durations varied from 30 seconds to 5 minutes.Prior to recording, vitelline membranes were manually removed from oocytes following placement in a hypertonic solution that contained (in mM): 200 sodium methyl sulphate, 20 KCl, 10 HEPES, 1 MgCl 2 ; pH 7.4 with NaOH.Recordings were made using an AxoPatch 1D amplifier (Molecular Devices).Electrodes were made using thick walled borosilicate glass (GC150F-7.5;Harvard Apparatus) using a P-87 electrode puller (Sutter Instrument, Novato, CA, USA) and their tips fire polished to give resistances of 7 to 12 MΩ when filled with internal solution containing (in mM): 2.5 NaCl, 141 K-gluconate, 10 HEPES, 11 EGTA; pH 7.4 with KOH.Electrode tips were coated in silicone elastomer ("sylgard 184"; Dow Corning, Wiesbaden, Germany) to reduce capacitance.Application of solutions was controlled manually.Data were prefiltered at 2 kHz (-3 dB; 8th order Bessel filter) and digitized at 20 kHz via a Micro 1401 analogue-digital interface (Cambridge Electronic Design) using WinEDR software (v 3.2.7).Patches were voltage-clamped at -60 or -100 mV. WinEDR v3.3.7 was used to idealise the traces (using a transition threshold of 50% of the unitary conductance level and a 100 μs open and shut resolution), to fit Gaussian curves to amplitude histograms and to fit exponential curves to dwell time durations (all curves fitted using iterative maximum likelihoods).Any openings where more than one channel was open simultaneously were discarded.Conductance was calculated by dividing the current amplitude by the holding potential. Whole-cell recordings in cultured neurons Recordings from cultured neurons were made at room temperature with neurons superfused (at a flow rate of 2 mL/min) with external recording solution composed of (in mM) 150 NaCl, 2.8 KCl, 10 HEPES, 2 CaCl 2 , 10 glucose, 0.1 glycine, 0.003 tetrodotoxin; pH 7.35 using NaOH (300-330 mOsm).Transfected cells were identified by eGFP expression (excitation at 470 nm, coolLED pE-100 (coolLED Ltd, Andover, UK)).In total, 150 μM NMDA was applied briefly twice to elicit desensitization, then reapplied until a steady state response was achieved (around 10 seconds of application) at which point 1 mM MgCl 2 was co-applied until a new steady state was achieved.Experiments were then repeated after perfusion with 3 μM ifenprodil for one minute.Application of solutions was controlled manually.Patch-pipettes were made from thick-walled borosilicate glass (GC150F-7.5;Harvard Apparatus) using a P-87 puller (Sutter Instruments) to give a resistance of 2-4 MΩ when filled with internal solution containing (in mM): 141 K-gluconate, 2.5 NaCl, 10 HEPES, 11 EGTA; pH 7.3 with KOH (300 mOsm).Currents were recorded using an Axopatch 200B amplifier (Molecular Devices).Data were filtered at 2 kHz and digitized at 20 kHz via a National Instruments BNC-2090A analogue-digital interface (National Instruments) using WinEDR software (v 3.2.7).Neurons were voltage-clamped at -65 mV, and recordings were rejected if the holding current was greater than 150 pA or if the series resistance was greater than 30 MΩ, or increased by greater than 20% during the course of the recording.Capacitance was calculated by calculating the area under the current response to a 5 mV test pulse plotted against time (giving the charge) and dividing by the voltage of the test pulse.Capacitance provides an estimate of cell surface area.Current density was then calculated as current/capacitance. Data analysis Bar graphs depict individual cells (circles), means (columns) and standard error of the mean (SEM; error bars).R (v 3.1.2;R Core Team, 2014) was used to perform statistical tests.Comparisons between multiple means were performed by ANOVA.Comparisons between two means were performed by independent, two-tailed, Welch t-tests (which do not assume equal variance between groups), unless otherwise stated.The significance level used was p <0.05, corrected for multiple comparisons using the Bonferroni method.In figures, * indicates corrected significance levels of p <0.05, indicates p <0.01 and * indicates p <0.001. Results The GluN2A R586K mutation has no effect on Mg 2+ block or current density We first expressed GluN2A WT and GluN2A R586K -containing NMDARs in X. laevis oocytes and made two-electrode voltage clamp recordings of block by Mg 2+ of glutamate-evoked currents.We found no effect of GluN2A R586K on block by Mg 2+ (Figure 1A-C). To assess for any neuron-specific consequences of the variant, we then used transient transfection to over-express GluN2A R586K in cultured primary mouse cortical neurons.The interpretation of results in neurons is complicated by each cell giving a NMDAevoked response that arises in part from endogenous NMDAR subunits and in part from transfected subunits.The age of culture used (DIV 9) was chosen so that virtually all the endogenous GluN2 subunits were GluN2B (McKay et al., 2012).The GluN2B selective negative allosteric modulator ifenprodil could therefore be used to suppress the contribution of endogenous NMDARs and to confirm the transfection of the subunit of interest (GluN2B diheteromers show 80% block, but GluN2A diheteromers minimal block at the concentration of ifenprodil used (3 μM) (Williams, 1993)). GluN2A R586K expression was confirmed by reduced ifenprodil sensitivity compared to control transfections (Figure 2A-C; Table 1).As with oocytes, no effect was found on Mg 2+ block in neurons (Figure 2A,B,D; Table 1).Current density was also unaffected by the GluN2A R586K mutation (Figure 2A,B,E; Table 1). The GluN2A R586K mutation has no effect on single-channel properties Finally, we used outside-out patches pulled from oocytes to assess the impact of GluN2A R586K on NMDAR single-channel conductance and mean open time.We found GluN2A R586K to have no effect (Figure 3A-D). Discussion In the present study, we investigated the functional consequences of a GluN2A point variation found previously in a person with epilepsy/EEG abnormalities and intellectual disability.Using heterologous systems, we showed that the GluN2A R586K variant had no effect on any of the properties assessed: Mg 2+ block, current density, conductance or mean open time. The lack of functional consequences on permeation and Mg 2+ block seen with the GluN2A R586K variant is in contrast to existing findings for several disease-associated NMDAR mutations, which have been found to have a substantial impact on NMDAR properties, including Mg 2+ block (Burnashev & Szepetowski, 2015).This is not unsurprising given the diversity of mutations found in NMDAR subunits, with some resulting in complete loss of function of the allele, some impacting residues of established functional importance, and some causing conservative changes in residues of unknown function, such as studied here. It