Ontolomics-P: Advancing Proteomics Data Interpretation through GPT-4o Reannotated Topic Ontology and Data-Driven...

Area of Science:

• Proteomics
• Systems Biology
• Bioinformatics

Background:

• Functional enrichment analysis (e.g., Gene Ontology) is crucial for interpreting proteomics data.
• Conventional methods often yield redundant terms, hindering biological insights.
• Lack of specialized tools complicates exploration of protein expression in specific contexts.

Purpose of the Study:

• To develop Ontolomics-P, a web-based tool for advanced proteomics data interpretation.
• To address challenges of redundant functional annotations and limited exploration tools.
• To integrate quantitative proteomic data for data-driven biological context exploration.

Main Methods:

• Integration of topic modeling (Latent Dirichlet Allocation) with Gene Ontology semantic similarity.
• Refinement and reannotation of topics using the GPT-4o language model.
• Incorporation of quantitative proteomic data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) database across 10 cancer types.

Main Results:

• Consolidation of redundant Gene Ontology terms into coherent, interpretable topics.
• Creation of a novel topics database for precise functional insights.
• Demonstrated streamlining of workflows and provision of actionable insights through case studies.

Conclusions:

• Ontolomics-P significantly advances proteomics data interpretation.
• The tool offers innovative solutions for functional annotation, quantitative exploration, and visualization.
• Ontolomics-P empowers researchers to accelerate discoveries in systems biology and beyond.