Proxima

“Molecular glue” compounds and other proximity modulators, such as PROTACs, are potentially powerful medicines because they can induce or block the formation of the protein complexes that act as signals directing many cellular functions (or malfunc­tions). To date, however, there’s been no systematic way to identify or design such proximity-based ther­a­peu­tics. Proxima is changing that, using a vast, proprietary atlas of protein cross-linking data to train state-of-the-art machine learning models. These models can simul­ta­ne­ously simulate how proteins co-fold and design small molecules that can slip between them and change their functions.