Learning Collective Variables from BioEmu with Time-Lagged Generation

TL;DR

This paper introduces BioEmu-CV, a framework that automatically learns collective variables from BioEmu to improve enhanced sampling in molecular dynamics, focusing on capturing slow dynamics for protein folding studies.

Contribution

It proposes a novel method to learn time-lagged collective variables from BioEmu, enabling better sampling of slow molecular processes without manual CV selection.

Findings

Effective CVs for fast-folding proteins were learned.

Improved free energy estimation with on-the-fly sampling.

Enhanced transition path sampling in molecular dynamics.

Abstract

Molecular dynamics is crucial for understanding molecular systems but its applicability is often limited by the vast timescales of rare events like protein folding. Enhanced sampling techniques overcome this by accelerating the simulation along key reaction pathways, which are defined by collective variables (CVs). However, identifying effective CVs that capture the slow, macroscopic dynamics of a system remains a major bottleneck. This work proposes a novel framework coined BioEmu-CV that learns these essential CVs automatically from BioEmu, a recently proposed foundation model for generating protein equilibrium samples. In particular, we re-purpose BioEmu to learn time-lagged generation conditioned on the learned CV, i.e., predict the distribution of molecular states after a certain amount of time. This training process promotes the CV to encode only the slow, long-term information while disregarding fast, random fluctuations. We validate our learned CV on fast-folding proteins with two key applications: (1) estimating free energy differences using on-the-fly probability enhanced sampling and (2) sampling transition paths with steered molecular dynamics. Our empirical study also serves as a new systematic and comprehensive benchmark for MLCVs on fast-folding proteins larger than Alanine Dipeptide.