CoChat
Sign Up
PreprintCited by 7Protein Structure and DynamicsComputational Drug Discovery MethodsAdvanced Physical and Chemical Molecular Interactions

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

How good is Generative Diffusion Model for Enhanced Sampling of Protein Conformations Across Scales and in All-atom Resolution?

Open access
Bera, Palash, Mondal, Jagannath·2025·The Journal of Chemical Physics
DOI 10.1101/2025.01.16.633470

Preprint — published as 10.1063/5.0279756 · Paper page

Open this paper in CoChat — free account

Abstract

Molecular dynamics (MD) simulations are fundamental for probing the structural dynamics of biomolecules, yet their efficiency is limited by the high computational cost of exploring long-timescale events. Generative machine learning (ML) models, particularly the Denoising Diffusion Probabilistic Model (DDPM), offer an emerging strategy to enhance conformational sampling. In this study, we evaluate the capabilities and limitations of DDPM in generating atomistically accurate conformational ensembles across proteins of varying size and structural order, ranging from the 20-residue folded Trp-cage and 58-residue BPTI to the 83-residue intrinsically disordered region Ash1 and the 140-residue intrinsically disordered protein α -Synuclein. Training DDPM on relatively short MD trajectories using both torsion angle and all-atom coordinate data, we demonstrate that it can reproduce key structural features such as secondary structure, radius of gyration, and contact maps, while effectively sampling sparsely populated regions of the conformational landscape. Notably, DDPM can also generate novel conformations, including transitions not explicitly observed in the training data. However, the model occasionally overlooks low-probability regions and may produce conformers with unclear physical relevance, warranting independent validation. These limitations are particularly evident in flexible systems such as IDPs. Overall, this work benchmarks DDPM as a viable tool for augmenting MD simulations, offering enhanced sampling with significant computational savings, while noting its limitations in capturing low-populated conformers. At the same time, it highlights the importance of rigorous validation and thoughtful interpretation when deploying generative models in computational biophysics.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1101/2025.01.16.633470

