CM-viewer: Visualizing interaction network of co-mutated and mutually exclusively mutated cancer genes

Abstract

Cancer genes usually play a crucial role in regulating cell growth. Normal cells transform into malignant tumors by the acquisition of accumulated genetic mutations that enable them to evade normal growth control. It is therefore important to understand the relationships between mutations during cancer development and progression. Although cancer genes with co-occurring and mutually exclusive mutations have already been studied on different scales, there is no timely updated interaction network available for co-mutated and mutually exclusively mutated cancer genes. Therefore, we firstly downloaded 567 cancer genes from COSMIC (catalogue of somatic mutations in cancer) cancer gene census. Secondly, somatic mutations of 71 cancer genomics projects were downloaded from the ICGC (International Cancer Genome Consortium) data portal. Thirdly, mutated cancer genes and affected donors were extracted from the ICGC data to form a mutation matrix where rows are genes, columns are donors, 1 denotes occurrence, and 0 denotes absence of mutation. Afterwards, co-mutated and mutually exclusively mutated cancer gene pairs were identified using DISCOVER (discrete independence statistic controlling for observations with varying event rates). Finally, CM-viewer was developed to visualize the interaction network of cancer genes with co-occurring and mutually exclusive mutations. It is an online visualization tool as well as a biological database. It promises to understand how gene mutations contribute to tumorigenesis and to identify key biomarkers and drug targets for cancer. CM-viewer is freely available at http://www.zhounan.org/comutgene.