was difficult to hypothesise what the likely impact of the GluN2A R586K variant might be, as little is known about the function of the M1-M2 linker region in which it is found.It is possible that the variant affects other properties that have not been investigated in this study.For example, conceivably it could influence trafficking or zinc affinity, which is known to be regulated by C-terminal phosphorylation (Zheng et al., 1998).Further, the impact of the sequence change could manifest through indirect mechanisms, such as an impact on mRNA transcript stability, or epigenetic modifications.Finally, a given variant may only have a physiological effect when partnered with a further variant (compound heterozygosity).In general, it is impossible to exclude that a variant has some functional effect.However, it is certainly possible that the variant does not alter NMDAR function, and that its presence in an individual with neurodevelopmental disorder was coincidental.This is particularly likely as the variant was inherited, and because it has also subsequently been found in two out of 60,706 individuals without severe paediatric disease collated by the Exome Aggregation Consortium (Lek et al., 2016) -although the existence of less severe neurodevelopmental disorders, such as adult onset epilepsy, in these individuals has not been excluded. In conclusion, our limited functional studies suggest that the GluN2A R586K variant is probably benign. 4. 5. The current manuscript describes the functional characterization of a GluN2A mutation, GluN2A in vitro , found in a patient with intellectual disability and either epilepsy or abnormal EEG .A subsequent number of mutations within the GRIN1, GRIN2A and GRIN2B genes were found in patients with neurological disorders . functional characterization revealed that some of these mutations had In vitro major impacts on NMDA receptor expression, biophysical and pharmacological properties (see, for example, Swanger et al. and Serraz et al. ).However, although predicted to be "possibly damaging" by Endele et al. , the functional effect of the GluN2A mutation had never been investigated.Since R586 is located in the M1-M2 loop, close to the pore, they logically investigated the permeation and block properties of the GluN1/ GluN2A mutant and found that magnesium block, conductance and mean open times were unaffected in Xenopus oocytes and cultured neurons.They thus concluded that this mutation is probably benign.The manuscript is written well and the experiments are clearly described in the experimental section.I have however several reservations concerning the experimental work and some suggestions for the discussion, which I hope might improve the quality of the manuscript. Magnesium block: The authors only test the effect of one concentration (1mM), at one holding potential.Although 1mM magnesium is close to the physiological concentration, it seems that this concentration is saturating for GluN1/GluN2A receptors at -60 mV, which may hide differences in magnesium sensitivities between wt and R586K receptors.The authors should test different concentrations and different holding potentials to be able to conclude that magnesium block is unaffected by the mutation.Ideally, full current-voltage relationships should be measured at different magnesium concentrations. Pore opening properties: In their single channel experiments, the authors measure the single channel conductance and the mean open time.What about the open probability?Is it also unaffected by the mutation? Limiting the functional studies to the sole permeation and block properties of NMDARs is very restrictive.Although the mutation is located far from the extracellular domains, where most pharmacological agents act, it could very well affect the pharmacological properties of the receptor.For example, it has been shown that phosphorylation of residues located in the intracellular C-terminal tail could influence affinity for zinc, which binds in the extracellular N-terminal domain .Therefore, to conclude that the mutation has no clear functional effect, the authors should have measured the mutant sensitivity for the agonists glutamate (or NMDA) and glycine, its kinetics of activation and deactivation, as well as its sensitivity for the endogenous allosteric modulators zinc and protons. Relative to point #3, in the discussion section, where the authors state "The lack of functional consequences of the GluN2A variant", they should say "The lack of consequences on the permeation and block properties of the GluN2A variant".Indeed, the variant could have functional consequences that were not investigated in the study. In the discussion, the authors mention that "It is also possible that the variant affects other 6. In the discussion, the authors mention that "It is also possible that the variant affects other properties that have not been investigated in this study".The authors should discuss in more details what properties could be affected by the variant (as previously mentioned, pharmacological properties, but also trafficking, downstream signaling…). Relative to points #3-5, the authors can certainly conclude that the mutation does not induce any dramatic effect on NMDAR function, which would show even at saturating agonist and magnesium concentrations.However, to conclude that it is "probably benign" would require further characterization of this variant.It is now also recognised that such rare coding variants may not necessarily have a functional impact.One such category of coding variant commonly known as Variants of Unknown Significance (VUS), are typically composed of missense mutations (amino acid substitutions) and are often carried in a heterozygous state.Whether such variants have a functional effect, and potentially could be pathogenic, depend on a number of factors including the mutated gene and gene product, genetic sequence context of the variant, and, where in the encoded protein the amino acid substitution occurs.However, certain families of genes have consistently been indicated to contain genetic variation associated with specific diseases.One such family of genes encode for glutamate neurotransmitter receptor subunits and have been associated with risk for developing a variety of brain diseases. References The manuscript describes an functional study of a missense variant within the NMDA type in vitro glutamate receptor subunit, GluN2A (gene which had originally been identified in the GRIN2A), heterozygous state in an individual with epilepsy and/or abnormal EEG and reported by Endele et al., 2010.The variant was reported to be inherited and not arising which is often considered to have de novo, a higher probability of being a pathogenic mutation.Mutations within have previously been GRIN2A