Deep generative model for anticancer drug design: Application for development of novel drug candidates against chronic myeloid leukemia

Abstract

A generative hetero-encoder model for computer-aided design of potential inhibitors of Bcr-Abl tyrosine kinase, the enzyme playing a key role in the pathogenesis of chronic myeloid leukemia, was developed. Training and testing of this model were carried out on a set of chemical compounds containing 2-arylaminopyrimidine, the major pharmacophore present in the structures of many small-molecule inhibitors of protein kinases. The neural network was then used for generating a wide range of new molecules and subsequent analysis of their binding affinity to the target protein using molecular docking tools. As a result, the developed neural network was shown to be a promising mathematical model for de novo design of small-molecule compounds potentially active against Abl kinase, which can be used to develop potent broad-spectrum anticancer drugs