Improved Estimates of Folding Stabilities and Kinetics from Multiensemble Markov Models

Abstract

In the last few years, Markov State Model (MSM) approaches have been widely applied to understand folding mechanisms and predict long timescale dynamics from ensembles of short molecular simulations. Most MSM estimators enforce detailed balance, assuming that trajectory data is sampled at equilibrium. This is rarely the case for ab initio folding studies, however, and as a result, protein folding stabilities can be severely underestimated. To remedy this problem, we propose an enhanced-sampling protocol in which we (1) perform unbiased folding simulations and use sparse tICA to obtain features that best capture the slowest events in folding, (2) perform umbrella sampling along this reaction coordinate to observe folding and unfolding transitions, and (3) estimate the thermodynamics and kinetics of folding using multiensemble Markov models (MEMMs).