Long-timescale predictions from short-trajectory data: A benchmark analysis of the trp-cage miniprotein

TL;DR

This paper introduces an improved dynamical Galerkin approximation method for analyzing molecular dynamics, enabling accurate long-timescale predictions from short trajectories, validated on the trp-cage miniprotein.

Contribution

The authors reformulate the dynamical Galerkin approximation, reducing lag time dependence and providing new estimators and basis construction methods for molecular kinetics analysis.

Findings

Validated on trp-cage miniprotein folding/unfolding data

Demonstrated accurate long-timescale predictions from short trajectories

Provided a comprehensive strategy for reaction pathway characterization

Abstract

Elucidating physical mechanisms with statistical confidence from molecular dynamics simulations can be challenging owing to the many degrees of freedom that contribute to collective motions. To address this issue, we recently introduced a dynamical Galerkin approximation (DGA) [Thiede et al. J. Phys. Chem. 150, 244111 (2019)], in which chemical kinetic statistics that satisfy equations of dynamical operators are represented by a basis expansion. Here, we reformulate this approach, clarifying (and reducing) the dependence on the choice of lag time. We present a new projection of the reactive current onto collective variables and provide improved estimators for rates and committors. We also present simple procedures for constructing suitable smoothly varying basis functions from arbitrary molecular features. To evaluate estimators and basis sets numerically, we generate and carefully validate a dataset of short trajectories for the unfolding and folding of the trp-cage miniprotein, a well-studied system. Our analysis demonstrates a comprehensive strategy for characterizing reaction pathways quantitatively.