Network-based integration of multi-omics data for prioritizing cancer genes

Area of Science:

• Computational biology
• Genomics
• Cancer research

Background:

• Cancer is characterized by molecular aberrations like genomic changes, promoter hypermethylation, and microRNA dysregulation.
• Tumor heterogeneity complicates understanding how these events impact cellular transcriptome and proteome.
• Protein interaction networks offer a way to link aberration events to gene and protein expression changes.

Purpose of the Study:

• To develop a computational method for prioritizing cancer genes by integrating diverse molecular data.
• To explain tumor heterogeneity by identifying functional convergence of aberration events.

Main Methods:

• Developed NetICS (Network-based Integration of Multi-omics Data), a graph diffusion-based method.
• Integrated multi-omics data onto a directed functional interaction network.
• Prioritized genes based on mediator effect (proximity to aberrations and differential expression) and aggregated ranks.

Main Results:

• NetICS successfully prioritized cancer genes by integrating multi-omics data.
• The method demonstrated competitive performance in predicting known cancer genes.
• Generated robust gene lists from TCGA data across five cancer types, aiding in explaining heterogeneity.

Conclusions:

• NetICS provides a framework for understanding cancer gene function and tumor heterogeneity.
• The integration of multi-omics data through network analysis is crucial for cancer research.
• NetICS aids in identifying key genes driving cancer development and progression.