OmixLitMiner

OmixLitMiner retrieves and analyzes PubMed literature to prioritize biomolecule leads from proteomics and genomics datasets for follow-up experimentation.

Key Features:

• Implementation: Implemented in R for computational processing and analysis of literature queries.
• Literature retrieval: Performs systematic literature retrieval from PubMed.
• Identifier handling: Leverages UniProt protein identifiers, gene names, and gene synonyms for query construction.
• Contextual keywords: Integrates user-defined contextual keywords based on gene ontology to tailor search specificity.
• Search flexibility: Supports strict and lenient search strategies to adjust sensitivity of literature retrieval.
• Output categorization: Categorizes outputs into three groups reflecting the degree of characterization of genes or proteins.
• Decision support: Automates retrieval and analysis to aid prioritization of novel biomolecule leads while excluding well-characterized markers.
• Retrospective validation: Demonstrated on three cancer proteomics publications and one cancer genomics publication to corroborate study decisions and identify additional candidates.

Scientific Applications:

• Proteomics lead prioritization: Prioritizes protein candidates from proteomics experiments for downstream experimental validation.
• Genomics candidate filtering: Filters and ranks gene candidates from genomics datasets using literature evidence.
• Follow-up experiment planning: Guides selection of targets for follow-up experiments by indicating degree of prior characterization.
• Retrospective analysis: Validates and extends findings from published cancer proteomics and genomics studies by identifying additional biomolecule candidates.

Methodology:

Implemented in R; performs systematic PubMed queries using UniProt identifiers, gene names and synonyms combined with user-defined gene ontology keywords; supports strict or lenient search strategies and categorizes results into three characterization-based groups.