Characterizing metastable states with the help of machine learning

TL;DR

This paper introduces a machine learning approach combined with the variational conformation dynamics method to efficiently identify and characterize metastable states in complex atomistic simulation data, applicable to both biased and unbiased simulations.

Contribution

It presents a novel integration of variational dynamics and machine learning for rapid, hierarchical analysis of metastable states in molecular simulations, improving over traditional methods.

Findings

Successfully applied to protein systems Chignolin and Bovine Pancreatic Trypsin Inhibitor

Can analyze data in seconds, significantly faster than existing methods

Applicable to both biased and unbiased simulation data

Abstract

Present-day atomistic simulations generate long trajectories of ever more complex systems. Analyzing these data, discovering metastable states, and uncovering their nature is becoming increasingly challenging. In this paper, we first use the variational approach to conformation dynamics to discover the slowest dynamical modes of the simulations. This allows the different metastable states of the system to be located and organized hierarchically. The physical descriptors that characterize metastable states are discovered by means of a machine learning method. We show in the cases of two proteins, Chignolin and Bovine Pancreatic Trypsin Inhibitor, how such analysis can be effortlessly performed in a matter of seconds. Another strength of our approach is that it can be applied to the analysis of both unbiased and biased simulations.