| "text": "This is an academic paper. This paper has corpus identifier PMC2532782\nAUTHORS: Nayeem Quayum, Alecksandr Kutchma, Suparna A. Sarkar, Kirstine Juhl, Gerard Gradwohl, Georg Mellitzer, John C. Hutton, Jan Jensen\n\nABSTRACT:\nObjective. We here describe the development of a freely available online database resource, GeneSpeed Beta Cell, which has been created for the pancreatic islet and pancreatic developmental biology investigator community. Research Design and Methods. We have developed GeneSpeed Beta Cell as a separate component of the GeneSpeed database, providing a genomics-type data repository of pancreas and islet-relevant datasets interlinked with the domain-oriented GeneSpeed database. Results. GeneSpeed Beta Cell allows the query of multiple published and unpublished select genomics datasets in a simultaneous fashion (multiexperiment viewing) and is capable of defining intersection results from precomputed analysis of such datasets (multidimensional querying). Combined with the protein-domain categorization/assembly toolbox provided by the GeneSpeed database, the user is able to define spatial expression constraints of select gene lists in a relatively rigid fashion within the pancreatic expression space. We provide several demonstration case studies of relevance to islet cell biology and development of the pancreas that provide novel insight into islet biology. Conclusions. The combination of an exhaustive domain-based compilation of the transcriptome with gene array data of interest to the islet biologist affords novel methods for multidimensional querying between individual datasets in a rapid fashion, presently not available elsewhere.\n\nBODY:\n1. INTRODUCTIONGenomics is playing a growing role\nin almost any biological experimentation. Based on presently available\ncommercial expression array technologies, an investigator is given almost\nfull-genome coverage of transcriptional changes that provides for novel methods\nfor gene identification and validation. However, exhaustive data mining from\ngenomics datasets is cumbersome, and to a large extent is outside the expertise\nof the individual experimenter. The greatest strength in genomics data analysis\nstems from multidimensional analysis, as such orthogonal comparison can bring\nout biologically relevant information not extractable from the individual\ndatasets alone. However, such multidimensional querying is often advised\nagainst, as individual genomics experiments are performed in different\nlaboratories, using dissimilar methodologies. Such array data should not be\nuploaded and analyzed concomitantly in the available software data analysis\nprograms commonly used. Prudent analysis of multiexperimental results would\ntherefore call for individual data analysis of experimental sets, and only\nparse for intersections/exclusions within the resulting gene lists. This is\npossible through genomics analysis platforms using separate gene list saving.\nHowever, the process is burdened by the fact that the analysis of a relevant\ndataset for orthogonal querying requires the identification of the existence of\nthe data, upload, normalization, and scaling of individual DNA chip scan files\nand thereafter selecting and executing a valid analysis for the particular\ndataset, followed by results storage. In practice, this is time consuming, and\ntoo overwhelming, for most biologists.In the islet and islet\ndevelopmental biology research fields, a continuously growing set of public\ngenomics data is becoming available. Also, initial problems in both genomics\nchip design and experimental execution are gradually being overcome. It is\ntherefore appreciated that a large, generally untapped resource is provided by\ngenomics analyses performed in islet-focused laboratories around the world. Two\nonline databases, T1dbase [1] and EpconDb [2], contain genome data\nrepository components within their sites. However, they do not provide advanced\nmultiexperimental querying options that would allow generation of gene lists\nbetween experiments. Acknowledging this, we set forth to create a resource that\nwould consolidate diabetes-research relevant genomics data and allow rapid\nmultidimensional analysis between such datasets. To do this, we created an\nonline genomics data repository, which we term GeneSpeed Beta Cell. This was\ndeveloped as an additional component of the GeneSpeed resource [3], see \nFigure 1. GeneSpeed Beta\nCell (http://genespeed.ccf.org/betaCell/) contains two forms of \ndata: normalized and similarly scaled genomics data relevant for the islet or pancreatic \ndevelopmental biologist. Secondly, it contains