reported to cause a range of epilepsy syndromes consistent with an inherited autosomal dominant Mendelian manner and also to contribute to risk for common complex neurodevelopmental diseases such as severe intellectual disability .The question of whether this VUS has a functional effect is indeed very pertinent. We find that the manuscript is written well with the aims and methods appropriately detailed.We have some comments regarding the experimental work and some suggestions for inclusion in the introduction and discussion sections, which we hope will contribute to the quality of the manuscript. Abstract The abstract needs to include the human gene name ( ) and that rare variants have been GRIN2A identified within NMDA subunit genes in individuals with neurodevelopmental disorders as well as in Introduction It would be useful to have more of an overview of how genetic variation within is thought to GRIN2A contribute to brain disease, i.e. genotype-phenotype relationships (e.g.type of mutations, mode of inheritance, 'causative' for disease or associated with increased risk of disease). It should be stated in the introduction that tools had been used to predict the function in silico consequence of this variant (as presented by Endele 2010, supplementary Table 2) and that such et al., prediction tools including others not used by Endele indicate that this variant is predicted to be et al., consistently 'possibly damaging' as opposed to benign.This provides good justification to study the variant.It is also probably worth pointing out the severity of the amino acid substitution in the studied variant, i.e. it is a conservative change. Where discussing NMDARs as a "molecular coincidence detector", it might be worth mentioning the requirement for glycine as co-agonist to glutamate too. Methods The methods were very well explained and are entirely repeatable. Results The first part of the results section shows the sensitivity of the NMDARs containing WT or mutated GluN2A to Mg .The chosen concentration of Mg is indeed around the expected physiological concentration, but because it is more-or-less a saturating inhibitory concentration it would be harder to pick up any changes in sensitivity.Repeating the experiment with a lower concentration of Mg or better still obtaining an IC would give more substance to the claim that this property is unchanged.The same is true for the experiments with transfected neurons where inhibition is very close to 100%. The methods section states that oocytes were clamped at several holding potentials but data is only shown for -60 mV.It would strengthen the claim that Mg sensitivity is unchanged if data for other holding potential is shown and/or discussed.Small changes to the voltage-dependence of inhibition could be responsible for defective physiological function of NMDARs.The responses of transfected neurons to NMDA in Fig. 2A-B may imply small alterations to desensitisation properties.Was there any measurable difference? In the single channel study, was there any difference in the mean closed time, and what was the outcome of the dwell time (open and closed) analysis mentioned in the methods? Discussion In paragraph 2, more details are needed to explain why there may have been differences between the present findings and those previously reported for mutations found to have an impact on NMDA subunit properties.Indeed, with the breadth of type of mutations identified for different brain diseases and their mode of inheritance (loss of function, gain of function, het state) would you expect the same in terms of observable change of function? The authors state that this variant has been identified in two individuals out of 60,706 in the ExAC database and hence the authors suggest this provides additional evidence that this variant is not pathogenic and the occurrence in an individual with epilepsy was coincidental.However, the ExAC consortium includes genome data from individuals belonging to disease cohort studies as well as healthy consortium includes genome data from individuals belonging to disease cohort studies as well as healthy 'control' individuals.All that is currently published about these two individuals who have this variant is that they do not have severe paediatric disease.It may be that these individuals belong to disease cohorts and have a diagnosis of a brain disease including neurodevelopmental disorders. It is quite possible that the genotype-phenotype relationship is not as straight forward as the original hypothesis, i.e. that this variant in isolation has a functional effect and that the effect on subunit properties is directly through an amino acid change.For example, several coding variants are thought to have an effect in tandem with other coding mutations (compound heterozygous mutations).Alternatively, other genomic/epigenetic mechanisms may be disrupted by this variant and underlie transcription and translation processes which affect subunit abundance. We have read this submission.We believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.Explanation of why this particular mutation was selected over the many choices in key regions of the GluN2A subunit would be useful. As the authors allude to, this not a comprehensive study of the short and long term functional property changes which may be linked to the mutation. No competing interests were disclosed. Competing Interests: I have read this submission.I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Figure 2 . Figure 2. NMDARs containing GluN2A R586K show normal block by Mg 2+ and normal current density when expressed in neurons.(A and B) Representative whole-cell voltage-clamp recordings from day in vitro (DIV) 9 neurons transfected with an inert control, GluN2A WT or GluN2A R586K , showing inhibition of current evoked by saturating NMDA (150 μM) by Mg 2+ (1 mM) before and after one minute of ifenprodil (3 μM) application.Holding potential -65 mV.Responses are normalised to the peak amplitude.(C) Summary data showing percentage inhibition by ifenprodil, recorded from neurons as shown in A and B.A one-way ANOVA showed a significant effect of transfected subunit (F(2, 30) = 91, p = 1.8e-13), with post hoc t-tests (p corr = 0.15) showing that neurons transfected with GluN2A WT showed lower ifenprodil sensitivity than control transfection cells (t(15.7)= 12.6, p = 1.3e-9), as did neurons transfected with GluN2A R586K (t(18.3)= 13.8, p = 3.9e-11), and with no difference between GluN2A WT and GluN2A R586K (t(20.3)= 0.2, p = 0.88).These results confirm that the GluN2A R586K subunits were successfully trafficked to the membrane in neurons.