GO2MSIG

GO2MSIG generates MSigDB-format gene set collections by leveraging the Gene Ontology (GO) hierarchical structure and association data to produce gene sets suitable for Gene Set Enrichment Analysis (GSEA, Broad Institute).

Key Features:

• Cross-species data integration: Uses the GO association database (covering over 200,000 species), Entrez gene2go tables, and array annotation files to assemble GO-based gene sets for multiple organisms.
• MSigDB output compatibility: Produces gene sets in MSigDB format compatible with GSEA and other software that accept MSigDB-formatted collections such as ErmineJ.
• Evidence-code filtering: Allows restriction of gene–GO mappings by GO evidence codes.
• Gene descriptor remapping: Remaps gene descriptors and synonymous GO terms to preferred identifiers.
• Set-size filtering: Filters gene sets by minimum and maximum set size.
• GO hierarchy caching: Caches the GO term hierarchy to expedite repeated queries.
• Up-to-date annotation generation: Generates gene sets from current GO annotation sources to reflect recent GO updates.

Scientific Applications:

• Gene set enrichment analysis (GSEA): Supplies MSigDB-format gene sets for enrichment testing across species using GSEA.
• Functional genomics: Facilitates identification of biological processes and pathways associated with gene expression or genomic signals.
• Comparative genomics: Enables cross-species comparisons by generating consistent GO-based gene sets for different organisms.
• Systems biology: Supports pathway- and network-level analyses using GO-derived gene sets.
• Transcriptomic and genomic studies: Provides tailored gene sets for interpretation of transcriptomic and other genome-scale experiments.

Methodology:

Integrates annotation sources (GO association database, Entrez gene2go, array annotation files), uses GO hierarchical relationships to construct gene sets, and applies evidence-code filtering, descriptor/synonym remapping, set-size filtering, and GO hierarchy caching.