HPIminer: A text mining system for building and visualizing human protein interaction networks and pathways

Area of Science:

• Biochemistry
• Bioinformatics
• Computational Biology

Background:

• Understanding protein-protein interactions (PPI) and pathways is crucial for deciphering cellular functions and disease mechanisms.
• Biomedical literature contains extensive, yet largely untapped, information on human PPI and pathways.
• Automated systems are needed to extract and visualize this complex biological information.

Purpose of the Study:

• To develop HPIminer, a text mining system for visualizing human protein interactions and pathways from biomedical literature.
• To integrate information from literature with curated databases for a comprehensive network view.
• To identify novel protein interactions from literature that are not present in existing databases.

Main Methods:

• HPIminer was developed as an automated text mining system.
• The system extracts human PPI information and interaction pairs from biomedical literature.
• Extracted data is visualized using information from the HPRD and KEGG databases.

Main Results:

• HPIminer successfully extracts and visualizes human PPI and associated pathways.
• The system integrates data from both literature and curated databases (HPRD, KEGG).
• Novel protein interactions discovered solely from literature are highlighted, offering new insights.

Conclusions:

• HPIminer is the first system to construct interaction networks from both literature and curated databases.
• The system provides a more informative network visualization compared to existing tools.
• HPIminer enhances the understanding of human protein interactions and their roles in diseases.