Water plays an important role in small-molecule binding to proteins; however, water is difficult to account for in molecular docking.  We incorporated a term into the DOCK3.7 scoring function to account for the energetics of water interacting with the protein receptor binding site using Grid Inhomogeneous Solvation Theory (GIST). By considering the regions of the site that are displaced upon ligand binding, we can calculate the energetic cost/benefit of removing waters at those locations. After retrospective testing, we performed prospective docking screens on a hydrated engineered cavity in the Cytochrome c Peroxidase protein. We screened over 1 million fragment molecules with and without a GIST energetic penalty/reward included in our scoring function. We bought 17 molecules that exhibit differences between these two screens for experimental testing determining crystallographic structures and experimental binding affinities, including molecules that show water-mediated interactions. Beyond our successes in finding new chemotypes at high hit rates that recapitulate crystallographic poses, we observe illuminating failures pointing us to ways to improve our methods. Overall, including the GIST component in docking benefits our scoring performance.

Testing a desolvation term in molecular docking on a model cavity

253rd National Meeting (2017)

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