pith. sign in

arxiv: 2502.05909 · v3 · submitted 2025-02-09 · ⚛️ physics.atom-ph · physics.chem-ph· physics.comp-ph

Towards a Universal Foundation Model for Protein Dynamics: A Multi-Chain Tree-Structured Framework with Transformer Propagators

Jinzhen Zhu This is my paper

Pith reviewed 2026-05-23 03:25 UTC · model grok-4.3

classification ⚛️ physics.atom-ph physics.chem-phphysics.comp-ph
keywords protein dynamicscoarse-grained molecular dynamicstransformer propagatortree-structured representationstochastic differential equationsmolecular simulationmulti-chain proteins
0
0 comments X

The pith

A tree-structured representation and Transformer propagators enable a universal coarse-grained model for protein dynamics that runs over 10,000 times faster than all-atom simulations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper aims to create a single framework that can simulate the dynamics of any protein or multi-chain assembly much faster than traditional methods. It does this by first converting atomic positions into a compact tree of collective variables that still allow rebuilding the full structure accurately. Then a Transformer model learns to evolve these variables over time as if solving stochastic equations. If successful, this would let researchers generate long trajectories for many systems quickly, opening the way to study larger or more complex molecular behaviors without prohibitive computing costs.

Core claim

The central claim is that a hierarchical tree-structured coarse-grained representation (TSCG) combined with a Transformer-based propagator for stochastic differential equations provides a universal model for protein dynamics. This representation maps Cartesian coordinates to minimal interpretable collective variables that reconstruct full-atom structures at sub-angstrom precision for arbitrary proteins and multi-chain systems. The model treats collective variables as sequences to generalize across lengths and configurations, achieving 10,000 to 20,000 times acceleration over traditional MD while producing trajectories statistically consistent in RMSD profiles and structural ensembles.

What carries the argument

The tree-structured coarse-grained (TSCG) representation, which encodes protein Cartesian coordinates into a minimal set of collective variables that support both high-precision reconstruction and temporal evolution modeled by Transformer propagators.

If this is right

  • Microsecond-long trajectories can be generated in minutes rather than requiring extensive computational resources.
  • Statistical properties such as RMSD profiles and structural ensembles remain consistent with those from all-atom molecular dynamics.
  • The approach applies to multi-chain assemblies without requiring protein-specific network designs.
  • Collective variables represented as sequences allow handling of arbitrary sequence lengths in a single model.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The framework could potentially extend to simulating dynamics of other macromolecules if the tree structure can be adapted to their topologies.
  • Integration with experimental observables might allow the model to be calibrated directly against data from techniques like cryo-EM or NMR.
  • Such speedups might enable screening of many protein variants or conditions in a high-throughput manner for functional studies.

Load-bearing premise

The tree-structured mapping produces a minimal set of collective variables that fully capture the essential dynamics and allow accurate reconstruction for any protein system.

What would settle it

A test case where the reconstructed atomic structures from the coarse-grained variables show root-mean-square deviations exceeding one angstrom from the original all-atom coordinates, or where the simulated trajectories diverge significantly in their statistical distributions from reference molecular dynamics runs.

Figures

Figures reproduced from arXiv: 2502.05909 by Jinzhen Zhu.

Figure 1
Figure 1. Figure 1: FIG. 1: This illustration depicts coordinate transform (upper) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: RMSD profiles of molecular dynamics simulations [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: The structure reconstruction of a two-chain protein [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: RMSD profiles of molecular dynamics simulations [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: RMSD profiles of molecular dynamics simulations [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
read the original abstract

Simulating large-scale protein dynamics using traditional all-atom molecular dynamics (MD) remains computationally prohibitive. We present a unified, universal framework for coarse-grained molecular dynamics (CG-MD) that achieves high-fidelity structural reconstruction and generalizes across diverse protein systems. Central to our approach is a hierarchical, tree-structured protein representation (TSCG) that maps Cartesian coordinates into a minimal set of interpretable collective variables. We extend this representation to accommodate multi-chain assemblies, demonstrating sub-angstrom precision in reconstructing full-atom structures from coarse-grained nodes. To model temporal evolution, we formulate protein dynamics as stochastic differential equations (SDEs), utilizing a Transformer-based architecture as a universal propagator. By representing collective variables as language-like sequences, our model transcends the limitations of protein-specific networks, generalizing to arbitrary sequence lengths and multi-chain configurations. The framework achieves an acceleration of over 10,000 to 20,000 times compared to traditional MD, generating microsecond-long trajectories within minutes. Our results show that the generated trajectories maintain statistical consistency with all-atom MD in RMSD profiles and structural ensembles. This universal model provides a salable solution for high-throughput protein simulation, offering a significant leap toward a foundation model for molecular dynamics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript proposes a universal coarse-grained MD framework centered on a hierarchical tree-structured coarse-grained (TSCG) representation that maps Cartesian coordinates of proteins (including multi-chain assemblies) to a minimal set of collective variables, with claimed sub-angstrom reconstruction fidelity. Protein dynamics are formulated as SDEs and propagated by a Transformer architecture that treats collective variables as language-like sequences, purportedly enabling generalization across arbitrary sequence lengths and systems. The work claims 10,000–20,000× acceleration relative to all-atom MD, with generated microsecond trajectories produced in minutes while preserving statistical consistency in RMSD profiles and structural ensembles.

