Discovering functional evolutionary dependencies in human cancers.

Area of Science:

• Genomics
• Computational Biology
• Cancer Research

Background:

• Cancer cells harbor genomic alterations conferring selective advantages.
• Predicting and validating these alterations, especially their combined effects, remains a significant challenge in cancer genomics.

Purpose of the Study:

• To infer functional genomic alterations and identify key combinations in human tumors.
• To validate computational predictions using experimental data from cancer cell lines.
• To understand how coexisting genomic alterations influence cellular response to perturbations.

Main Methods:

• Utilized a Bayesian inference framework to analyze >9,000 human tumor datasets.
• Validated predictions through high-throughput genetic and pharmacological screenings on 2,000 cancer cell lines.
• Investigated mutually exclusive and co-occurring alterations to identify functional redundancies and synergistic interactions.

Main Results:

• Identified specific combinations of cancer genomic alterations, including PIK3CA and NFE2L2 co-alteration in squamous cell carcinomas.
• Demonstrated that mutually exclusive alterations can lead to functional redundancy, while co-occurring alterations can increase oncogene addiction.
• Validated computational predictions with experimental data, highlighting the framework's efficacy.

Conclusions:

• Developed a framework integrating computational, experimental, and clinical data to study combinatorial effects of cancer genomic alterations.
• The study provides insights into evolutionary trajectories and therapeutic vulnerabilities driven by specific genomic event combinations.
• This approach aids in understanding complex cancer genomics and predicting responses to targeted therapies.