Network modeling of kinase inhibitor polypharmacology reveals pathways targeted in chemical screens

摘要: Small molecule screens are widely used to prioritize pharmaceutical development. However, determining the pathways targeted by these molecules is challenging, since compounds often promiscuous. We present a network strategy that takes into account polypharmacology of small in order generate hypotheses for their broader mode action. report screen kinase inhibitors increase efficacy gemcitabine, first-line chemotherapy pancreatic cancer. Eight emerge known affect 201 kinases, which only three kinases have been previously identified as modifiers gemcitabine toxicity. In this work, we use SAMNet algorithm identify linking and genetic toxicity with transcriptional epigenetic changes induced measure using DNaseI-seq RNA-seq. uses constrained optimization connect genes from complementary datasets through set protein-protein protein-DNA interactions. The resulting recapitulates including DNA repair, cell proliferation epithelial-to-mesenchymal transition. predict important roles response, six already shown modify cancer ten novel candidates. Our work reveals role activity chemosensitizing agents.