Significance. If the central claims of sub-angstrom reconstruction, cross-system generalization, and orders-of-magnitude acceleration with statistical fidelity were substantiated by rigorous validation, the framework would constitute a substantial advance toward scalable, general-purpose simulators for protein dynamics, with clear implications for high-throughput structural biology and drug discovery.

major comments (2)
  1. [Abstract] Abstract: the central claims of 10,000–20,000× acceleration, microsecond trajectories generated in minutes, and statistical consistency with all-atom MD in RMSD and ensembles are stated without any accompanying methods, datasets, error bars, baselines, or quantitative results; no sections, equations, tables, or figures are supplied to support or allow verification of these load-bearing assertions.
  2. [Abstract] Abstract: the assertion that the TSCG representation 'maps Cartesian coordinates into a minimal set of interpretable collective variables' that 'suffice to reconstruct full-atom structures at sub-angstrom precision' and 'capture the essential dynamics for arbitrary proteins' is presented without any derivation, loss function, or reconstruction-error analysis, rendering the weakest assumption untestable.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their review and for identifying areas where the abstract could be more clearly linked to the supporting material in the manuscript. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claims of 10,000–20,000× acceleration, microsecond trajectories generated in minutes, and statistical consistency with all-atom MD in RMSD and ensembles are stated without any accompanying methods, datasets, error bars, baselines, or quantitative results; no sections, equations, tables, or figures are supplied to support or allow verification of these load-bearing assertions.

    Authors: The abstract is written as a concise summary of the principal results. The underlying methods, training and test datasets, error bars, baselines, quantitative acceleration measurements, and statistical comparisons (RMSD profiles and ensemble properties) are provided in full in the Methods and Results sections, supported by the relevant equations, tables, and figures that enable verification of the reported performance. revision: no

  2. Referee: [Abstract] Abstract: the assertion that the TSCG representation 'maps Cartesian coordinates into a minimal set of interpretable collective variables' that 'suffice to reconstruct full-atom structures at sub-angstrom precision' and 'capture the essential dynamics for arbitrary proteins' is presented without any derivation, loss function, or reconstruction-error analysis, rendering the weakest assumption untestable.

    Authors: The TSCG hierarchical mapping, its formulation as a minimal set of collective variables, the loss function employed for reconstruction, and the quantitative reconstruction-error analysis establishing sub-angstrom fidelity are derived and presented in the main text (with multi-chain extensions). Generalization across arbitrary proteins and sequence lengths is demonstrated via cross-system experiments reported in the Results section. revision: no

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

No equations, derivation steps, or self-citations are visible in the supplied abstract. The full manuscript text was not provided, preventing inspection of TSCG construction, SDE formulation, or any claimed predictions. Without load-bearing steps that reduce to inputs by construction, the derivation cannot be assessed as circular; the framework description remains self-contained at the level of available information.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Review performed on abstract alone; ledger entries are inferred from claims rather than explicit statements in the manuscript.

invented entities (1)
  • TSCG (tree-structured coarse-grained) representation no independent evidence
    purpose: Maps Cartesian coordinates to minimal collective variables for reconstruction and dynamics
    Central new construct introduced to enable the universal framework

pith-pipeline@v0.9.0 · 5752 in / 1184 out tokens · 34873 ms · 2026-05-23T03:25:00.100401+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

61 extracted references · 61 canonical work pages · 10 internal anchors

  1. [1]

    merlin.mbs aapmrev4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked

    FUNCTION id.bst "merlin.mbs aapmrev4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked" ENTRY address archive archivePrefix author bookaddress booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key language month note number organization pages primaryClass publisher school SLACcitation series title translat...