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 89

  1. 10.1006/jmbi.1999.3110 · DOI · Paper page
  2. 10.1016/j.febslet.2005.03.072 · DOI · Paper page
  3. 10.1038/s41580-023-00673-0 · DOI
  4. 2023, Conformational plasticity in α-synuclein and how crowded environment modulates it, The Journal of Physical Chemistry B, 127, 4032, 10.1021/acs.jpcb.3c00982 · DOI
  5. 10.1146/annurev.biophys.37.032807.125924 · DOI
  6. 10.1016/j.bbadis.2011.10.002 · DOI
  7. 10.1016/S0896-6273(02)00872-3 · DOI
  8. 10.1021/cr500288y · DOI
  9. 10.1038/42166 · DOI · Paper page
  10. Shaw, D. E. ; Dror, R. O. ; Salmon, J. K. ; Grossman, J. ; Mackenzie, K. M. ; Bank, J. A. ; Young, C. ; Deneroff, M. M. ; Batson, B. ; Bowers, K. J . Millisecond-scale molecular dynamics simulations on Anton. Proceedings of the conference on high performance computing networking, storage and analysis. 2009; pp 1–11. · DOI
  11. Shaw, D. E. ; Grossman, J. ; Bank, J. A. ; Batson, B. ; Butts, J. A. ; Chao, J. C. ; Deneroff, M. M. ; Dror, R. O. ; Even, A. ; Fenton, C. H . Anton 2: raising the bar for performance and programmability in a special-purpose molecular dynamics supercomputer. SC’14: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. 2014; pp 41–53. · DOI
  12. 10.1002/jcc.24030 · DOI
  13. 10.1007/BF00124016 · DOI
  14. 10.1002/wcms.31 · DOI
  15. 10.1016/0021-9991(77)90121-8 · DOI · Paper page
  16. 10.1002/wcms.66 · DOI
  17. 10.1126/science.271.5251.997 · DOI
  18. 10.1016/S0959-440X(00)00194-9 · DOI
  19. 10.1016/j.cplett.2006.05.062 · DOI
  20. 10.1073/pnas.0914540107 · DOI
  21. 2014, On-the-fly learning and sampling of ligand binding by high-throughput molecular simulations, Journal of chemical theory and computation, 10, 2064, 10.1021/ct400919u · DOI
  22. 2019, How effectively can adaptive sampling methods capture spontaneous ligand binding?, Journal of chemical theory and computation, 15, 2053, 10.1021/acs.jctc.8b00913 · DOI
  23. 10.1016/S0009-2614(99)01123-9 · DOI
  24. 10.1016/S0006-3495(03)74897-8 · DOI
  25. 10.1063/1.1755656 · DOI
  26. 10.1016/S1574-1400(09)00504-0 · DOI
  27. 10.1146/annurev-biophys-070816-033834 · DOI
  28. 10.1016/S0006-3495(96)79552-8 · DOI
  29. Mardt, A. ; Pasquali, L. ; Wu, H. ; Noé, F. VAMPnets for deep learning of molecular kinetics. Nature communications 2018, 9, 5. · DOI
  30. Wehmeyer, C. ; No, F . Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics. The Journal of Chemical Physics 2018, 148, 241703. · DOI
  31. Prinz, J.-H. ; Wu, H. ; Sarich, M. ; Keller, B. ; Senne, M. ; Held, M. ; Chodera, J. D. ; Schtte, C. ; No, F . Markov models of molecular kinetics: Generation and validation. The Journal of Chemical Physics 2011, 134, 174105. · DOI · Paper page
  32. 10.1021/jacs.7b12191 · DOI
  33. 10.1073/pnas.1704803114 · DOI
  34. 10.1109/5.18626 · DOI · Paper page
  35. No, F. ; Wu, H. ; Prinz, J.-H. ; Plattner, N . Projected and hidden Markov models for calculating kinetics and metastable states of complex molecules. The Journal of Chemical Physics 2013, 139, 184114. · DOI · Paper page
  36. 10.1038/nbt1004-1315 · DOI
  37. 10.1126/science.1127647 · DOI · Paper page
  38. 10.1016/j.neucom.2013.09.055 · DOI
  39. 10.1093/nar/gkae749 · DOI
  40. Kingma, D. P. ; Welling, M. Auto-Encoding Variational Bayes. 2022; https://arxiv. org/abs/1312.6114.
  41. 10.1561/2200000056 · DOI · Paper page
  42. 10.1016/j.str.2019.03.018 · DOI
  43. 2021, Predicting new protein conformations from molecular dynamics simulation conformational landscapes and machine learning, Proteins: structure, function, and bioinformatics, 89, 915, 10.1002/prot.26068 · DOI
  44. Gupta, A. ; Dey, S. ; Hicks, A. ; Zhou, H.-X . Artificial intelligence guided conformational mining of intrinsically disordered proteins. Communications biology 2022, 5, 610. · DOI · Paper page
  45. Goodfellow, I. ; Pouget-Abadie, J. ; Mirza, M. ; Xu, B. ; Warde-Farley, D. ; Ozair, S. ; Courville, A. ; Bengio, Y. Generative adversarial nets. Advances in neural information processing systems 2014, 27 .
  46. Brock, A. ; Donahue, J. ; Simonyan, K. Large Scale GAN Training for High Fidelity Natural Image Synthesis. 2019; https://arxiv.org/abs/1809.11096. · arXiv
  47. Rezende, D. ; Mohamed, S . Variational inference with normalizing flows. International conference on machine learning. 2015; pp 1530–1538.
  48. 2021, Smooth normalizing flows, Advances in Neural Information Processing Systems, 34, 2796
  49. 2021, Deep generative modelling: A comparative review of vaes, gans, normalizing flows, energy-based and autoregressive models, IEEE transactions on pattern analysis and machine intelligence, 44, 7327
  50. Noé, F. ; Olsson, S. ; Köhler, J. ; Wu, H. Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning. Science 2019, 365, eaaw1147. · DOI
  51. Hochreiter, S. Long Short-term Memory. Neural Computation MIT-Press 1997, · DOI · Paper page
  52. Sundermeyer, M. ; Schlüter, R. ; Ney, H. Lstm neural networks for language modeling. Interspeech. 2012; pp 194–197. · DOI · Paper page
  53. 2020, Learning molecular dynamics with simple language model built upon long short-term memory neural network, Nature communications, 11, 5115, 10.1038/s41467-020-18959-8 · DOI
  54. 2022, Path sampling of recurrent neural networks by incorporating known physics, Nature Communications, 13, 7231, 10.1038/s41467-022-34780-x · DOI
  55. Vaswani, A. ; Shazeer, N. ; Parmar, N. ; Uszkoreit, J. ; Jones, L. ; Gomez, A. N. ; Kaiser, Lt . ; Polosukhin, I . Attention is all you need. Advances in neural information processing systems 2017, 30 .
  56. 2025, Accurate Prediction of the Kinetic Sequence of Physicochemical States Using Generative Artificial Intelligence, Chem. Sci, 16, 8735, 10.1039/D5SC00108K · DOI
  57. Sohl-Dickstein, J. ; Weiss, E. ; Maheswaranathan, N. ; Ganguli, S . Deep unsupervised learning using nonequilibrium thermodynamics. International conference on machine learning. 2015; pp 2256–2265.
  58. 10.48550/ARXIV.2006.11239 · DOI · Paper page
  59. 2021, Diffusion models beat gans on image synthesis, Advances in neural information processing systems, 34, 8780
  60. 2024, Protein structure generation via folding diffusion, Nature communications, 15, 1059, 10.1038/s41467-024-45051-2 · DOI
  61. 2024, Direct conformational sampling from peptide energy landscapes through hypernetwork-conditioned diffusion, Nature Machine Intelligence, 6, 775, 10.1038/s42256-024-00860-4 · DOI
  62. 10.1038/s41586-023-06728-8 · DOI · Paper page
  63. 2023, Score-based generative modeling for de novo protein design, Nature Computational Science, 3, 382, 10.1038/s43588-023-00440-3 · DOI · Paper page
  64. 2024, Predicting equilibrium distributions for molecular systems with deep learning, Nature Machine Intelligence, 6, 558, 10.1038/s42256-024-00837-3 · DOI
  65. Abramson, J. ; Adler, J. ; Dunger, J. ; Evans, R. ; Green, T. ; Pritzel, A. ; Ronneberger, O. ; Willmore, L. ; Ballard, A. J. ; Bambrick, J . Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 2024, 1–3. · DOI · Paper page
  66. 10.1126/science.abj8754 · DOI · Paper page
  67. 10.1038/s41586-023-06415-8 · DOI · Paper page
  68. 10.1146/annurev-physchem-083122-125941 · DOI
  69. Ribeiro, J. M. L. ; Bravo, P. ; Wang, Y. ; Tiwary, P . Reweighted autoencoded variational Bayes for enhanced sampling (RAVE). The Journal of Chemical Physics 2018, 149, 072301. · DOI
  70. Aranganathan, A. ; Gu, X. ; Wang, D. ; Vani, B. P. ; Tiwary, P . Modeling Boltzmann-weighted structural ensembles of proteins using artificial intelligencebased methods. Current Opinion in Structural Biology 2025, 91, 103000. · DOI
  71. Janson, G. ; Valdes-Garcia, G. ; Heo, L. ; Feig, M . Direct generation of protein conformational ensembles via machine learning. Nature Communications 2023, 14, 774. · DOI · Paper page
  72. 2024, Transferable deep generative modeling of intrinsically disordered protein conformations, PLOS Computational Biology, 20, e1012144, 10.1371/journal.pcbi.1012144 · DOI · Paper page
  73. 10.1073/pnas.2203656119 · DOI
  74. 2024, Neural Thermodynamic Integration: Free Energies from Energy-Based Diffusion Models, The Journal of Physical Chemistry Letters, 15, 11395, 10.1021/acs.jpclett.4c01958 · DOI
  75. Benayad, Z. ; Stirnemann, G. Hamiltonian replica exchange augmented with diffusion-based generative models and importance sampling to assess biomolecular conformational basins and barriers. 2025; https://arxiv.org/abs/2505.08357. · DOI
  76. Ronneberger, O. ; Fischer, P. ; Brox, T . U-net: Convolutional networks for biomedical image segmentation. Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18. 2015; pp 234–241. · DOI · Paper page
  77. Kingma, D. P. ; Ba, J. Adam : A Method for Stochastic Optimization. 2017; https://arxiv.org/abs/1412.6980. · arXiv
  78. 10.1073/pnas.1800690115 · DOI
  79. 2021, Deep learning the structural determinants of protein biochemical properties by comparing structural ensembles with DiffNets, Nature communications, 12, 3023, 10.1038/s41467-021-23246-1 · DOI
  80. 10.1073/pnas.2321971121 · DOI
  81. 10.1021/jacs.1c07591 · DOI
  82. 2018, Atomic resolution mechanism of ligand binding to a solvent inaccessible cavity in T4 lysozyme, PLoS computational biology, 14, e1006180, 10.1371/journal.pcbi.1006180 · DOI
  83. 10.1021/jacs.8b10840 · DOI
  84. 10.1021/jacs.3c06144 · DOI
  85. 2020, Fuzzy association of an intrinsically disordered protein with acidic membranes, JACS Au, 1, 66
  86. 10.1073/pnas.2407744121 · DOI
  87. 10.1073/pnas.1418155112 · DOI
  88. Strodel, B . Energy landscapes of protein aggregation and conformation switching in intrinsically disordered proteins. Journal of molecular biology 2021, 433, 167182. · DOI
  89. 2020, Simulation of FUS protein condensates with an adapted coarse-grained model, Journal of chemical theory and computation, 17, 525

