Identification of Common Driver Gene Modules and Associations between Cancers through Integrated Network Analysis

Area of Science:

• Oncology
• Bioinformatics
• Systems Biology

Background:

• High-throughput biological data offers insights into tumor evolution.
• Identifying common driver genes and cancer associations is crucial for understanding cancer.
• Existing methods face challenges in deciphering these complex relationships.

Purpose of the Study:

• To develop a computational tool, ComCovEx, for identifying common cancer driver gene modules between two cancer types.
• To investigate associations between different cancer types using biological network analysis.
• To provide novel insights into the pathological basis and potential therapeutic strategies for associated cancers.

Main Methods:

• ComCovEx utilizes an exclusivity-coverage iteration strategy within local signaling networks.
• It identifies common driver gene modules with significant coverage and exclusivity for paired cancers.
• Cancer pair associations are statistically evaluated using Fisher's exact test.

Main Results:

• ComCovEx was applied to 11 The Cancer Genome Atlas (TCGA) cancer datasets.
• The analysis identified 13 significantly associated cancer pairs.
• Numerous biologically significant common gene modules were discovered for these associated cancers.

Conclusions:

• The identified cancer associations and common gene modules provide a deeper understanding of shared pathological mechanisms.
• ComCovEx offers a valuable approach for uncovering novel relationships between cancer types.
• The findings suggest potential new avenues for cancer diagnosis and drug treatment development in associated cancers.