IDSL.GOA: Gene Ontology Analysis for Interpreting Metabolomic datasets

Area of Science:

• Metabolomics and Bioinformatics
• Systems Biology
• Genomics and Proteomics

Background:

• Metabolomic data interpretation often relies on pathway analysis, but database inconsistencies limit comprehensive understanding.
• Gene Ontology (GO) term analysis is standard for gene, transcript, and protein data but lacks application to metabolomics.
• Existing pathway databases have varying metabolite coverage and definitions, potentially leading to missed biological insights.

Approach:

• Developed a novel knowledgebase (KB) integrating metabolic GO terms, genes, EC annotations, and metabolites.
• Created the Gene Ontology Analysis by the Integrated Data Science Laboratory for Metabolomics and Exposomics (IDSL.GOA) online tool for metabolite GO over-representation analysis.
• Applied IDSL.GOA to a case study of older vs. young female brain cortex metabolomes.

Key Points:

• The IDSL.GOA KB contains 2,393 metabolic GO terms, 3,144 genes, 1,492 EC annotations, and 2,621 metabolites.
• Analysis of brain cortex metabolomes identified 82 significantly overrepresented GO terms (FDR <0.05).
• IDSL.GOA uncovered key metabolic processes not present in other pathway databases, demonstrating its unique interpretive power.

Conclusions:

• Metabolite list interpretation should incorporate GO terms for a more comprehensive analysis, complementing traditional pathway mapping.
• The IDSL.GOA tool offers a standardized and accurate method for applying GO analysis to metabolomic datasets.
• This approach enhances biological discovery by providing deeper insights into metabolic functions and pathways.