A literature based method for identifying gene-disease connections.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Identifying genes linked to specific diseases is crucial for understanding disease mechanisms and developing targeted therapies.
• The existing literature contains vast amounts of gene-disease association data, but extracting this information systematically is challenging.
• Gene symbol ambiguity and the presence of aliases complicate automated information retrieval.

Purpose of the Study:

• To develop and present a robust statistical method for the rapid identification of gene-disease associations from scientific literature.
• To address challenges in gene symbol disambiguation and the handling of gene symbol aliases.
• To demonstrate the method's utility by identifying genes associated with breast cancer.

Main Methods:

• A novel statistical approach was developed to analyze scientific literature for gene-disease relationships.
• The method incorporates a strategy for managing gene symbol aliases.
• A technique for disambiguating gene symbols from other abbreviations was implemented.
• The relevance of identified genes to a query disease was statistically computed.

Main Results:

• The presented method successfully identified known gene sets associated with specific diseases from the literature.
• The approach demonstrated effectiveness in handling gene symbol variations and disambiguating gene names.
• The methodology was successfully applied to identify genes relevant to breast cancer.

Conclusions:

• The developed statistical method provides an efficient and comprehensive tool for discovering gene-disease associations.
• This approach enhances the systematic extraction of valuable genetic information from biomedical literature.
• The method has significant implications for accelerating research in areas like cancer genomics and personalized medicine.