precalculated analyses, which include pairwise\nand self-organizing neural network clustering results applied to relevant data\nseries. On the analysis side, GeneSpeed Beta Cell provides “My Gene Workspace”\nwhere gene list overlap can be evaluated. It also provides access to any search\nparameter in the GeneSpeed environment, including precalculated data on tissue\nspecificity (Shannon entropy) and wide-tissue\nbatch expression queries. Together, the unified environment within the\nGeneSpeed database provides for some unique capacities not found elsewhere. We\nhere describe the use of GeneSpeed Beta Cell by addressing a set of novel, and biologically\nrelevant, questions appealing to the islet biologist.2. GENESPEED BETA CELL DATABASE STRUCTURE\nAND NAVIGATION2.1. Multiexperimental viewing“GeneSpeed Beta Cell” is a gene\narray data repository linked to the GeneSpeed environment. For a more detailed\ndescription of the domain-based gene categorization afforded by GeneSpeed,\nplease refer to [3] and the online background and\ntutorials. GeneSpeed Beta Cell consists of a central “experiment selector page”\n(Figure 2), listing experiments by relevance to embryonic development, adult\nislet studies, adult whole pancreas studies, experiments using cell lines, and\nsolid tumor data. Currently, the site is being expanded with gene chip data of\ndeveloping nonpancreatic endoderm. For each group, experiments are separated\ninto human and mouse studies. At current, Affymetrix-type data is supported,\ngiven that the body of relevant genomics data is the largest on this particular\nplatform; but as the parent GeneSpeed database is not platform-specific, we\nhave also made it capable of operating with the Illumina BeadChip type datasets.\nAll current Affymetrix-type datasets in the repository were obtained as\nunnormalized raw cel files, and were normalized using MAS5.0 using identical\nsettings and similarly scaled for cross-experimental comparisons (see methods).\nThe available data can be viewed through multiexperimental viewing. Any saved\ngene list can be viewed for any of the available datasets. This is a fast and\nconvenient way to display the normalized expression values of defined gene\nlists between independent experiments performed in different laboratories. The\nresulting display page is constructed to facilitate horizontal glancing of expression\nvalues, while maintaining the individuality of experiments. As there is no\nlimit as to the number of genes shown or number of experiments selected, the\nresulting page view can be quite large. To assist the identification of the\nrespective column (tissue type/experimental condition) and row (gene symbol), a\nhovering tool supplying this information was implemented. Also, for quick\nanalysis of the gene ID, each cell is hyperlinked to the respective Unigene\npage for that gene. If a gene within a gene list does not contain a respective\nprobeset, the cell content is displayed as N.A. The multiexperiment viewer\nfacilitates table sorting based on each component in selected datasets. This\nprovides, for example, quickly arranging genes in a larger gene list according\nto expression levels for any tissue/condition selected (e.g., Figure 3 shows a\nlist of homeodomain-class genes sorted based on expression at the E12.5\ngestational time point in pancreatic development).2.2. Multidimensional analysis in “My Gene Workspace”To enhance online capabilities, we\ndeveloped tools for multidimensional analysis. The multidimensional analysis\ntool operates within a “My Gene Workspace” environment (Figure 4), which is\narray-platform independent as it stores genes by Unigene identifier. “My Gene\nWorkspace” allows for temporary storage of gene lists, naming such, and\nselecting individual lists to be combined using the Boolean operators AND or\nOR. Hereby, intersections (AND), or additive combinations (OR) can be performed\non the selected gene lists, for further logical operations or visualized using\nthe multiexperimental viewer. Several means of populating the workspace is\npossible. The user is provided with a “permanent list” account, in which work\nbetween sessions can be saved. Lists can here be grouped according to project\nname. Gene lists from the permanent account can be ported to the workspace or\ngene family choices from a concurrent GeneSpeed query that can be directly imported. In addition,\ngene list results from precalculated analyses based on the available datasets\ncan be added. The final option provides a highly useful method to dynamically aggregate\nand compare results from individual experimental data that was not initially\ndesigned for a