(D) Summary data showing percentage inhibition by Mg 2+ in the presence and absence of ifenprodil, recorded from neurons as shown in A and B. A two-way ANOVA showed a significant effect of transfected subunit (F(2, 30) = 7.5, p = 0.0022) on Mg 2+ block and of the presence/absence of ifenprodil (F(1, 30) = 17.1, p = 0.0003), with a significant interaction (F(2, 30) = 4.4, p = 0.021).However, post hoc t-tests (p corr = 0.15) showed no difference in Mg 2+ block in the absence of ifenprodil in neurons transfected with GluN2A R586K compared to GluN2A WT (t(20.3)= 0.86, p = 0.40), and no difference between GluN2A WT and neurons transfected with an inert control (t(13.8)= 2.2, p = 0.049).There was also no reduction in Mg 2+ block associated with GluN2A R586K vs GluN2A WT in the presence of ifenprodil, when a greater proportion of response was attributable to the transfected subunits of interest (t(20.2) = 0.87, p = 0.40).(E) Summary data showing current density evoked by NMDA in the presence and absence of ifenprodil, recorded from neurons as shown in A and B. A two-way ANOVA showed a significant effect of transfected subunit (F(2, 30) = 7.5, p = 0.0023) and of the presence/absence of ifenprodil (F(1, 30) = 139, p = 9e-13), with a significant interaction (F(2, 30) = 16.8, p = 1.3e-5).However, post hoc t-tests (p corr = 0.15) showed no difference in current density in the absence of ifenprodil in neurons transfected with GluN2A R586K compared to GluN2A WT (t(20.9)= 0.5, p = 0.61), and no difference between GluN2A WT and neurons transfected with an inert control (t(18.8)= 1.2, p = 0.24).There was also no reduction in current density associated with GluN2A R586K vs GluN2A WT in the presence of ifenprodil (t(21) = 0.44, p = 0.66).Data are means ± SEM.Circles represent individual cells.See Table1for means and sample sizes. National Institute of Health and Medical Research), Paris, France Are all the source data underlying the results available to ensure full reproducibility?Yes Are the conclusions drawn adequately supported by the results?Are the conclusions drawn adequately supported by the results?Partly No competing interests were disclosed.Competing Interests: Referee Expertise: Structure-function, pharmacology, neuroscience I have read this submission.I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.School of Life Sciences, University of Nottingham, Nottingham, UK University of Nottingham, Nottingham, UK Recent large scale exome sequencing studies of disease and biomedically relevant quantitative traits have revealed a high frequency of rare coding loss of function mutations and amino acid changing missense variants across the genome . subunit genes in individuals with neurodevelopmental disorders as well as in unaffected individuals. School of Biological and Biomedical Science, Durham University, Durham, UK This is a well written, straight-forward study which has been performed well. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Partly Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate Koide T, Wang C, Kimura H, Xing J, Kushima I, Ishizuka K, Mori D, Sekiguchi M, Ikeda M, Aizawa M, Tsurumaru N, Iwayama Y, Yoshimi A, Arioka Y, Yoshida M, Noma H, Oya-Ito T, Nakamura Y, Kunimoto S, Aleksic B, Uno Y, Okada T, Ujike H, Egawa J, Kuwabara H, Someya T, Yoshikawa T, Iwata N, Ozaki N: Mutation screening of GRIN2B in schizophrenia and autism spectrum disorder in a Japanese population.	2022-07-14T19:15:11.673Z	2017-03-17T00:00:00.000	{ "year": 2017, "sha1": "5d39fe7cf8f108f57167a5ec769164ec7db30973", "oa_license": "CCBY", "oa_url": "https://wellcomeopenresearch.org/articles/2-20/v2/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "5942299981b51bbe00bc1479cecb3f33b5ed4b2f", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
18314264	pes2o/s2orc	v3-fos-license	Meta-analyses of Blood Homocysteine Levels for Gender and Genetic Association Studies of the MTHFR C677T Polymorphism in Schizophrenia Previous studies suggest that elevated blood homocysteine levels and the methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism are risk factors for schizophrenia. However, the effects of gender and MTHFR C677T genotypes on blood homocysteine levels in schizophrenia have not been consistent. We first investigated whether plasma total homocysteine levels were higher in patients with schizophrenia than in controls with stratification by gender and by the MTHFR C677T genotypes in a large cohort (N = 1379). Second, we conducted a meta-analysis of association studies between blood homocysteine levels and schizophrenia separately by gender (N = 4714). Third, we performed a case-control association study between the MTHFR C677T polymorphism and schizophrenia (N = 4998) and conducted a meta-analysis of genetic association studies based on Japanese subjects (N = 10 378). Finally, we assessed the effect of plasma total homocysteine levels on schizophrenia by a mendelian randomization approach. The ANCOVA after adjustment for age demonstrated a significant effect of diagnosis on the plasma total homocysteine levels in all strata, and the subsequent meta-analysis for gender demonstrated elevated blood homocysteine levels in both male and female patients with schizophrenia although antipsychotic medication might influence the outcome. The meta-analysis of the Japanese genetic association studies demonstrated a significant association between the MTHFR C677T polymorphism and schizophrenia. The mendelian randomization analysis in the Japanese populations yielded an OR of 1.15 for schizophrenia per 1-SD increase in plasma total homocysteine. Our study suggests that increased plasma total homocysteine levels may be associated with an increased risk of schizophrenia. Previous studies suggest that elevated blood homocysteine levels and the methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism are risk factors for schizophrenia. However, the effects of gender and MTHFR C677T genotypes on blood homocysteine levels in schizophrenia have not been consistent. We first investigated whether plasma total homocysteine levels were higher in patients with schizophrenia than in controls with stratification by gender and by the MTHFR C677T genotypes in a large cohort (N = 1379). Second, we conducted a meta-analysis of association studies between blood homocysteine levels and schizophrenia separately by gender (N = 4714). Third, we performed a case-control association study between the MTHFR C677T polymorphism and schizophrenia (N = 4998) and conducted a meta-analysis of genetic association studies based on Japanese subjects (N = 10 378). Finally, we assessed the effect of plasma total homocysteine levels on schizophrenia by a mendelian randomization approach. The ANCOVA after adjustment for age demonstrated a significant effect of diagnosis on the plasma total homocysteine levels in all strata, and the subsequent meta-analysis for gender demonstrated elevated blood homocysteine levels in both male and female patients with schizophrenia although antipsychotic medication might influence the outcome. The meta-analysis of the Japanese genetic association studies demonstrated a significant association between the MTHFR C677T polymorphism and schizophrenia. The mendelian randomization analysis in the Japanese populations yielded an OR of 1.15 Introduction Schizophrenia is a devastating psychiatric disorder with a median lifetime prevalence rate of 0.7%-0.8%. 