Learning Structure, Energy, and Dynamics: A Survey of Artificial Intelligence for Protein Dynamics
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 68 of its 152 references that are in the map. First posted on arXiv 28 Apr 2026.
Builds on
68 of 152 references.
References · 68
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2016Density estimation using Real NVPcited by 2,042
- 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 708
- 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 182
- 2023Equivariant flow matchingcited by 135
- 2022Flow Annealed Importance Sampling Bootstrapcited by 127
- 2024Iterated Denoising Energy Matching for Sampling from Boltzmann Densitiescited by 116
- 2024AlphaFold Meets Flow Matching for Generating Protein Ensemblescited by 110
- 2024Normalizing Flows are Capable Generative Modelscited by 91
- 2022SPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentialscited by 82
- 2023Two for One: Diffusion Models and Force Fields for Coarse-Grained Molecular Dynamicscited by 72
- 2018Machine Learning of coarse-grained Molecular Dynamics Force Fieldscited by 69
38 more
- 2023Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamicscited by 66
- 2020Coarse Graining Molecular Dynamics with Graph Neural Networkscited by 56
- 2021Smooth Normalizing Flowscited by 53
- 2024Generative Modeling of Molecular Dynamics Trajectoriescited by 51
- 2024Transferable Boltzmann Generatorscited by 46
- 2024Protein Conformation Generation via Force-Guided SE(3) Diffusion Modelscited by 45
- 2025Scalable Equilibrium Sampling with Sequential Boltzmann Generatorscited by 38
- 2022Machine Learning Coarse-Grained Potentials of Protein Thermodynamicscited by 37
- 2022Flow-matching -- efficient coarse-graining of molecular dynamics without forcescited by 36
- 2023Str2Str: A Score-based Framework for Zero-shot Protein Conformation Samplingcited by 29
- 2025Progressive Inference-Time Annealing of Diffusion Models for Sampling from Boltzmann Densitiescited by 26
- 2017Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape explorationcited by 24
- 2023Navigating protein landscapes with a machine-learned transferable coarse-grained modelcited by 23
- 2025Consistent Sampling and Simulation: Molecular Dynamics with Energy-Based Diffusion Modelscited by 23
- 2024Structure Language Models for Protein Conformation Generationcited by 17
- 2024EquiJump: Protein Dynamics Simulation via SO(3)-Equivariant Stochastic Interpolantscited by 16
- 2025SimpleFold: Folding Proteins is Simpler than You Thinkcited by 16
- 2019DeepDriveMD: Deep-Learning Driven Adaptive Molecular Simulations for Protein Foldingcited by 15
- 2024BNEM: A Boltzmann Sampler Based on Bootstrapped Noised Energy Matchingcited by 15
- 2025Temperature-Annealed Boltzmann Generatorscited by 15
- 2026Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Computecited by 15
- 2021Chasing Collective Variables using Autoencoders and biased trajectoriescited by 13
- 2025Amortized Sampling with Transferable Normalizing Flowscited by 12
- 2018Automated design of collective variables using supervised machine learningcited by 10
- 2024Scalable Normalizing Flows Enable Boltzmann Generators for Macromoleculescited by 10
- 2024F3low: Frame-to-Frame Coarse-grained Molecular Dynamics with SE(3) Guided Flow Matchingcited by 9
- 2024Dynamic PDB: A New Dataset and a SE(3) Model Extension by Integrating Dynamic Behaviors and Physical Properties in Protein Structurescited by 9
- 2025Enhancing Diffusion-Based Sampling with Molecular Collective Variablescited by 9
- 2025Learning Boltzmann Generators via Constrained Mass Transportcited by 9
- 2023Differentiable Simulations for Enhanced Sampling of Rare Eventscited by 8
- 2025ConfRover: Simultaneous Modeling of Protein Conformation and Dynamics via Autoregressioncited by 8
- 2025Efficient Regression-Based Training of Normalizing Flows for Boltzmann Generatorscited by 8
- 2022Contrastive Learning of Coarse-Grained Force Fieldscited by 7
- 2025UniSim: A Unified Simulator for Time-Coarsened Dynamics of Biomoleculescited by 7
- 2025MD-LLM-1: A Large Language Model for Molecular Dynamicscited by 7
- 2025Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Samplingcited by 7
- 2026Learning Hamiltonian Flow Maps: Mean Flow Consistency for Large-Timestep Molecular Dynamicscited by 7
- 2024Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolationcited by 6
50 most cited of 68 All 68 (JSON)
Built on by
Citing papers per year: 2026: 1.
1 paper.