combined analysis. Such comparisons can be highly scientifically\nrelevant, and examples are provided later.As this latter method is based on\nprecalculated analysis of available datasets, a certain level of a priori\nchoice has been necessary to implement, as all permutations of possible data analysis\ncould not be practically implemented. Consequently, depending on the underlying\nexperimental conditions, the precomputed analysis is restricted to pairwise\nanalysis (although multiple pairwise comparisons are often provided for a given\ndataset), or a self-organizing cluster analysis (for series-type data such as\nexperimental time, drug concentration, or developmental time). Graphical\npresentation of each analysis is provided to help the user gauge gene numbers\ngiven the conditions chosen. For pairwise analysis, a volcano plot (plotting\nsignificance (p-value) versus fold-change) of the pairwise analysis result is\nshown. The default cutoff for gene selection is set at a false discovery rate\nof 0.1, but can be changed to the user's preference. Similarly, the fold-change\nrange can be freely set, allowing the user to port, that is, >2-fold\nupregulated genes in a given condition into the workspace. The graphical\npresentation of cluster analyses contains a cluster number, and number of genes\nwithin the cluster. The user is free to\nselect any number of clusters and port to the workspace. In this manner,\nvarious experimental conditions can be continuously ported to the workspace,\nand the experimental multi-intersectional analysis occurs there. There is no\nlimit to the number of gene lists present in the workspace. We should note that\nfor both the multidimensional viewing page, and for the multidimensional query\nform in the workspace, individual datasets are always kept separate (viewer),\ntreated as such (query page), and are never pooled. Cross-experimental pooling\nis not tolerable due to varying conditions in different laboratories during\ndata generation.2.3. Current experimental content of GeneSpeed Beta CellAs the available datasets and analyses grow on a daily basis, users should visit the site for a list of\ncurrently available datasets and analyses.2.4. GeneSpeed Beta Cell use-case scenariosSome biologically relevant use-case\nscenarios for the islet cell biologist are described in the following. Each of\nthese is available also as online tutorials at GeneSpeed Beta Cell at http://genespeed.ccf.org/betaCell/tutorial.jsp.\nAs for any bioinformatics-based method application, the results are provided as candidate gene lists, corresponding to genes/probesets fulfilling input\ncriteria. The further validation of such lists using noninformatics-based techniques is a\ngeneral requirement. In the following demonstrations, the end-result gene lists\nare often supported by previous published data from other sources, hereby\nproviding the validation required for the particular demonstration scenarios.\nExample 1(Compiling lists of islet-expressed transcription\nfactors (online tutorial 1)). We wish to address the issue of\ndefining islet-expressed transcription factor (TF) encoding genes. To do this,\nwe will utilize the predefined transcription factor categorization provided by\nthe GeneSpeed database, assemble a nonredundant list of TF encoding genes, and\nfind those reduced in Ngn3 null pancreas. First, we select “new search,”\nand desired species “mouse” from the drop-down menu. Next, we select “search by\ntranscription factor classification” within the GeneSpeed search options. As 5\nmajor domain family groupings exist for the transcription factor type genes, we\nwill need to iterate the following procedure for each, but will here limit the\nfamilies to the “Basic,” “Beta-Scaffold,” and “HTH” superfamilies. These\nfamilies contain, for example, the leucine zipper, bHLH, and homeodomain\ntranscription factor families, but not the Zn-finger class. Selecting “Basic”\nas the first type, we ctrl-select all the subfamily members of the basic TF\nsuperfamily. Displaying the result provides 685 hits. These correspond to every\ninstance where the Unigene database of the mouse contains a homology hit for\nany of the domain types associated the “basic” superfamily. However, as the\ndatabase has no preset lower E-score cutoff, several false positives exist in\nthis list (see discussion of how to set an E-score cutoff on the description\npages at GeneSpeed for a full explanation). To eliminate low-scoring similarity\nhits, we set the E-score cutoff at E10-6, and redo the search. Now, a resulting\nlist of 167 genes is detected. We save these to the user account under an\narbitrary name (All_TFs). This