1 Accumulating evidence has shown that alterations in 1-carbon metabolism might play an important role in the pathogenesis of schizophrenia. 2,3 A number of studies have been conducted to evaluate the association between blood homocysteine levels and schizophrenia. The majority of these studies have demonstrated elevated blood homocysteine levels in patients with schizophrenia compared with controls. However, several studies have reported no significant diagnostic differences in the blood homocysteine levels between the 2 groups. [25][26][27][28][29][30][31] To date, 1 study has examined an association between blood homocysteine levels and schizophrenia by conducting a meta-analysis of 8 case-control studies (a total number of 812 cases with schizophrenia and 2113 control subjects) and demonstrated that a 5 μmol/l increase in homocysteine concentration was associated with a higher risk of schizophrenia (OR = 1.7; 95% CI = 1.27-2.29). 32 However, this meta-analysis was performed without consideration of the effect of gender. Higher blood homocysteine levels in men than in women have been reported, 33 and the results of the previous association studies between blood homocysteine levels and schizophrenia with stratification by gender are inconclusive. In some studies, elevated blood homocysteine levels were observed in only male patients with schizophrenia and not in female patients, 5,6,10,19 whereas other studies have demonstrated that both male and female patients with schizophrenia had increased blood homocysteine levels. 11,17 Blood homocysteine levels are also influenced by genetic variations. 34,35 Among these variants, 1 common functional single nucleotide polymorphism (SNP) of the methylenetetrahydrofolate reductase (MTHFR) gene, C677T (rs1801133), has been investigated well. The MTHFR C677T polymorphism results in amino acid substitution (Ala222Val) and causes a reduction of enzyme activity and higher homocysteine levels. 36 The results of the previous association studies between blood homocysteine levels and schizophrenia with stratification by C677T genotypes are inconclusive. A significant diagnostic difference in blood homocysteine levels has been found only in the subjects carrying the CT genotype or only in the subjects carrying the TT genotype. 7,18 However, Feng et al 16 showed a significant diagnostic difference for both CT and TT genotype carriers. Many genetic case-control association studies between the MTHFR C677T polymorphism and schizophrenia have been performed in various populations, and the results of these association studies are not consistent. Only 1 study of the Japanese population reported a significant association between the MTHFR C677T polymorphism and schizophrenia, while the other 3 studies of the Japanese population have not replicated this positive finding. [37][38][39][40] However, several meta-analyses of association studies have revealed a significant association between this SNP and schizophrenia. 39,[41][42][43][44][45][46][47] In this study, we first investigated whether plasma total homocysteine levels were higher in patients with schizophrenia than in nonpsychiatric controls with stratification by gender and by the MTHFR C677T genotypes in a large cohort (N = 1379). Second, we conducted a meta-analysis of association studies between the blood homocysteine levels and schizophrenia separately by gender to evaluate a precise estimation of the association (N = 4714). Third, we performed a case-control association study between the MTHFR C677T polymorphism and schizophrenia (N = 4998) and carried out a metaanalysis of genetic association studies of this SNP with schizophrenia based on Japanese subjects to determine whether the MTHFR C677T polymorphism was genetically implicated in schizophrenia in the Japanese population (N = 10 378). Finally, we assessed the effect of plasma total homocysteine levels on schizophrenia by a mendelian randomization (MR) approach, a useful tool for assessing causal associations in observational data. 48,49 Methods Subjects of the Association Study Between the Plasma Total Homocysteine Levels and Schizophrenia Three hundred and eighty-one patients with schizophrenia (225 men, mean age: 58.2 ± 9.3 y; 156 women, mean age: 59.4 ± 9.7 y) were recruited from Tokushima University Hospital in Japan. The diagnosis of schizophrenia was made according to Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria by at least 2 expert psychiatrists on the basis of extensive clinical interviews and a review of medical records. None of the patients had any psychiatric comorbidity or cardiovascular diseases. All patients were treated with various antipsychotic drugs. The mean chlorpromazine equivalent dose was 689.6 ± 581.3 mg/d. Nine hundred and ninety-eight control subjects (331 men, mean age: 38.3 ± 12 y; 667 women, mean age: 43.0 ± 11.9 y) were selected from volunteers who were recruited from hospital staff, students, and company employees documented to be free from psychiatric problems and past histories of mental illness. All subjects who participated in this study were of Japanese origin. All subjects signed written informed consent approved by the institutional ethics committees of the University of Tokushima Graduate School. Of 1379 subjects used in this association study, 1357 with genomic DNA (377 patients and 980 controls) were used in the next genetic association study. Subjects of the Association Study Between the MTHFR C677T Polymorphism and Schizophrenia Two case-control sets were used: the Tokushima sample set (A southern island of Japan) and the Osaka sample set (Midwestern Japan). Both sets have been described in previous studies. 50,51 For the Tokushima sample set, 1149 patients with schizophrenia (676 males, 473 females, mean age: 54.6 ± 14.9 y) were recruited from the Tokushima and Kochi University Hospitals in Japan. The diagnosis of schizophrenia was made according to DSM-IV criteria. A total of 2742 control subjects (1230 males, 1512 females, mean age: 38.8 ± 12.6 y) were selected from volunteers. For the Osaka sample set, 621 patients with schizophrenia (302 males, 319 females, mean age: 46.5 ± 15.8 y) were recruited from Osaka University Hospital in Japan. The diagnosis of schizophrenia was made according to DSM-IV criteria. A total of 486 control subjects (231 males, 255 females, mean age: 35.0 ± 12.7 y) were selected from volunteers. All subjects signed written informed consent approved by the institutional ethics committees of the University of Tokushima Graduate School, Kochi Medical School, and University of Osaka Graduate School. Plasma Total Homocysteine Analysis Plasma total homocysteine levels were measured by high-performance liquid chromatography. Homocysteine was labeled with 4-fluoro-7-sulfamoylbenzofurazan and detected by a fluorescent detector according to the method of a previous study. 