Predicting Ligand Binding Kinetics Using a Markovian Milestoning with Voronoi Tessellations Multiscale Approach
Preprints are manuscripts made publicly available before they have been submitted for formal peer review and publication. They might contain new research findings or data. Preprints can be a draft or final version of an author's research but must not have been accepted for publication at the time of submission.
Accurate and efficient computational predictions of ligand binding kinetics can be useful to inform drug discovery campaigns, particularly in the screening and lead optimization phases. Simulation Enabled Estimation of Kinetic Rates, SEEKR, is a multiscale molecular dynamics, Brownian dynamics, and milestoning simulation approach for calculating receptor-ligand association and dissociation rates. Here we present the implementation of a Markovian milestoning with Voronoi tessellations approach that significantly reduces the simulation cost of calculations as well as further improving their parallelizability. The new approach is applied to a host-guest system to assess its effectiveness for rank-ordering compounds by kinetic rates and to the model protein system, trypsin, with the noncovalent inhibitor benzamidine. For both applications, we demonstrate that the new approach requires up to a factor of 10 less simulation time to achieve results with comparable or increased accuracy.