Computational Predictions of Drug-Protein Binding Kinetics with a Hybrid Molecular Dynamics, Brownian Dynamics, and Milestoning Approach

Abstract

Molecular binding events are a fundamental aspect of cellular processes and are therefore of particular interest for the development of novel therapeutics. Simulation based approaches that can estimate binding and unbinding kinetics are desirable, as they can provide critical detail on the molecular level. We present recent developments to our hybrid molecular dynamics, Brownian dynamics, and milestoning software for estimating protein-ligand association and dissociation rates, called SEEKR. This approach reduces the simulation time required to observe binding and unbinding events and also benefits from an embarrassingly parallel implementation. We describe the effectiveness of the approach for rank-ordering ligands by dissociation rates and binding free energies. We also describe improvements focused on making such calculations feasible for larger systems of pharmaceutical relevance. This includes implementation changes to reduce simulation cost and on-the-fly convergence monitoring.