Modeling Mutual Exclusivity of Cancer Mutations

Abstract

In large collections of tumor samples, it has been observed that sets of genes that are commonly involved in the same cancer pathways tend not to occur mutated together in the same patient. Such gene sets form mutually exclusive patterns of gene alterations in cancer genomic data. Computational approaches that detect mutually exclusive gene sets, rank and test candidate alteration patterns by rewarding the number of samples the pattern covers and by punishing its impurity, i.e., additional alterations that violate strict mutual exclusivity. However, the extant approaches do not account for possible observation errors. In practice, false negatives and especially false positives can severely bias evaluation and ranking of alteration patterns. To address these limitations, we develop a fully probabilistic, generative model of mutual exclusivity, explicitly taking coverage, impurity, as well as error rates into account, and devise efficient algorithms for parameter estimation and pattern ranking. Based on this model, we derive a statistical test of mutual exclusivity by comparing its likelihood to the null model that assumes independent gene alterations. Using extensive simulations, the new test is shown to be more powerful than a permutation test applied previously. When applied to detect mutual exclusivity patterns in glioblastoma and in pan-cancer data from twelve tumor types, we identify several significant patterns that are biologically relevant, most of which would not be detected by previous approaches. Our statistical modeling framework of mutual exclusivity provides increased flexibility and power to detect cancer pathways from genomic alteration data in the presence of noise. A summary of this paper appears in the proceedings of the RECOMB 2014 conference, April 2–5.

A Modeling Approach to Explain Mutually Exclusive and Co-Occurring Genetic Alterations in Bladder Tumorigenesis

“…The underlying molecular network should, at least partly, explain such observations. So far, these patterns have been explained in terms of linear pathways: co-occurring mutations tend to target genes in parallel signaling pathways, whereas mutual exclusive alterations may implicate genes involved either in a common pathway, or in different progression pathways, i.e., in different tumor types (1)(2)(3)(4)(5). Mutual exclusivity could also involve genes that are synthetically lethal (6).…”

Sports genetics moving forward: lessons learned from medical research

“…Beyond trait heterogeneity, any single trait is likely to have genetic (different genes) and allelic (different SNPs in the same gene) variability. Analytically modelling the joint effects of multiple genes or allele features without exhaustive multiple testing is an active area of research (27,62). Moreover, interpretation of variation in regulatory regions including the potential for epistasis and epigenetics will be an ongoing effort for the community.…”

CoMEt: A Statistical Approach to Identify Combinations of Mutually Exclusive Alterations in Cancer

“…These observations motivated the development of probabilistic models of mutual exclusivity. These include the Dendrix++ algorithm (an early version of the approach that we present in this paper) and the muex algorithm [23]. Dendrix++ uses a statistical score and was used in TCGA acute myeloid leukemia study [3].…”

“…Dendrix++ uses a statistical score and was used in TCGA acute myeloid leukemia study [3]. The muex algorithm [23] uses a generative model of mutual exclusivity and a likelihood ratio test to score the mutual exclusivity of combinations of mutations. However, we find that the muex score is sensitive to high frequency mutations (see section Comparisons to other methods on real data).…”