LipiDisease: associate lipids to diseases using literature mining

Area of Science:

• Biochemistry and Molecular Biology
• Bioinformatics
• Genomics and Proteomics

Background:

• Lipids are crucial for cellular functions, and their dysregulation is linked to diseases like obesity, diabetes, and cancer.
• Investigating lipid profiles offers insights into cellular mechanisms and potential therapeutic interventions.
• Current research lacks a comprehensive resource for analyzing cumulative lipid-disease associations.

Purpose of the Study:

• To develop a web server, LipiDisease, for disease enrichment analysis based on lipid sets.
• To provide a tool that predicts and ranks associations between lipids and diseases using biomedical literature data.

Main Methods:

• The LipiDisease web server was implemented to analyze millions of PubMed records.
• It predicts associations between lipids and diseases using statistical analysis and random simulations.
• The server currently incorporates data on 4798 lipids and 4270 diseases.

Main Results:

• LipiDisease provides a ranked list of lipid-disease associations based on statistical significance (false discovery rates).
• The tool leverages a vast corpus of biomedical literature to identify these associations.
• The database is designed to expand as new research on lipids and diseases is published.

Conclusions:

• LipiDisease offers a novel resource for exploring lipid-disease relationships.
• This tool can advance our understanding of lipidomics in various pathological conditions.
• The web server facilitates research into potential lipid-targeted interventions for diseases.