    work page 2010

  2. [2]

    merlin.mbs aipauth4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked

    FUNCTION id.bst "merlin.mbs aipauth4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked" ENTRY address archive archivePrefix author bookaddress booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key language month note number organization pages primaryClass publisher school SLACcitation series title translat...

    work page 2010

  3. [3]

    merlin.mbs aipnum4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked

    FUNCTION id.bst "merlin.mbs aipnum4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked" ENTRY address archive archivePrefix author bookaddress booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key language month note number organization pages primaryClass publisher school SLACcitation series title translati...

    work page 2010

  4. [4]

    author author S. A. \ Hollingsworth \ and\ author R. O. \ Dror ,\ title title Molecular Dynamics Simulation for All , \ 10.1016/j.neuron.2018.08.011 journal journal Neuron \ volume 99 ,\ pages 1129--1143 ( year 2018 ) NoStop

    work page doi:10.1016/j.neuron.2018.08.011 2018

  5. [5]

    author author M. I. \ Zimmerman \ and\ author G. R. \ Bowman ,\ 10.1016/bs.mie.2016.05.032 title Methods in Enzymology ,\ edition 1st \ ed.,\ Vol.\ volume 578 \ ( publisher Elsevier Inc. ,\ year 2016 )\ pp.\ pages 213--225 NoStop

    work page doi:10.1016/bs.mie.2016.05.032 2016

  6. [6]

    Bergdorf , author A

    author author M. Bergdorf , author A. Robinson-Mosher , author X. Guo , author K.-H. \ Law , author D. E. \ Shaw , \ and\ author D. E. \ Shaw ,\ title title Desmond/GPU Performance as of April 2021 , \ @noop \ volume 1 ( year 2021 ) NoStop

    work page 2021

  7. [7]

    author author D. E. \ Shaw , author P. J. \ Adams , author A. Azaria , author J. A. \ Bank , author B. Batson , author A. Bell , author M. Bergdorf , author J. Bhatt , author J. Adam Butts , author T. Correi , author R. M. \ Dirks , author R. O. \ Dror , author M. P. \ Eastwoo , author B. Edwards , author A. Even , author P. Feldmann , author M. Fenn , au...

    work page doi:10.1145/3458817.3487397 2021

  8. [8]

    author author R. C. \ Bernardi , author M. C. \ Melo , \ and\ author K. Schulten ,\ title title Enhanced sampling techniques in molecular dynamics simulations of biological systems , \ 10.1016/j.bbagen.2014.10.019 journal journal Biochimica et Biophysica Acta - General Subjects \ volume 1850 ,\ pages 872--877 ( year 2015 ) NoStop

    work page doi:10.1016/j.bbagen.2014.10.019 2014

  9. [9]

    author author W. G. \ Noid ,\ title title Perspective: Coarse-grained models for biomolecular systems , \ 10.1063/1.4818908 journal journal Journal of Chemical Physics \ volume 139 ( year 2013 ),\ 10.1063/1.4818908 NoStop

    work page doi:10.1063/1.4818908 2013

  10. [10]

    Liwo , author C

    author author A. Liwo , author C. Czaplewski , author J. Pillardy , \ and\ author H. A. \ Scheraga ,\ title title Cumulant-based expressions for the multibody terms for the correlation between local and electrostatic interactions in the united-residue force field , \ 10.1063/1.1383989 journal journal Journal of Chemical Physics \ volume 115 ,\ pages 2323-...

    work page doi:10.1063/1.1383989 2001

  11. [11]

    Mim , author H

    author author C. Mim , author H. Cui , author J. A. \ Gawronski-Salerno , author A. Frost , author E. Lyman , author G. A. \ Voth , \ and\ author V. M. \ Unger ,\ title title Structural basis of membrane bending by the N-BAR protein endophilin , \ 10.1016/j.cell.2012.01.048 journal journal Cell \ volume 149 ,\ pages 137--145 ( year 2012 ) NoStop

    work page doi:10.1016/j.cell.2012.01.048 2012

  12. [12]

    author author J. W. \ Chu \ and\ author G. A. \ Voth ,\ title title Allostery of actin filaments: Molecular dynamics simulations and coarse-grained analysis , \ 10.1073/pnas.0503732102 journal journal Proceedings of the National Academy of Sciences of the United States of America \ volume 102 ,\ pages 13111--13116 ( year 2005 ) NoStop

    work page doi:10.1073/pnas.0503732102 2005

  13. [13]

