Learning Structure, Energy, and Dynamics: A Survey of Artificial Intelligence for Protein Dynamics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tang, Haocheng, Shi, Liang, Zhang, Ya-Shi, Liu, Xixian, Tang, Jian, Lu, Jiarui
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915962620477440
author Tang, Haocheng
Shi, Liang
Zhang, Ya-Shi
Liu, Xixian
Tang, Jian
Lu, Jiarui
author_facet Tang, Haocheng
Shi, Liang
Zhang, Ya-Shi
Liu, Xixian
Tang, Jian
Lu, Jiarui
contents Protein dynamics underlie many biological functions, yet remain difficult to characterize due to the high computational cost of molecular dynamics simulations and the scarcity of dynamic structural data. This survey reviews recent advances in artificial intelligence for protein dynamics from three perspectives: learning from structural ensembles and trajectories, learning from physical energy signals, and learning to accelerate molecular simulations. We summarize representative methods for conformation ensemble generation, trajectory generation, Boltzmann generators, physics-aware adaptation, machine learning potentials, coarse-grained modeling, and collective variable discovery. We further discuss available datasets and key open challenges, such as scalability, thermodynamic consistency, kinetic fidelity, and integration with experimental constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25244
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning Structure, Energy, and Dynamics: A Survey of Artificial Intelligence for Protein Dynamics
Tang, Haocheng
Shi, Liang
Zhang, Ya-Shi
Liu, Xixian
Tang, Jian
Lu, Jiarui
Biomolecules
Machine Learning
Protein dynamics underlie many biological functions, yet remain difficult to characterize due to the high computational cost of molecular dynamics simulations and the scarcity of dynamic structural data. This survey reviews recent advances in artificial intelligence for protein dynamics from three perspectives: learning from structural ensembles and trajectories, learning from physical energy signals, and learning to accelerate molecular simulations. We summarize representative methods for conformation ensemble generation, trajectory generation, Boltzmann generators, physics-aware adaptation, machine learning potentials, coarse-grained modeling, and collective variable discovery. We further discuss available datasets and key open challenges, such as scalability, thermodynamic consistency, kinetic fidelity, and integration with experimental constraints.
title Learning Structure, Energy, and Dynamics: A Survey of Artificial Intelligence for Protein Dynamics
topic Biomolecules
Machine Learning
url https://arxiv.org/abs/2604.25244