EPIMUTESTR: a nearest neighbor machine learning approach to predict cancer driver genes from the evolutionary action of coding variants. Issue 12 (12th April 2022)

• Record Type: Journal Article
• Authors: Parvandeh, Saeid; Donehower, Lawrence A; Panagiotis, Katsonis; Hsu, Teng-Kuei; Asmussen, Jennifer K; Lee, Kwanghyuk; Lichtarge, Olivier
• Is Part Of: Nucleic acids research. Volume 50:Issue 12(2022)
• Page Start: e70
• Page End: e70
• Publication Date: 2022-04-12
• Subjects: Nucleic acids -- Periodicals; Molecular biology -- Periodicals; 572.805
• DOI: 10.1093/nar/gkac215
• Languages: English
• ISSNs: 0305-1048

Abstract

Abstract: Discovering rare cancer driver genes is difficult because their mutational frequency is too low for statistical detection by computational methods. EPIMUTESTR is an integrative nearest-neighbor machine learning algorithm that identifies such marginal genes by modeling the fitness of their mutations with the phylogenetic Evolutionary Action (EA) score. Over cohorts of sequenced patients from The Cancer Genome Atlas representing 33 tumor types, EPIMUTESTR detected 214 previously inferred cancer driver genes and 137 new candidates never identified computationally before of which seven genes are supported in the COSMIC Cancer Gene Census. EPIMUTESTR achieved better robustness and specificity than existing methods in a number of benchmark methods and datasets.