Abstract Comprehensive data-driven discovery of cancer driver genes, including tumor suppressor genes (TSGs) and oncogenes (OGs), is imperative for cancer prevention, diagnosis, and treatment. Although epigenetic alterations are important contributors to tumor initiation and progression, most known driver genes were identified based on genetic alterations alone, and it remains unclear to what the extent epigenetic features would facilitate the identification and characterization of cancer driver genes. Here we developed a prediction algorithm DORGE (Discovery of Oncogenes and tumor suppressoR genes using Genetic and Epigenetic features), which integrates the most comprehensive collection of tumor genetic and epigenetic data to identify TSGs and OGs, particularly those with rare mutations. DORGE identified histone modifications as strong predictors for TSGs, and it found missense mutations, super enhancer percentages, and methylation differences between cancer and normal samples as strong predictors for OGs. We extensively validated novel cancer driver genes predicted by DORGE using independent functional genomics data. We also found that the dual-functional genes, which are both TSGs and OGs predicted by DORGE, are enriched at hubs in protein-protein interaction and drug-gene networks. Overall, our study has deepened the understanding of epigenetic mechanisms in tumorigenesis and revealed a previously undetected repertoire of cancer driver genes.

DORGE: Discovery of Oncogenes and Tumor SuppressoR Genes Using Genetic and Epigenetic Features