Integrative Analysis Reveals Novel Driver Genes and Molecular Subclasses of Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is a heterogeneous disease with various genetic and epigenetic abnormalities. Previous studies of HCC driver genes were majority based on the frequency of mutations and copy number alterations. The contribution of other omics and the functional impact of driver genes remain poorly understood. Here we analyze the genomic and epigenomic data from 377 HCC patients to identify driver genes regulating gene expression modules.The integrative approach shows significant advantages over single-platform analysis on identifying cancer drivers. A representative candidate, SFN(stratifin), knockout shows a markedly slower proliferation in HCC cell lines. Systematic integration of driver genes uncovers four HCC subclasses with distinct prognostic implications and molecular characteristics. Additionally, we find loss of enhancer that significantly reduces expression occurs more frequently than gain of enhancer in HCC compared to normal, indicating the important regulation by epigenome in addition to genomic, methylome and transcriptome. Our study represents the most comprehensive discovery of HCC driver genes and also highlights the power of integrating multi-omics data.