CHAT

CHAT annotates PubMed abstracts with a Hallmarks of Cancer-aligned taxonomy to organize and evaluate cancer-related literature and support identification of biomarkers and therapeutic targets.

Key Features:

• Automatic text mining: Retrieves and categorizes cancer-related references from PubMed using automatic text mining methods.
• Hallmarks of Cancer taxonomy: Employs an extensive taxonomy aligned with the Hallmarks of Cancer to structure literature annotations.
• PubMed abstract annotation: Annotates PubMed abstracts with hallmark categories for structured representation of literature.
• Literature-to-hallmark alignment: Aligns scientific literature with hallmark categories to organize existing knowledge.
• Correlation identification: Identifies correlations between hallmarks and extrinsic factors, biomarkers, and therapeutic targets.
• Evaluation: Was assessed through intrinsic evaluation and case studies demonstrating efficiency and accuracy in classifying cancer-related literature.

Scientific Applications:

• Knowledge organization: Facilitates retrieval and structured overview of cancer literature by hallmark categories.
• Biomarker discovery: Supports identification of novel biomarkers associated with specific Hallmarks of Cancer.
• Therapeutic target identification: Aids in pinpointing therapeutic targets linked to particular hallmark alterations.
• Environmental and lifestyle studies: Enables analysis of correlations between extrinsic factors and hallmark alterations.
• Research prioritization: Helps researchers identify critical areas for further investigation within cancer research.

Methodology:

CHAT creates an extensive taxonomy aligned with the Hallmarks of Cancer and applies automatic text mining to retrieve, categorize, and annotate PubMed abstracts, with performance assessed via intrinsic evaluation and case studies.