process is repeated for the TF superfamilies\nmentioned above, where the individual results is added to the All_TF's list,\nconsequently providing a list of >1600 individual Unigenes. These are next\nimported into the “My Gene Workspace.” To extract genes unique to pancreatic\nislets in the developing pancreas, we will take advantage of the available\ndataset for Ngn3-null embryonic pancreas, which is listed under the experiment\nlisting page of GeneSpeed Beta Cell. A pair-wise analysis is provided comparing\nE15.5 Wt and E15.5 Ngn3 Null pancreas. The Ngn3-deficient\npancreas is excellent to define endocrine-specificity, as the organ is devoid\nof endocrine cells. Selecting genes upregulated >1.5 fold, P < .25,\na second list is imported into the workspace as Down_in_Ngn3. This list\ncontains 114 genes. Obtaining the intersection between the TF_ALL and Down_in_Ngn3 lists provide a total of 8 transcription factor encoding genes lost in Ngn3\nmutant E15.5 pancreas: Ngn3, NeuroD, Isl1, Pax6, Arx, MafB, Nkx2.2, Insm1, and HIF1a.\n\nExample 2 (Multidimensional intersection analysis to\ndefine developmentally regulated expression of protein\nkinase-encoding genes (online tutorial 2)). We here will seek\nto discover kinase-encoding genes that are enriched in either early or late\npancreatic development. A similar study has not been done before. To perform\nthis task, we first need to compile a list of all protein kinase-type genes in\nthe mouse transcriptome. Using a text-search for a gene known as a protein\nkinase (e.g., insr), we obtain two hits: Insr and Insrr. Both of\nthese are receptor tyrosine kinases, and display the presence of the\nTyr_pkinase domain (IPR001245) with an E-score at 1E10−145. We also\nnote that the generic kinase domain (IPR000719) is detected in both at 1E10−24.\nBy checking the “InterPro\nsub-search” box for the IPR001245 domain, and execute the search:\n“refine by subsearch,” we obtain a nonredundant list of Unigene clusters having\nsimilarity to the IPR001245 domain. This provides 480 hits, covering all\nkinase-domain forms (S/T as well as Y-kinase types). To curate against low-similarity\nhits, we manually set the E-score threshold at <1E-6. The resulting list\ncontains bona-fide 432 kinase-containing genes, which we subsequently save as “Kinase_all”\nto our account. Many of these genes represent genes with no previous annotation\nas being of the kinase-domain containing type, and may not have been named yet.\nNext, we wish to identify which of these kinase-encoding genes display a\ndownward trend during pancreatic development. To do this, we move to the\n“search GeneSpeed Beta Cell,” and expand the “Embryonic studies” dataset tab.\nSelecting the “kinetic series of mouse pancreatic development 1” precomputed\ncluster analysis, we are provided with the results of a Kohonen's\nself-organizing cluster analysis in a graphical format. Gene clusters with a\ndownward trend during pancreatic development are selected (cluster\n3,4,5,8,9,15,20) and combined using the selection tool provided. The results\nare saved as Genes_Trend_Down to the workspace. Within the workspace the\nintersection between the Genes_Trend_Down and the Kinase_all lists are obtained using the Boolean operator AND. The resulting list contains\n138 kinase-type genes. The list can be saved, or gene expression of the\nparticular genes can be displayed in some or all mouse array experiments in the\nGeneSpeed Beta Cell database. The latter may provide important clues as to\ntissue-specific expression of individual members. Finalizing this demonstration,\nwe wish to address the identity of kinase-encoding genes that are upregulated\nover time in the developing pancreas. By repeating the above method for\nkinase-type genes displaying upregulation (Cluster 6,11,16,17,21,22, generating\nlist Gene_Trend_Up), only 27 genes are identified. We can conclude that\nmore kinase signaling diversity exists prior to rather than after the secondary\ntransition in the mouse pancreas.\n\nExample 3(Defining human islet-specific expression using\nShannon Entropy with exocrine elimination (online tutorial\n3)). This example uses the available\ndataset on human tissues, as provided by the Novartis Genomics Institute (http://symatlas.gnf.org/SymAtlas/about.jsp).\nA tissue set consisting of 79 human tissues and 61 different mouse tissues, mostly\nadult solid organs, has been generated in duplicate using the Affymetrix GNF1\nplatform. To provide a measure of tissue expression selectivity, we adopted the\nmethod of Shannon entropy determination, as previously described by Schug et\nal. [4]. Shannon entropy provides quantitative measures of expression using a bit-rate scale.