ENQUIRE

ENQUIRE reconstructs and expands co-occurrence networks of genes and biomedical ontologies (MeSH) from user-selected literature corpora by iteratively querying PubMed to reveal significant interrelations relevant to molecular networks and disease.

Key Features:

• Co-occurrence network reconstruction: Reconstructs co-occurrence networks of genes and MeSH terms from a user-selected input corpus.
• Iterative PubMed query generation: Iteratively generates PubMed queries based on significant interrelations identified within networks to expand the corpus.
• Text mining and automatic querying: Integrates text mining with automatic querying to extract co-occurrence information from literature.
• Network-based statistics: Applies network-based statistics to identify significant interrelations and mitigate literature biases.
• Context specificity and confidence scoring: Produces high-confidence co-occurrence gene networks that preserve context specificity of the input topics.
• Enrichment and linkage detection: Identifies interlinked genes and enriched pathways across analyzed topics such as cancer, cell differentiation, and immunity.

Scientific Applications:

• Molecular network analysis in disease: Expands and analyzes gene–MeSH co-occurrence relationships to study molecular networks linked to diseases.
• Literature annotation and curation augmentation: Reveals non-obvious interrelations in PubMed literature to support literature annotation and augment manual curation.
• Hypothesis generation and target identification: Supports generation of hypotheses and identification of candidate molecular targets for experimental follow-up.
• Domain-specific discovery: Has been applied to cancer, cell differentiation, and immunity to identify topic-specific interlinked genes and enriched pathways.

Methodology:

Reconstructs and expands gene–MeSH co-occurrence networks from user-selected corpora, iteratively generates PubMed queries based on significant network interrelations, and applies text mining, automatic querying, and network-based statistics to expand and score networks until an interconnected framework relevant to the initial corpus is achieved.