Using mutual exclusivity of mutations

CoMEt (the Combinations of Mutually Exclusive Alterations) ⁴ uses mutual exclusivity technique to detect cancer driver modules. Because different cancer patients have different combinations of genomic alterations which develop the disease, CoMEt detects combinations of alterations (i.e. modules of mutated genes) in the same pathway, which are mutual exclusive across samples. The method uses the exact statistical test to test mutual exclusivity and it does simultaneous analysis for mutually exclusive alterations specific to cancer subtypes. The advantage of this method is that it has a low computational complexity. Similarly, WeSME ²⁰ also assesses the mutual exclusivity of mutations of genes to detect cancer drivers. However, instead of evaluating genes in the same pathway, WeSME only considers gene pairs and the gene pairs whose mutations have a significantly mutual exclusivity are considered as modular candidate cancer drivers.

MEMo (Mutual Exclusivity Modules) ¹⁷ applies mutual exclusivity technique in biological networks to identify oncogenic network modules. According to ¹⁷, although individual tumours of the same cancer type may have different genomic alterations, these alterations just happen in a restricted number of pathways. In addition, alterations in the same pathway are not likely to exist in the same patient. Based on these, MEMo does correlation analysis and applies statistical tests to detect network modules based on three criteria: (1) genes in a network module are altered across the sample; (2) member genes tend to join into the same biological process; (3) alterations in modules are mutually exclusive. The method is applied to the glioblastoma multiforme (GBM) dataset and detects successfully known network modules, i.e., groups of cancer drivers, in GBM.