\nFor each gene, the Shannon entropy (H\ngene)\ndefines the degree of ordered expression; as a rule, the lower the H\ngene,\nthe fewer tissues in the total set express the gene in question. To identify\nthose tissues showing uniqueness in expression, the measure Q\ntissue can be used. Again, as a rule, the lower the Q\ntissue value, the more\nspecific the gene is expressed in that particular tissue. A rank order of the\nlowest Q\ntissue values thus provides a list of those genes that have\nthe highest selectivity for the tissue in question. Shannon entropy computations were performed for all tissues in the above datasets. As\nthe human, but not the mouse, datasets contain array data for islets, the present\nexample is currently limited to human. From the GeneSpeed search query, we\nselect species “homo sapiens,” and thereafter “search by expression.” We select\nthe human GNF1A chip, and “calculated Shannon entropy” using the drop-down boxes. The following page contains selector boxes\nfor each tissue in the set. For pancreatic islets, we select a Q\ntissue value of < 1.7 × H\ngene. The user should have experiment with the Q\ntissue setting; the lower values (approaching H\ngene) provide smaller\nnumbers, but more tissue-specific genes. Relaxing the value towards a value of\n2 × H\ngene provides more exhaustive, albeit less selectively\nexpressed genes. At Q\nislet at 1.7 × H\ngene, we identify 96\nprobesets (as more than one probeset may exist for each gene, the actual number\nof genes identified is often slightly lower). Selecting to show all Shannon entropy values, we next copy the entire table\ninto Excel, in order to rank-order the hits. Not surprisingly, the\ntop of this list consists of Glucagon (GCG), Insulin (INS), IAPP,\nbut we also notice that not far from the top, some nonendocrine-type genes, such as PNLIPRP1 and CPA2 are present. The reason for this is due to exocrine\ncontamination. The majority of such genes are easily removed by eliminating all genes in which\nQ\npanc < Q\nislet. Ranking the resulting genes provides a\nlist of genes showing the highest selectivity for pancreatic islets compared to\n78 other human tissues (the 30 top-ranked genes of this list is shown in Table\n1). The results include the complement of endocrine terminal products (Ins, Gcg, Sst, Ppy), four Reg-type genes\n(Reg1b, Reg3g, Reg3a, Regl), several\nwell-known endocrine transcripts (Pcsk1 (PC1/3), Iapp, Slc30a8), secretogranins\n(Scgb2a1, Scgn, Scg5, Scg2, Scg3),\nand transcription factors known to function in the islet (FoxA2, Nkx2.2, Isl1).\n\nExample 4(Pituitary versus pancreatic islets: finding common\nneuroendocrine properties (online tutorial 4)). It is known that neuroendocrine\ncell types shares certain characteristics related to production and release of\nsecreted products. The pituitary and islets are highly enriched in cells\nproducing polypeptide hormones. Using Shannon entropy, we will here ask what are the genes that may be in common between pituitary and\npancreatic islets and not expressed widely elsewhere. Similar to above,\nwe select Shannon entropy query in GeneSpeed,\nand input a slightly relaxed Q\ntissue value of 1.8 for both pituitary\nand pancreatic islets. Individually, Q\nIslet < 1.8 × H\ng and Q\npituitary < 1.8 × H\ng identify 292 and 222\nprobesets, respectively. The intersection is 21 probesets, corresponding to 19\nindividual genes (Table 2, 3 probesets for GNAS, guanine nucleotide\nbinding protein were identified). Three genes encode known granule-type\nproteins (ChgrA, secretogranin 2 (SCG2), secretogranin 5 (SCG5)).\nTwo transcription factors are found: InsM1 and ZNF91. The proprotein convertase\nsubtilisin/kexin type-1 inhibitor (PCSKN1) and the peptidylglycine\nalpha-amidating monooxygenase are also present. Other products include CACNA1F (Calcium channel, voltage-dependent, alpha 1F), CNGA3 (Cyclic nucleotide\ngated channel alpha 3), the transmembrane protein TMEM30 as well as\nseveral uncharacterized genes. Many of these genes represent expected hits, and\nshow the value of combining parameters such as tissue uniqueness and\noverlapping gene expression to derive a meaningful candidate repertoire for\nfurther scrutiny.\n3. DISCUSSIONOf the current available places for\ngenomics data reposition, the NCBI GEO (gene expression omnibus, [5]) is presently the most\nexhaustive. The development of GEO proceeds to include data analysis of public\narray-type experiments, which also include those deposited on islets, or\ndeveloping pancreas. The tools are currently limited to analyses performed\nwithin individual experiments, and data results cannot be ported between experiments.