BioTextQuest+

BioTextQuest+ performs biomedical literature mining and integrated bioinformatics analyses to identify bioentities and organize literature-based knowledge for functional interpretation.

Key Features:

• Enhanced querying capabilities: Queries PubMed and OMIM to retrieve abstracts matching specified queries.
• Bioentity recognition and functional annotation: Identifies genes, proteins, molecular functions, pathways, and biological processes within retrieved documents for functional annotation.
• Vector Space Model and similarity computation: Represents text records as term-based vectors and computes pairwise document similarities.
• Document clustering and term co-occurrence analysis: Applies clustering algorithms to organize documents into clusters and analyzes term co-occurrence patterns within clusters.
• Integration with biological repositories and software tools: Connects core functionalities with external biological repositories and software tools to access additional bioinformatics services.
• Advanced parameterization for expert analyses: Provides configurable analysis parameters to tailor computational processing.

Scientific Applications:

• Author disambiguation: Supports disambiguation of authors across biomedical literature.
• Functional term enrichment: Facilitates identification of enriched molecular functions, pathways, or biological processes.
• Knowledge acquisition and concept discovery: Enables extraction of concepts and the construction of literature-derived knowledge.
• Disease association analysis: Aids in linking major human diseases, exemplified by obesity and ageing, through literature-derived relationships.

Methodology:

Abstracts are collected from Medline literature and OMIM based on queries; relevant terms are identified to represent text records within a Vector Space Model; pairwise document similarities are calculated and clustering algorithms transform similarities into organized clusters, and term co-occurrence patterns are analyzed.