Metadata

Authors
Bera, P., & Mondal, J.
Affiliations
Tata Institute of Fundamental Research (IN)
Corresponding author
  • Bera, Palash
  • Mondal, Jagannath
ORCID
Venue
The Journal of Chemical Physics
Publisher
openRxiv
Published
2025-01-21
Volume
163
Issue
11
ISSN
0021-9606 (ISSN-L)
DOI
10.1101/2025.01.16.633470
OpenAlex
W4414279885
Cited by
7
References
1
Citations by year
2026: 5 · 2025: 2
FWCI
2.11
Type
journal article
Language
en
Open access
Licence
Locations
  • The Journal of Chemical Physics (publishedVersion)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc)
  • PubMed (publishedVersion)
Funders
Department of Atomic Energy, Government of India (RTI 4007)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
  • Physical Sciences › Chemistry › Physical and Theoretical Chemistry › Advanced Physical and Chemical Molecular Interactions
Keywords
denoising diffusion probabilistic models, generative models, protein conformational sampling, intrinsically disordered proteins, alpha-synuclein, aprotinin, Trp-cage, ASH1, secondary structure, radius of gyration, contact maps, dihedral angle
MeSH terms
Machine Learning, Diffusion, Protein Conformation, Proteins, alpha-Synuclein, Molecular Dynamics Simulation, Intrinsically Disordered Proteins

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review