Ennet: exert enhaner-only somatic mutations to discover potential cancer-driving biological networks

Whole genome sequencing technology has facilitated the discovery of a large number of somatic mutations in enhancers (SMEs), whereas the utility of SMEs in tumorigenesis has not been fully explored. Here we present Ennet, a method to comprehensively investigate SMEs enriched networks (SME-networks) in cancer by integrating SMEs, enhancer-gene interactions and gene-gene interactions. Using Ennet, we performed a pan-cancer analysis in 2004 samples from 8 cancer types and found many well-known cancer drivers were involved in SME-networks, including ESR1, SMAD3, MYC, EGFR, BCL2 and PAX5 et al. Meanwhile, Ennet also identified many new networks with less characterization but have potentially important roles in cancer, including the biggest SME-network in medulloblastoma (MB), which contains genes enriched in the glutamate receptor and nervous development pathways. Interestingly, SME-networks are specific across cancer types, and the vast majority of the genes identified by Ennet have few mutations in gene bodies. Collectively, our work shows that using enhancer-only somatic mutations can be an effective way to discover potential cancer-driving networks. Ennet provides a new perspective to explore new mechanisms for tumor progression from SMEs.