Ultra-High throughput virtual screening using active learning

With the size of commercial libraries now reaching several billions of compounds, screening vast chemical spaces is becoming an attainable prospect. We have developed a new Hit identification platform underpinned by
cutting edge Machine Learning (MolPAL), structure-enabled virtual screening and biophysical technology (GCI) aiming to maximise this ever expanding commercially available chemical universe. Our platform not only provides
rapid access to structurally diverse and high quality Hits, it also produces valuable information on Hit binding kinetics and topologies without the need to acquire expensive X-ray, NMR or Cryo-EM structures. We exemplify this
technology within Concept Life Sciences for the identification of novel BRPF1b inhibitors. Our platform identified 22 low µM affinity BRPF1b binders, from 51 virtual hits; a 43% hit rate. Further characterisation of selected Hits
provided new start points for BPRF1b inhibition and exemplifies the impact our platform can have on Hit ID campaigns

Please fill in this form to access the full resource