BioContext

BioContext extracts and contextualizes molecular process information from biomedical literature.

Key Features:

• Integration of text mining tools: Integrates multiple text mining tools to perform entity recognition, biomolecular event extraction, and contextualization at scale.
• Large-scale corpus coverage: Applied to 10.9 million MEDLINE abstracts and 234,000 open-access full-text articles from PubMed Central.
• Event extraction scale: Identified over 36 million mentions representing approximately 11.4 million distinct biomolecular events.
• Gene and protein recognition: Identifies more than 290,000 distinct genes and proteins with over 80 million mentions across the literature.
• Identifier linking: Links recognized entities to Entrez Gene identifiers where possible.
• Context capture: Captures contextual information including anatomical locations and event qualifiers such as negation and speculation.

Scientific Applications:

• Comprehensive literature overviews: Provides comprehensive overviews of biomolecular processes related to specific genes or anatomical sites.
• Hypothesis generation: Supports hypothesis generation by aggregating event mentions across large literature corpora.
• Systems biology: Supports systems biology studies through extraction of large-scale biomolecular events and entity relationships.
• Event occurrence analysis: Enables analysis of the occurrence and nature of biological events across the literature.

Methodology:

Integrates multiple text mining tools to perform entity recognition, biomolecular event extraction, and contextualization, applies these methods to MEDLINE abstracts and PubMed Central open-access full-text articles, and links entities to Entrez Gene identifiers where possible.