Proceeding talk – Theme: Systems.
Abstract
The somatic mutations that drive cancer development tend to be mutually exclusive across tumors, providing a signal for distinguishing driver mutations from numerous random passenger mutations. The number of mutations per tumor can vary over several orders of magnitude, which can confound the exclusivity signal. However, current statistical tests for exclusivity that incorporate gene and sample mutational frequency are computationally expensive and limited in the tail of the distribution. We formulate a weighted exact test for mutational exclusivity that conditions on the number of mutated samples in each gene and per-gene, per-sample mutation probabilities. We provide an exact formula and a fast and accurate approximation for the weighted test. We analyze hundreds of colorectal and endometrial samples from The Cancer Genome Atlas which have large variation in mutation rates. In both cancer types, the weighted test identifies mutually exclusive mutations in cancer genes with fewer false positives than earlier approaches.
Authors
Mark Leiserson, Brown University, United States
Matthew Reyna, Brown University, United States
Benjamin Raphael, Brown University, United States