52 MTHFR Genotyping We genotyped the MTHFR C677T polymorphism by using a commercially available TaqMan probe with the Applied Biosystems 7500 Fast Real Time PCR System, according to the protocol recommended by the manufacturer (Applied Biosystems, California, USA). Twelve percent of the genotypes were genotyped again, and there were no mismatches between the 2 genotyping steps. Study Selection for the Meta-analysis of Association Studies Between Blood Homocysteine Levels and Schizophrenia Eligible studies were identified using the PubMed search engine with the terms "homocysteine," "hyperhomocysteinemia," and "schizophrenia." We also conducted an additional manual search of reference lists and review articles. Studies meeting the following criteria were included for further meta-analysis: (1) included laboratory assessment of serum or plasma homocysteine levels, (2) performed a case-control study (schizophrenia vs control), (3) provided raw data of homocysteine levels separately by gender, and (4) published in an English language. The 2 reviewers (N.S. and K.K.) selected the articles independently according to the inclusion criteria, and then discussed the articles until they reached a consensus on every study used for the meta-analysis. Statistical Methods A linear regression analysis was used to examine the effects of diagnosis, age, gender, and the MTHFR C677T genotypes on the plasma total homocysteine. An ANCOVA was performed to examine the presence of the differences in the plasma total homocysteine between the 2 groups (schizophrenia vs control) separately by gender and by the 3 MTHFR C677T genotypes (total of 6 strata). Allelic and genotypic frequencies of the MTHFR C677T polymorphism in patients and control subjects were compared using the χ 2 test. In order to quantify the strength of association between plasma total homocysteine and schizophrenia, an MR approach was used, as in a previous study. 53 The risk estimate in gene-schizophrenia association for the TT genotypes of the MTHFR C677T polymorphism (vs the CC genotypes; OR SCZ/TT ) was from the current meta-analysis of the Japanese genetic association studies. For gene-homocysteine association, the effect of the TT genotypes (vs the CC genotypes) on plasma total homocysteine levels (beta hcy/TT ) was estimated using the Japanese control subjects from the Tokushima homocysteine study under a multivariate linear regression model including age and gender as covariates. The effect for the TT genotypes in the gene-homocysteine association was expressed as 1-SD increase in plasma total homocysteine. From these 2 estimates, an MR estimate of the effect of plasma total homocysteine on the risk of schizophrenia (OR SCZ/hcy ) was calculated as follows: log OR SCZ/hcy = (log OR SCZ/ TT )/beta hcy/TT . The MR estimate represented the OR for schizophrenia risk per 1-SD increase in plasma total homocysteine. The standard error (SE) of the MR estimate was calculated by the Delta method. 54,55 Meta-analysis The meta-analysis of association studies between the blood homocysteine levels and schizophrenia was performed on the standardized mean differences (SMD). The meta-analysis of association studies between the MTHFR C677T polymorphism and schizophrenia was performed for 5 genetic models, recessive (CC/CT vs TT genotypes), dominant (CC vs CT/TT genotypes), codominant (CC vs TT genotypes), codominant (CT vs TT genotypes), and allele frequency (C-allele vs T-allele), as had been done in a previous study. 45 Heterogeneity was assessed using the I 2 statistic. If heterogeneity across studies was found, then a random-effects model was applied; otherwise, a fixedeffects model was applied. Publication bias was assessed using funnel plots and a regression test. 56 OR and 95% CI were calculated by "metafor," an R package. Differences in the Plasma Total Homocysteine Levels Between Patients With Schizophrenia and Controls A linear regression analysis showed significant effects of diagnosis (higher in schizophrenic patients than in controls), MTHFR C677T genotypes (higher in CT carriers than in CC carriers, and higher in TT carriers than in CC carriers), age (decreases with age), and gender (higher in males than in females) on the plasma total homocysteine levels (diagnosis P = 3.4 × 10 -29 , genotype P [CT vs CC] = 4.7 × 10 -3 , genotype P [TT vs CC] = 5.8 × 10 -43 , age P = 1.0 × 10 -2 , and gender P = 1.3 × 10 -26 ). Next, an ANCOVA was performed to examine the presence of the differences between the 2 groups in the plasma total homocysteine separately by gender and by the 3 MTHFR C677T genotypes (total of 6 strata), and a significant effect of diagnosis (higher in patients with schizophrenia than in the control) was still observed in all strata (diagnosis P of malegenotype CC = 2.4 × 10 -8 , diagnosis P of male-genotype CT = 3.2 × 10 −10 , diagnosis P of male-genotype TT = 2.3 × 10 -4 , diagnosis P of female-genotype CC = 1.1 × 10 -8 , diagnosis P of female-genotype CT = 3.2 × 10 -8 , and diagnosis P of female-genotype TT = 1.2 × 10 -5 , respectively) after adjustment for age ( figure 1). A Meta-analysis of the Blood Homocysteine Levels in Schizophrenia We performed a meta-analysis of previous association studies between the blood homocysteine levels and schizophrenia separately by gender. The studies included in this meta-analysis are shown in the supplementary table 1. For the meta-analysis of males, data were obtained from 12 association studies, 5,6,8,10,11,13,17,20,21,25,28 including our data, for a total of 1079 patients with schizophrenia and 1559 control subjects. As shown in figure 2A, the randomeffects model showed that the blood homocysteine levels were significantly higher in male patients with schizophrenia than in the male controls (SMD = 0.76; 95% CI = 0.30-1.22; P = 1.2 × 10 -3 ) with significant heterogeneity among studies (I 2 = 96.3%; P < .05). The funnel plot analysis indicated no evidence of publication bias in the male association studies (P = .13; supplementary figure 1). For the meta-analysis of females, data were obtained from 10 association studies, 5,6,8,10,11,13,17,25,28 including our data, for a total of 615 patients with schizophrenia