    Yu , author A

    author author A. Yu , author A. J. \ Pak , author P. He , author V. Monje-Galvan , author L. Casalino , author Z. Gaieb , author A. C. \ Dommer , author R. E. \ Amaro , \ and\ author G. A. \ Voth ,\ title title A multiscale coarse-grained model of the SARS-CoV-2 virion , \ 10.1016/j.bpj.2020.10.048 journal journal Biophysical Journal \ volume 120 ,\ pages...

    work page doi:10.1016/j.bpj.2020.10.048 2020

  14. [14]

    author author G. Tóth ,\ title title Effective potentials from complex simulations: a potential-matching algorithm and remarks on coarse-grained potentials , \ 10.1088/0953-8984/19/33/335222 journal journal Journal of Physics: Condensed Matter \ volume 19 ,\ pages 335222 ( year 2007 ) NoStop

    work page doi:10.1088/0953-8984/19/33/335222 2007

  15. [15]

    Li \ and\ author S

    author author W. Li \ and\ author S. Takada ,\ title title Characterizing protein energy landscape by self-learning multiscale simulations: Application to a designed -hairpin , \ 10.1016/j.bpj.2010.08.041 journal journal Biophysical Journal \ volume 99 ,\ pages 3029--3037 ( year 2010 ) NoStop

    work page doi:10.1016/j.bpj.2010.08.041 2010

  16. [16]

    Wang , author Y

    author author D. Wang , author Y. Wang , author J. Chang , author L. Zhang , author H. Wang , \ and\ author W. E ,\ title title Efficient sampling of high-dimensional free energy landscapes using adaptive reinforced dynamics , \ 10.1038/s43588-021-00173-1 journal journal Nature Computational Science \ volume 2 ,\ pages 20--29 ( year 2022 ) ,\ http://arxiv...

    work page doi:10.1038/s43588-021-00173-1 2022

  17. [17]

    Reinforced dynamics for enhanced sampling in large atomic and molecular systems

    author author L. Zhang , author H. Wang , \ and\ author E. Weinan ,\ title title Reinforced dynamics for enhanced sampling in large atomic and molecular systems , \ 10.1063/1.5019675 journal journal Journal of Chemical Physics \ volume 148 ( year 2018 ),\ 10.1063/1.5019675 ,\ http://arxiv.org/abs/1712.03461 arXiv:1712.03461 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1063/1.5019675 2018

  18. [18]

    Klein , author A

    author author L. Klein , author A. Y. K. \ Foong , author T. E. \ Fjelde , author B. Mlodozeniec , author M. Brockschmidt , author S. Nowozin , author F. No \' e , \ and\ author R. Tomioka ,\ title title Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics , \ http://arxiv.org/abs/2302.01170 \ ( year 2023 ) ,\ http...

    work page arXiv 2023

  19. [19]

    Zhu \ and\ author J

    author author J. Zhu \ and\ author J. Ma ,\ title title A unified framework for coarse grained molecular dynamics of proteins , \ https://arxiv.org/abs/2403.17513v1 \ ( year 2024 ) ,\ http://arxiv.org/abs/2403.17513 arXiv:2403.17513 NoStop

    work page arXiv 2024

  20. [20]

    Attention Is All You Need

    author author A. Vaswani , author N. Shazeer , author N. Parmar , author J. Uszkoreit , author L. Jones , author A. N. \ Gomez , author . Kaiser , \ and\ author I. Polosukhin ,\ title title Attention is all you need , \ @noop journal journal Advances in Neural Information Processing Systems \ volume 2017-December ,\ pages 5999--6009 ( year 2017 ) ,\ http:...

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  21. [21]

    Van Der Spoel , author E

    author author D. Van Der Spoel , author E. Lindahl , author B. Hess , author G. Groenhof , author A. E. \ Mark , \ and\ author H. J. \ Berendsen ,\ title title GROMACS: Fast, flexible, and free , \ 10.1002/jcc.20291 journal journal Journal of Computational Chemistry \ volume 26 ,\ pages 1701--1718 ( year 2005 ) NoStop

    work page doi:10.1002/jcc.20291 2005

  22. [22]

    Lindahl , author B

    author author E. Lindahl , author B. Hess , \ and\ author D. van der Spoel ,\ title title GROMACS 3.0: A package for molecular simulation and trajectory analysis , \ 10.1007/S008940100045 journal journal Journal of Molecular Modeling \ volume 7 ,\ pages 306--317 ( year 2001 ) NoStop

    work page doi:10.1007/s008940100045 2001

  23. [23]

    author author H. J. \ Berendsen , author D. van der Spoel , \ and\ author R. van Drunen ,\ title title GROMACS: A message-passing parallel molecular dynamics implementation , \ 10.1016/0010-4655(95)00042-E journal journal Computer Physics Communications \ volume 91 ,\ pages 43--56 ( year 1995 ) NoStop

    work page doi:10.1016/0010-4655(95)00042-e 1995

  24. [24]

