Learning Biomolecular Motion: The Physics-Informed Machine Learning Paradigm

TL;DR

This paper reviews how physics-informed machine learning integrates data and physical laws to improve biomolecular modeling, addressing long-standing challenges in kinetics, rare events, and free-energy estimation.

Contribution

It surveys recent advances in physics-informed neural networks and operator learning, highlighting their potential to solve the biomolecular closure problem with mechanistic interpretability.

Findings

Enhanced modeling of long-timescale kinetics

Improved estimation of rare events and free energy

Frameworks that combine data-driven inference with physical constraints

Abstract

The convergence of statistical learning and molecular physics is transforming our approach to modeling biomolecular systems. Physics-informed machine learning (PIML) offers a systematic framework that integrates data-driven inference with physical constraints, resulting in models that are accurate, mechanistic, generalizable, and able to extrapolate beyond observed domains. This review surveys recent advances in physics-informed neural networks and operator learning, differentiable molecular simulation, and hybrid physics-ML potentials, with emphasis on long-timescale kinetics, rare events, and free-energy estimation. We frame these approaches as solutions to the "biomolecular closure problem", recovering unresolved interactions beyond classical force fields while preserving thermodynamic consistency and mechanistic interpretability. We examine theoretical foundations, tools and frameworks, computational trade-offs, and unresolved issues, including model expressiveness and stability. We outline prospective research avenues at the intersection of machine learning, statistical physics, and computational chemistry, contending that future advancements will depend on mechanistic inductive biases, and integrated differentiable physical learning frameworks for biomolecular simulation and discovery.