and 1461 control subjects. As shown in figure 2B, the randomeffects model showed that the blood homocysteine levels were significantly higher in female patients with schizophrenia than in the female controls (SMD = 0.50; 95% CI = 0.31-0.70; P = 5.9 × 10 -7 ) with significant heterogeneity among the studies (I 2 = 65.7%; P < .05). The funnel plot analysis indicated no evidence of publication bias in the female association studies (P = .73; supplementary figure 2). A Case-Control Association Study Between the MTHFR C677T Polymorphism and Schizophrenia Two case-control data sets were evaluated: one is the Tokushima sample set (case = 1149, control = 2742), and the other is the Osaka sample set (case = 621, control = 486). The genotypic distributions of rs1801133 did not deviate significantly from the Hardy-Weinberg equilibrium (HWE) in the control groups of these 2 sample sets (P > .05). Significant difference was observed between the controls and patients with schizophrenia in the allelic frequencies of the Tokushima sample set (P = .025). On the other hand, no significant differences were observed in the genotypic and allelic frequencies of the Osaka sample set (genotype P = .98, and allele P = .97, respectively). A Meta-analysis of Genetic Association Studies Between the MTHFR C677T Polymorphism and Schizophrenia Six association studies on Japanese subjects, including the 2 data sets from this study, were used for the meta-analysis. [37][38][39][40] The genotypic distributions and allelic frequencies of rs1801133 in each study are shown in the supplementary table 2. The diagnosis of schizophrenia was made according to DSM-IV criteria in 5 studies, and it was made according to DSM-III criteria in the remaining study. A total of 4316 cases and 6062 controls were included in this analysis. Genotypic distribution of this SNP did not deviate significantly from the HWE in any control group across the 6 studies (P > .05). Significant heterogeneity was not detected at this SNP among the studies for the 5 genetic models (P > .05). The funnel plot analysis indicated no evidence of publication bias for all genetic models (P > .05). The results of ORs and CI analyzed by the fixed-effects model for all genetic models and the risk of schizophrenia are shown in table 1. Of these 5 genetic models, significant associations were found in 4 models. The highest OR was observed in the codominant model (CC vs TT genotypes; figure 3; OR = 1.16, 95% CI = 1.03-1.31, P = 1.4 × 10 -2 , in the fixed-effects model). Effect of Plasma Total Homocysteine Levels on Schizophrenia Risk in an MR Study From the current meta-analysis of the Japanese genetic association studies, the pooled OR (the TT vs CC genotypes) for the effect of the MTHFR C677T polymorphism on schizophrenia risk was 1.16 (95% CI = 1.03-1.31). In multivariate gene-homocysteine association analysis in the Fig. 1. Differences in the plasma total homocysteine levels between patients with schizophrenia and controls separately by gender and by the methylenetetrahydrofolate reductase (MTHFR) C677T genotypes. The ANCOVA demonstrated that the plasma total homocysteine levels were significantly higher in patients with schizophrenia than in controls in each of the 6 strata (*age-adjusted P < .001). Japanese control subjects genotyped (n = 980 with a SD for plasma total homocysteine levels of 4.84 nmol/ml), the effect on plasma total homocysteine levels, expressed as 1-SD increase in homocysteine, was estimated to be 1.14 (95% CI = 0.96-1.33; P = 1.1 × 10 -29 ) for the TT genotypes of the MTHFR C677T polymorphism (vs the CC genotypes), after adjustment for age and gender. When combining these 2 estimates by an MR approach, the effect of plasma total homocysteine on schizophrenia risk was statistically significant, representing the OR of 1.14 (95% CI = 1.03-1.27; P = 1.6 × 10 -2 ) for schizophrenia per 1-SD increase in plasma total homocysteine ( figure 4). Discussion In this study, we performed an association study between the plasma total homocysteine and schizophrenia with stratification by gender and by the MTHFR C677T genotypes and demonstrated significantly elevated plasma total homocysteine levels in patients with schizophrenia compared with controls, in both male and female subjects. The subsequent meta-analysis for gender supported this finding. To our knowledge, this is the first study to conduct a meta-analysis according to gender. The significant decrease in plasma total homocysteine levels with age in our multivariate linear regression . 3. A meta-analysis of genetic association studies on Japanese subjects between the MTHFR C677T polymorphism and schizophrenia. Six association studies on Japanese subjects, including the 2 data sets from this study, were used for the meta-analysis (N = 10 378). Significant associations between the MTHFR C677T polymorphism and schizophrenia were found in 4 models, and the result of the codominant model (CC vs TT genotypes) is shown (OR = 1.16, 95% CI = 1.03-1.31, P = 1.4 × 10 -2 in the fixed-effects model). analysis was not consistent with a previous finding. 33 This discrepancy might be caused by disease status and gender of the subjects analyzed. When we examined the relationship between age and the plasma total homocysteine levels in a univariate linear regression model, a significant negative correlation was observed only in male subjects with schizophrenia (supplementary figure 3). Consistent with this finding, Levine et al 5 reported that the difference in the plasma total homocysteine levels between patients and controls was attributable to young male patients with schizophrenia. We also conducted a meta-analysis of genetic association studies between the MTHFR C677T polymorphism and schizophrenia based on Japanese subjects and demonstrated that the MTHFR C677T polymorphism was a risk factor for developing schizophrenia in the Japanese population, which is consistent with the results of previous meta-analyses. 39,[41][42][43][44][45][46][47]57 On the other hand, this polymorphism has not been identified as a risk locus for schizophrenia in the large genome-wide association studies. 58,59 This discrepancy might be caused by ethnic differences (risk allele [T] frequencies from HapMap; Japanese 0.39, Caucasian 0.31, African American 0.12, and Mexican American 0.41) and a lack of adequate statistical power to detect the relatively small genetic effect of this polymorphism on schizophrenia at the genomewide significant threshold. Importantly, we demonstrated that increased homocysteine levels may be associated with an increased risk of developing schizophrenia by an MR approach. Hyperhomocysteinemia has been proposed as being part of the pathophysiology of schizophrenia due to its various biological effects, such as acting as a partial antagonist of the glutamate site of the N-methyl-d-aspartate receptor 23 and causing subtle placental vascular damage that interferes with oxygen delivery to the fetus, 60 DNA damage and cell cytotoxicity, 61 neuronal apoptosis, 62 and mitochondrial nitric oxide accumulation. 