    Lu , author J

    author author G. Lu , author J. Qi , author Z. Chen , author X. Xu , author F. Gao , author D. Lin , author W. Qian , author H. Liu , author H. Jiang , author J. Yan , \ and\ author G. F. \ Gao ,\ title title Enterovirus 71 and Coxsackievirus A16 3C Proteases: Binding to Rupintrivir and Their Substrates and Anti-Hand, Foot, and Mouth Disease Virus Drug De...

    work page doi:10.1128/jvi.00787-11 2011

  25. [25]

    author author J. W. \ Neidigh , author R. M. \ Fesinmeyer , \ and\ author N. H. \ Andersen ,\ title title Designing a 20-residue protein , \ 10.1038/nsb798 journal journal Nature Structural Biology \ volume 9 ,\ pages 425--430 ( year 2002 ) NoStop

    work page doi:10.1038/nsb798 2002

  26. [26]

    author author J. Zhu ,\ title title Theoretical investigation of the Freeman resonance in the dissociative ionization of H2+ , \ 10.1103/PhysRevA.103.013113 journal journal Physical Review A \ volume 103 ,\ pages 013113 ( year 2021 ) NoStop

    work page doi:10.1103/physreva.103.013113 2021

  27. [27]

    Zhu \ and\ author A

    author author J. Zhu \ and\ author A. Scrinzi ,\ title title Electron double-emission spectra for helium atoms in intense 400-nm laser pulses , \ 10.1103/PhysRevA.101.063407 journal journal Physical Review A \ volume 101 ,\ pages 063407 ( year 2020 ) NoStop

    work page doi:10.1103/physreva.101.063407 2020

  28. [28]

    author author J. Zhu ,\ title title Quantum simulation of dissociative ionization of H 2 + in full dimensionality with a time-dependent surface-flux method , \ 10.1103/PhysRevA.102.053109 journal journal Physical Review A \ volume 102 ,\ pages 053109 ( year 2020 ) NoStop

    work page doi:10.1103/physreva.102.053109 2020

  29. [29]

    Parsons , author J

    author author J. Parsons , author J. B. \ Holmes , author J. M. \ Rojas , author J. Tsai , \ and\ author C. E. M. \ Strauss ,\ title title Practical Conversion from Torsion Space to Cartesian Space for In Silico Protein Synthesis , \ 10.1002/jcc.20237 journal journal Wiley InterScience \ volume 0211458 ( year 2005 ),\ 10.1002/jcc.20237 NoStop

    work page doi:10.1002/jcc.20237 2005

  30. [30]

    Xu , author Q

    author author G. Xu , author Q. Wang , \ and\ author J. Ma ,\ title title OPUS-Rota4: a gradient-based protein side-chain modeling framework assisted by deep learning-based predictors , \ 10.1093/bib/bbab529 journal journal Briefings in Bioinformatics \ volume 23 ,\ pages 1--10 ( year 2022 ) NoStop

    work page doi:10.1093/bib/bbab529 2022

  31. [31]

    Jumper , author R

    author author J. Jumper , author R. Evans , author A. Pritzel , author T. Green , author M. Figurnov , author O. Ronneberger , author K. Tunyasuvunakool , author R. Bates , author A. Z \' i dek , author A. Potapenko , author A. Bridgland , author C. Meyer , author S. A. \ Kohl , author A. J. \ Ballard , author A. Cowie , author B. Romera-Paredes , author ...

    work page doi:10.1038/s41586-021-03819-2 2021

  32. [32]

    author author M. K. \ Scherer , author B. Trendelkamp-Schroer , author F. Paul , author G. P \' e rez-Hern \' a ndez , author M. Hoffmann , author N. Plattner , author C. Wehmeyer , author J. H. \ Prinz , \ and\ author F. No \' e ,\ title title PyEMMA 2: A Software Package for Estimation, Validation, and Analysis of Markov Models , \ 10.1021/acs.jctc.5b00...

    work page doi:10.1021/acs.jctc.5b00743 2015

  33. [33]