\nHowever, no other resource exists with a similar exhaustive compilation of DNA microarray-type datasets,\nand as such, GEO represents a growing and increasingly important pillar for\narray data compilation. In contrast to the more universal user-base that GEO\nseeks to cover, certain resources have also been made available and dedicated\nto the islet community. T1Dbase (http://www.t1dbase.org/)\nwas specifically developed to catalogue information on the genetics of type-I\ndiabetes, and contains extensive information on candidate gene regions [1]. It also contains a\nmicroarray repository and a recently developed Gene Atlas search\nfunction, aimed at providing a rapid visualization of gene expression in\nislets. The strength of the environment lies in the use of Gaggle [6], which is a Java-based\ncommunicator interface to several bioinformatics tools. However, to use this\nrequires a significant knowledge of the Gaggle-implemented tools such as the\nTIGR tmev (http://www.tm4.org/mev.html) or R\n(http://www.r-project.org/),\nnotwithstanding a rather complicated data upload scheme. Despite its strength,\nthis may therefore represent a time-consuming and intellectual barrier to most\nbiologists using the resource irregularly. Another comparable resource, the\nEpconDb (http://www.cbil.upenn.edu/epcondb42/)\n[2], originally generated by the\nEndocrine Pancreas Consortium and funded through the NIH Beta Cell Biology\nConsortium (http://www.betacell.org/), also\nprovides microarray chip repository support. Recently, precalculated analysis\nresults for select experiments are also provided. The structure of the EpConDb\nresource centers on the GUS (genome unified schema), which includes the DOTS\ndatabase. DOTS shares significant similarities to the NCBI-devised Unigene EST\ndatabase, but extend to include splice site data, as well as promoter\ndefinition.The GeneSpeed Beta\nCell site seeks to complement these resources on particularly two fronts: to\nprovide more extensive orthogonal analysis between array experiments and to\nprovide a functional gene list operator workspace, which neither the T1dbase\nnor Epcondb sites allow. To achieve the former, we focused on providing a\nlarger degree of relevant precomputed analyses of array experiments providing\nthese in an easy-to-query format. To achieve the latter, we developed a gene\nlist workspace that would allow for platform-to-platform compatibility using\nthe common Unigene denominator, which is the nexus of the GeneSpeed MySQL\ndatabase. The current version of the database provides certain features not\nfound elsewhere, some of which has been addressed through the demonstration\ncases. Yet, the database is a currently developing structure that in its\npresent form is useful, but easily imagined improved. Therefore, we are\ncurrently focusing on key aspects for the further development of the GeneSpeed\nenvironment. These include the identification of additional relevant microarray\nexperiments; filling out “missing links” by performing stop-gap-type microarray\nexperiments for populating critical, but missing, areas of the pancreatic\nexpression space; improving the search and query formats for user-friendliness;\nand finally developing an export/import interface for pathway analysis programs\nsuch as Ingenuity Pathway Analysis (IPA).The usefulness of GeneSpeed\nBeta Cell database is dependent on the amount of available genomics data\ncontent. A linear increase in number of available datasets and accompanying precomputed\nanalyses translates into an exponentially growing set of query combinations.\nThere are obvious gaps in the available datasets, as multiple null mutations\nhave been created for several key developmental regulators during pancreatic\ndevelopment, and several mutant models resulting in diabetes due to beta-cell\ndysfunction have also been reported, all of which would represent valuable data\nin the present environment. Therefore, we are asking the islet research\ncommunity to share available datasets for multidimensional analysis. Also, we\nwill continue to upload publicly available datasets from the GEO environment.For a wet-biology laboratory like our own, the present\nincarnation of the database has provided means of moving forward in otherwise\ndifficult-to-execute bioinformatics-based questions. We hope that the same\nappreciation may pioneer gene identification challenges in other laboratories\nhereby helping the diabetes research community.4. METHODS4.1. Genomics data incorporation and analysisThe “GeneSpeed Beta Cell” environment was developed using the J2EE platform on