Identification of Cancer Driver Modules by Combining Network Functional and Topology Information.

Xin Chu,Feng Li,Hongyu Duan,Junliang Shang,Juan Wang,Jin-Xing Liu

DOI: https://doi.org/10.1109/bibm55620.2022.9995423

2022-01-01

Abstract:Accurate identification of cancer driver modules or pathways is important for controlling disease progression and timely treatment. In recent years, most approaches have been based on mutation data combined with gene interaction networks to identify cancer driver modules, but cancer-related genes tend to interact with each other, and the mutations they experience disruption their neighbors. Therefore, we propose a framework that combines network function and topological information to quantify the extent to which mutated genes disrupt their neighbors. Firstly, similarity in protein-protein interaction networks binds to high coverage and high mutual exclusivity of mutant genes, which are used to obtain the impact of the interaction between two mutant genes on biological function. Secondly, we quantified the degree of gene disruption by mutant genes in their neighborhood using an adaptive spread strength measure to obtain the gene spread strength network (GSSN). Finally, the module is extended using CFinder strategy to obtain the optimal driving module. We apply our method to 12 cancer datasets, and the experimental results show that our method outperforms the other three methods on most datasets. At the same time, we also analyze common and low-frequency driver modules in cancer.