goSTAG

goSTAG identifies representative Gene Ontology (GO) biological themes from gene lists by clustering enriched GO terms, constructing GO subtrees, and selecting subtree terms with the greatest connectivity to the GO root to summarize over-representation analysis (ORA) results.

Key Features:

• GO subtree utilization: Employs Gene Ontology (GO) subtrees to provide more nuanced annotation of genes within a set.
• Clustering by significance: Clusters GO terms based on p-values derived from enrichment significance tests.
• Subtree construction: Constructs a GO subtree for each cluster of enriched terms.
• Representative theme selection: Selects the term with the most paths to the GO root within each subtree as the representative biological theme for that cluster.
• Reduced subjectivity and comparability: Tags clusters with biologically meaningful themes to reduce subjectivity and facilitate comparison of enriched categories across multiple analyses.
• High-throughput compatibility: Accepts gene lists generated from microarray and RNA sequencing (RNA-Seq) experiments for GO enrichment and summarization.

Scientific Applications:

• Drug exposure transcriptomics: Applied to microarray gene expression data from rat bone marrow exposed to cancer therapeutic drugs to discern the biological impact of drug administration order, highlighting pathways such as negative regulation of MAP kinase activity and nucleoside triphosphate catabolic process.
• ORA result summarization: Summarizes and annotates ORA outputs from high-throughput genomic studies to reveal coherent biological themes.

Methodology:

Performs GO enrichment analysis on input gene lists; clusters enriched GO terms using p-values from enrichment tests; constructs a GO subtree for each cluster; and selects the subtree term with the most paths to the GO root as the biological theme for that cluster.