FACTA plus

FACTA plus extracts and identifies indirect associations between biomedical concepts from MEDLINE abstracts for literature-based discovery and exploration of gene, disease, and chemical relationships.

Key Features:

• Real-time text mining: Pre-indexes words and biomedical concepts from MEDLINE to support queries using free keywords or Boolean combinations and enable immediate retrieval.
• Biomolecular event detection: Detects biomolecular events in text using a machine learning model.
• Co-occurrence statistics: Applies statistical analysis of term co-occurrence to uncover hidden associations between biomedical concepts.
• Visualization of associations: Presents discovered associations with category labels and measures of relative importance to aid interpretation.
• Concept indexing from multiple databases: Uses concept IDs and names/synonyms collected from UniProt, BioThesaurus, UMLS, KEGG, and DrugBank for comprehensive concept mapping.

Scientific Applications:

• Gene–disease and drug–target discovery: Identify potential gene-disease links and drug-target interactions that are not explicitly stated in the literature.
• Event-centric literature analysis: Explore biomolecular events and their implications as described in MEDLINE abstracts.
• Network visualization for hypothesis generation: Visualize complex networks of relationships to support hypothesis generation and experimental design.

Methodology:

The system combines text mining with pre-indexing of words and biomedical concepts from MEDLINE, machine learning-based biomolecular event detection, statistical co-occurrence analysis, and real-time processing with visualization.