Mutation-profile-based methods for understanding selection forces in cancer somatic mutations: a comparative analysis

// Zhan Zhou 1, 4, * , Yangyun Zou 1, * , Gangbiao Liu 1 , Jingqi Zhou 1 , Jingcheng Wu 4 , Shimin Zhao 2 , Zhixi Su 1 and Xun Gu 3, 1 1 Ministry of Education Key Laboratory of Contemporary Anthropology, School of Life Sciences, Fudan University, Shanghai, China 2 State Key Laboratory of Genetic Engineering, Collaborative Innovation Center of Genetics and Development, School of Life Sciences, Fudan University, Shanghai, China 3 Department of Genetics, Development and Cell Biology, Program of Bioinformatics and Computational Biology, Iowa State University, Ames, Iowa, USA 4 College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China * These authors have contributed equally to this work Correspondence to: Zhixi Su, email: zxsu@fudan.edu.cn Xun Gu, email: xgu@iastate.edu Keywords: cancer somatic mutations, mutation profile, natural selection, cancer-associated genes, evolution Received: October 26, 2016      Accepted: July 12, 2017      Published: July 19, 2017 ABSTRACT Human genes exhibit different effects on fitness in cancer and normal cells. Here, we present an evolutionary approach to measure the selection pressure on human genes, using the well-known ratio of the nonsynonymous to synonymous substitution rate in both cancer genomes ( C N / C S ) and normal populations ( p N / p S ). A new mutation-profile-based method that adopts sample-specific mutation rate profiles instead of conventional substitution models was developed. We found that cancer-specific selection pressure is quite different from the selection pressure at the species and population levels. Both the relaxation of purifying selection on passenger mutations and the positive selection of driver mutations may contribute to the increased C N / C S values of human genes in cancer genomes compared with the p N / p S values in human populations. The C N / C S values also contribute to the improved classification of cancer genes and a better understanding of the onco-functionalization of cancer genes during oncogenesis. The use of our computational pipeline to identify cancer-specific positively and negatively selected genes may provide useful information for understanding the evolution of cancers and identifying possible targets for therapeutic intervention.