63 Homocysteine acts as a methyl donor when it is converted to S-adenosyl-methionine, and we recently demonstrated a significant association between the plasma total homocysteine and DNA methylation in schizophrenia, which suggests that homocysteine might play a role in the pathogenesis of schizophrenia via alterations to DNA methylation. 64 Homocysteine, S-adenosylmethionine, DNA methylation, MTHFR, folate, and vitamin B12 are involved in 1-carbon metabolism, and abnormalities of these components in schizophrenia have been reported in previous studies. 2,3,65 These lines of evidence suggest that disrupted 1-carbon metabolism may be an important role in the pathophysiology of schizophrenia. The benefits of homocysteine-reducing strategies in schizophrenia have been shown in several studies. Levine et al 66 reported an improvement in the clinical symptoms of schizophrenic patients with hyperhomocysteinemia who were treated with folate, vitamin B12, and pyridoxine. Hill et al 67 reported an improvement in the negative symptoms of schizophrenic patients who were treated with folate when the MTHFR C677T genotype was taken into account. Roffman et al 68 reported an improvement in the negative symptoms of schizophrenic patients who were treated with folate and vitamin B12 when 4 variants in the FOLH1, MTHFR, MTR, and COMT genes were taken into account. Further research will be necessary to identify the features of patients with schizophrenia who would benefit from homocysteine-reducing treatments. In cardiovascular diseases, which are also associated with hyperhomocysteinemia, clinical trials to identify a subgroup that appeared to benefit from homocysteine-lowering intervention have been performed 69,70 although no benefits of homocysteine-lowering intervention on cardiovascular outcomes have been reported in randomized controlled trials. 71 Many studies have indicated the potential contributions of the MTHFR C677T polymorphism to the pathophysiology of schizophrenia. This risk SNP has been associated with schizophrenic negative symptoms, aggressive behavior, and various phenotypes related to schizophrenia, such as cognitive function, episodic memory, gray matter density, and prefrontal function. [72][73][74][75][76][77][78][79][80] Interestingly, pharmacogenetic studies have demonstrated that this risk SNP has also been involved in the antipsychotic drug response and metabolic syndrome treated with antipsychotics in schizophrenia. [81][82][83][84] There are some limitations to this study. First, we did not obtain genomic DNA from all participants in the association study between the plasma total homocyateine levels and schizophrenia. Second, all patients were treated with various antipsychotic drugs, and these medications might Fig. 4. Graphical representation of the mendelian randomization approach. The pooled OR (the TT vs CC genotypes) for the effect of the MTHFR C677T polymorphism on schizophrenia risk was 1.16 (95% CI = 1.03-1.31). The OR (the TT vs CC genotypes) for the effect of the MTHFR C677T polymorphism on plasma total homocysteine levels, expressed as 1-SD increase in homocysteine, was 1.14 (95% CI = 0.96-1.33; P = 1.1 × 10 -29 ). The effect of plasma total homocysteine on schizophrenia risk by a mendelian randomization analysis was statistically significant, representing the OR of 1.14 (95% CI = 1.03-1.27; P = 1.6 × 10 -2 ) for schizophrenia per 1-SD increase in plasma total homocysteine. influence the outcome. When we examined the relationship between equivalent dose and the plasma total homocysteine levels in subjects with schizophrenia by a univariate linear regression model, a positive correlation was observed in female subjects (P = .03). However, it did not reach statistical significance after correction for multiple comparisons. Third, there was heterogeneity among the studies in the meta-analysis for blood homocysteine, while significant heterogeneity was not observed in the meta-analysis for genetic association studies. This heterogeneity might be caused by other genetic variations, the clinical heterogeneity of the patients included, medications, and environmental factors, such as folic acid, vitamin B6, vitamin B12, obesity, smoking status, and caffeine consumption, although we did not take these confounding factors into consideration in our analysis. Fourth, the use of "well controls" in this case-control analysis might accentuate such confounding influences. 85 Fifth, this is a cross-sectional study, and MR has limitations on the ability to establish causal relationships between risk factors and outcomes. 86 So, the causality between schizophrenia and blood homocysteine levels must be still cautious. Notably, elevated maternal levels of homocysteine during the third trimester have been found to increase the risk of schizophrenia in the offspring. 60 Finally, hyperhomocysteinemia has been identified as an independent risk factor for several neurological disorders in addition to schizophrenia, such as depression and dementia. 87,88 Further studies to examine how hyperhomocysteinemia is involved in the pathophysiology of each disease will be necessary. In conclusion, to the best of our knowledge, this is the first meta-analysis of association studies between blood homocysteine levels and schizophrenia according to gender, and we demonstrated elevated blood homocysteine levels in both male and female subjects with schizophrenia. The meta-analysis of genetic association studies using the Japanese subjects provided stringent evidence of association between the MTHFR C677T polymorphism and schizophrenia. Our MR analysis using the Japanese subjects suggests that increased plasma total homocysteine levels may be associated with an increased risk of developing schizophrenia. Funding Core Research for Evolutional Science and Technology, Japan Science and Technology Agency; Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology (24791216); SENSHIN Medical Research Foundation; Research Group For Schizophrenia.	2017-04-14T04:57:23.325Z	2014-02-17T00:00:00.000	{ "year": 2014, "sha1": "608f19e2b5d073976fa1e44968bba8f7c0cc7141", "oa_license": "CCBYNCND", "oa_url": "https://academic.oup.com/schizophreniabulletin/article-pdf/40/5/1154/16976818/sbt154.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "5efadd02ae4eb8ae3bc6800a526b80332eda8dc6", "s2fieldsofstudy": [ "Psychology", "Medicine" ], "extfieldsofstudy": [ "Medicine", "Psychology" ] }

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.