Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration

TL;DR

This paper introduces a machine learning method using autoencoders to discover explicit collective variables for enhanced sampling in molecular dynamics, enabling faster exploration of complex biomolecular energy landscapes.

Contribution

It presents a novel approach combining autoencoders with molecular dynamics to automatically identify and utilize data-driven collective variables for accelerated sampling.

Findings

Substantial speedups in configurational space exploration.

Successful application to alanine dipeptide and Trp-cage.

Open-source implementation available within OpenMM.

Abstract

Macromolecular and biomolecular folding landscapes typically contain high free energy barriers that impede efficient sampling of configurational space by standard molecular dynamics simulation. Biased sampling can artificially drive the simulation along pre-specified collective variables (CVs), but success depends critically on the availability of good CVs associated with the important collective dynamical motions. Nonlinear machine learning techniques can identify such CVs but typically do not furnish an explicit relationship with the atomic coordinates necessary to perform biased sampling. In this work, we employ auto-associative artificial neural networks ("autoencoders") to learn nonlinear CVs that are explicit and differentiable functions of the atomic coordinates. Our approach offers substantial speedups in exploration of configurational space, and is distinguished from exiting approaches by its capacity to simultaneously discover and directly accelerate along data-driven CVs. We demonstrate the approach in simulations of alanine dipeptide and Trp-cage, and have developed an open-source and freely-available implementation within OpenMM.