Combining AI-driven virtual screening with biophysics to identify novel BRPF1b inhibitors

At the forefront of the hit identification paradigm we report the development of a new hit identification workflow combining an AI-driven virtual screen with cutting-edge biophysical techniques. Its application to the BRPF1b bromodomain enabled the rapid identification of micromolar binders with favourable in vitro ADMET parameters, from a virtual collection of nearly 25 million commercial compounds. Besides the excellent hit rate demonstrated with BRPF1b, our workflow provides full binding kinetics and predicts ligand binding topologies without the need to acquire new ligand-protein structures,hence not only accelerating hit identification campaigns, but also all subsequent stages of the drug discovery process.