Cancer driver gene discovery through an integrative genomics approach in a non-parametric Bayesian framework

• DOI: 10.1093/bioinformatics/btw662
• Publication Date: 2016-10-17T19:06:19Z
• Authors: Hai Yang; Qiang Wei; Xue Zhong; Hushan Yang; Bingshan Li
• Keywords: Leverage (statistics)

Comprehensive catalogue of genes that drive tumor initiation and progression in cancer is key to advancing diagnostics, therapeutics treatment. Given the complexity cancer, far from complete yet. Increasing evidence shows driver exhibit consistent aberration patterns across multiple-omics tumors. In this study, we aim leverage complementary information encoded each omics data identify novel through an integrative framework. Specifically, integrated mutations, gene expression, DNA copy numbers, methylation protein abundance, all available The Cancer Genome Atlas (TCGA) developed iDriver, a non-parametric Bayesian framework based on multivariate statistical modeling unsupervised fashion. iDriver captures inherent clusters aberrations constructs background distribution used assess calibrate confidence identified multi-dimensional genomic data.We applied method 4 types TCGA candidate are highly enriched with known drivers. (e.g.: P < 3.40 × 10 -36 for breast cancer). We particularly interested observed multiple lines supporting evidence. Using systematic evaluation independent aspects, 45 were not previously these types. finding has important implications integrating additional statistics can help drivers guide next stage genomics research.The C ++ source code freely at https://medschool.vanderbilt.edu/cgg/ [e-mail] or [e-mail] Bioinformatics online.