Type A γ-aminobutyric acid (GABAA) receptors are pentameric ligand-gated cloride channels and the main drivers of fast inhibitory neurotransmission in the vertebrate nervous system. Their dysfunction is implicated in a range of neurological disorders, including depression, epilepsy and schizophrenia. Boehringer Ingelheim and Saniona have entered into a collaboration with the objective to identify new negative allosteric modulators of the GABAA α5 receptor, starting from in vitro data for ~2000 cmpds. This legacy data set enabled an innovative medicinal chemistry strategy that was driven by computational approaches. We will describe how a combination of machine learning and of a modified Free Wilson approach guided Combinatorial Chemistry and Medicinal Chemistry synthesis campaigns. This new approach allowed a drastic reduction of the number of molecules that are typically prepared at this stage and an acceleration of the DMTA cycles. We quickly identified several “in vivo ready” compounds for rodent PK studies and, out of these, multiple advanced, selective and very potent candidates for a novel, high quality drug for CNS diseases.

Leveraging machine learning and the Free-Wilson approach in lead optimization: Efficient discovery of a new chemical class modulating the GABAA α5 receptor

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.