Network control principles for identifying personalized driver genes in cancer

Area of Science:

• Computational biology
• Cancer genomics
• Systems biology

Background:

• Understanding tumor heterogeneity requires identifying personalized driver genes (PDGs) to link patient-specific genotypes to phenotypes.
• Existing driver gene identification methods often rely on cohort data, neglecting individual patient information.
• Network control principles offer a novel computational approach for discovering cancer driver genes at the individual level.

Purpose of the Study:

• To explore network control methods for identifying personalized driver genes (PDGs) in cancer.
• To frame cancer progression as a network control problem, where PDGs are genes altered by oncogenic signals.
• To provide a comprehensive review and assessment of these methods for individual cancer analysis.

Main Methods:

• Reviewed network reconstruction methods applicable to single biological samples.
• Developed and applied novel network control methods to single-sample networks for PDG identification.
• Assessed the performance of network structure control-based PDG identification on TCGA cancer datasets.

Main Results:

• Demonstrated the utility of network control principles for identifying PDGs in individual cancer patients.
• Provided a performance evaluation of network control methods using publicly available TCGA data.
• Highlighted the potential of these methods for personalized cancer driver gene discovery.

Conclusions:

• Network control methods represent a promising avenue for personalized driver gene discovery in cancer.
• Future research should focus on applying these methods to diverse biological processes and complex diseases.
• The developed methods and data packages facilitate further investigation into individual-level cancer genomics.