a Linux\nserver. For Affymetrix-type genomics data, we compiled the CEL files (raw data)\nassociated with different experiments from different sources and normalized\nthem locally using MAS5.0 algorithm, using an identical scaling factor of 500,\nto ensure optimal comparability in a cross-experimental setting.The microarray experiments\ncurrently available can be grouped, and hence analyzed, according to\nexperimental design type. For time-series experiments (and drug-effect studies),\nan SOM neural network clustering algorithm was applied. The number of clusters\nselected is empirically based on individual results, selecting the minimal\nnumber adequately describing the data complexity. A graphical presentation is\nprovided of the log-transformed expression averages of genes within the\ncluster. Also, the total number of gene number contained/cluster is provided. R\nand Bioconductor [7] were used to accomplish this\ntask.For multicomponent analysis, which also\nincludes single pair-wise analysis, ANOVA testing was performed. For\nmultiple-condition datasets, several pair-wise analyses are provided. These\nresults are depicted through volcano plots. A false discovery rate (FDR) test\ncorrection on the ANOVA result at 10% significance is provided for each plot as\nthe default P-value setting. On the volcano plots, the boxed areas\noutlining a 10% FDR corrected P-value and the −2 to +2 fold regions of\nchange are shown.4.2. Account functionalityThere are two account types in\nGeneSpeed: “guest” and “registered user.” In order to use the workspace\nenvironment registration is required. A “registered user” can log back into\ntheir account to gain access to saved studies. Establishing a registered\naccount is free and can be done on the GeneSpeed registration page (http://genespeed.ccf.org/loginReq.php).\nAn automated password will be sent to the newly registered user. The\nconfidentiality of all registration information is strictly maintained and we\nwill only use such information to notify our users of any disruptions or\nmodifications of the GeneSpeed service. At present, only registered users are\nallowed to use the GeneSpeed Beta Cell database.The GeneSpeed account allows\nregistered users to save gene lists into a private account that is permanent\nand may only be viewed by the owner. The “My Gene Workspace,” on the other\nhand, maintains gene lists temporarily during the current login session; upon\nlogging out the content of the “workspace” will be deleted.4.3. Functional implementation of “My Gene Workspace”The “My Gene Workspace” logic was\ndeveloped using J2EE and sql-type queries. Facilitating cross-platform\ncomparisons, the workspace utilizes Unigene cluster Ids (UID). Consequently,\nthe probeset identification through the experimental analyses is translated\ninto corresponding UID upon transfer to the workspace. As a result, if more\nthan one probeset is detected for a given gene in the analysis, these probesets collapse into\nthe UID of that gene. Secondly, upon selection of the content of a gene list in\nthe workspace, followed by showing the content in the expression space, all probesets corresponding to the\nselected Unigene will be displayed. To reduce ambiguities, we update the system\ncontinuously upon the availability of updated mapping files from NetAffx Analysis Center\nserver. Given that the NCBI Unigene dataset is constantly evolving, updated\nmapping to the most recent UID is done every 6 months.\n\nREFERENCES:\n1. HulbertEMSminkLJAdlemECT1DBase: integration and presentation of complex data for type 1 diabetes researchNucleic Acids Research200735, database issueD742D74617169983\n2. MazzarelliJMBrestelliJGorskiRKEPConDB: a web resource for gene expression related to pancreatic development, beta-cell function and diabetesNucleic Acids Research200735, database issueD751D75517071715\n3. KutchmaAQuayumNJensenJGeneSpeed: protein domain organization of the transcriptomeNucleic Acids Research200735, database issueD674D67917132830\n4. SchugJSchullerWPKappenCSalbaumJMBucanMStoeckertCJJrPromoter features related to tissue specificity as measured by Shannon entropyGenome Biology200564, article R33\n5. BarrettTTroupDBWilhiteSENCBI GEO: mining tens of millions of expression profiles—database and tools updateNucleic Acids Research200735, database issueD760D76517099226\n6. ShannonPTReissDJBonneauRBaligaNSThe Gaggle: an open-source software system for integrating bioinformatics software and data sourcesBMC Bioinformatics20067, article 176\n7. ReimersMCareyVJBioconductor: an open source framework for bioinformatics and computational biologyMethods in Enzymology200641111913416939789" |
