@article {Lai082511, author = {Liming Lai and Jason Hennessey and Valerie Bares and Eun Woo Son and Yuguang Ban and Wei Wang and Jianli Qi and Gaixin Jiang and Arthur Liberzon and Steven Xijin Ge}, title = {GSKB: A gene set database for pathway analysis in mouse}, elocation-id = {082511}, year = {2016}, doi = {10.1101/082511}, publisher = {Cold Spring Harbor Laboratory}, abstract = {Interpretation of high-throughput genomics data based on biological pathways constitutes a constant challenge, partly because of the lack of supporting pathway database. In this study, we created a functional genomics knowledgebase in mouse, which includes 33,261 pathways and gene sets compiled from 40 sources such as Gene Ontology, KEGG, GeneSetDB, PANTHER, microRNA and transcription factor target genes, etc. In addition, we also manually collected and curated 8,747 lists of differentially expressed genes from 2,526 published gene expression studies to enable the detection of similarity to previously reported gene expression signatures. These two types of data constitute a Gene Set Knowledgebase (GSKB), which can be readily used by various pathway analysis software such as gene set enrichment analysis (GSEA). Using our knowledgebase, we were able to detect the correct microRNA (miR-29) pathway that was suppressed using antisense oligonucleotides and confirmed its role in inhibiting fibrogenesis, which might involve upregulation of transcription factor SMAD3. The knowledgebase can be queried as a source of published gene lists for further meta-analysis. Through meta-analysis of 56 published gene lists related to retina cells, we revealed two fundamentally different types of gene expression changes. One is related to stress and inflammatory response blamed for causing blindness in many diseases; the other associated with visual perception by normal retina cells. GSKB is available online at http://ge-lab.org/gs/, and also as a Bioconductor package (gskb, https://bioconductor.org/packages/gskb/). This database enables in-depth interpretation of mouse genomics data both in terms of known pathways and the context of thousands of published expression signatures.}, URL = {https://www.biorxiv.org/content/early/2016/10/24/082511}, eprint = {https://www.biorxiv.org/content/early/2016/10/24/082511.full.pdf}, journal = {bioRxiv} }