    Zang , author L

    author author T. Zang , author L. Yu , author C. Zhang , \ and\ author J. Ma ,\ title title Parallel continuous simulated tempering and its applications in large-scale molecular simulations , \ 10.1063/1.4890038 journal journal The Journal of Chemical Physics \ volume 141 ,\ pages 044113 ( year 2014 ) NoStop

    work page doi:10.1063/1.4890038 2014

  34. [34]

    Zang , author T

    author author T. Zang , author T. Ma , author Q. Wang , \ and\ author J. Ma ,\ title title Improving low-accuracy protein structures using enhanced sampling techniques , \ 10.1063/1.5027243 journal journal The Journal of chemical physics \ volume 149 ( year 2018 ),\ 10.1063/1.5027243 NoStop

    work page doi:10.1063/1.5027243 2018

  35. [35]

    Enhanced sampling and applications in protein folding in explicit solvent

    author author C. Zhang \ and\ author J. Ma ,\ title title Enhanced sampling and applications in protein folding in explicit solvent , \ 10.1063/1.3435332 journal journal The Journal of Chemical Physics \ volume 132 ,\ pages 244101 ( year 2010 ) ,\ http://arxiv.org/abs/1003.0464 arXiv:1003.0464 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1063/1.3435332 2010

  36. [36]

    o hler , author Y. Chen , author A. Kr \

    author author J. K \" o hler , author Y. Chen , author A. Kr \" a mer , author C. Clementi , \ and\ author F. No \' e ,\ title title Flow-Matching: Efficient Coarse-Graining of Molecular Dynamics without Forces , \ 10.1021/acs.jctc.3c00016 journal journal Journal of Chemical Theory and Computation \ volume 19 ,\ pages 942--952 ( year 2023 ) ,\ http://arxi...

    work page doi:10.1021/acs.jctc.3c00016 2023

  37. [37]

    Machine Learning of coarse-grained Molecular Dynamics Force Fields

    author author J. Wang , author S. Olsson , author C. Wehmeyer , author A. P \' e rez , author N. E. \ Charron , author G. De Fabritiis , author F. No \' e , \ and\ author C. Clementi ,\ title title Machine Learning of Coarse-Grained Molecular Dynamics Force Fields , \ 10.1021/acscentsci.8b00913 journal journal ACS Central Science \ volume 5 ,\ pages 755--...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1021/acscentsci.8b00913 2019

  38. [38]

    author author B. E. \ Husic , author N. E. \ Charron , author D. Lemm , author J. Wang , author A. P \' e rez , author M. Majewski , author A. Kr \" a mer , author Y. Chen , author S. Olsson , author G. De Fabritiis , author F. No \' e , \ and\ author C. Clementi ,\ title title Coarse graining molecular dynamics with graph neural networks , \ 10.1063/5.00...

    work page doi:10.1063/5.0026133 2020

  39. [39]

    author author J. Moult ,\ title title A decade of CASP: Progress, bottlenecks and prognosis in protein structure prediction , \ 10.1016/j.sbi.2005.05.011 journal journal Current Opinion in Structural Biology \ volume 15 ,\ pages 285--289 ( year 2005 ) NoStop

    work page doi:10.1016/j.sbi.2005.05.011 2005

  40. [40]

    Moult , author J

    author author J. Moult , author J. T. \ Pedersen , author R. Judson , \ and\ author K. Fidelis ,\ title title A large‐scale experiment to assess protein structure prediction methods , \ 10.1002/prot.340230303 journal journal Proteins: Structure, Function, and Bioinformatics \ volume 23 ,\ pages ii--iv ( year 1995 ) NoStop

    work page doi:10.1002/prot.340230303 1995

  41. [41]

    Ma , author T

    author author T. Ma , author T. Zang , author Q. Wang , \ and\ author J. Ma ,\ title title Refining protein structures using enhanced sampling techniques with restraints derived from an ensemble-based model , \ 10.1002/PRO.3486 journal journal Protein science : a publication of the Protein Society \ volume 27 ,\ pages 1842--1849 ( year 2018 ) NoStop

    work page doi:10.1002/pro.3486 2018

  42. [42]

    Xu , author Z

    author author G. Xu , author Z. Luo , author R. Zhou , author Q. Wang , \ and\ author J. Ma ,\ title title OPUS-Fold3: a gradient-based protein all-atom folding and docking framework on TensorFlow , \ 10.1093/bib/bbad365 journal journal Briefings in Bioinformatics \ volume 24 ,\ pages 1--8 ( year 2023 ) NoStop

    work page doi:10.1093/bib/bbad365 2023

  43. [43]

    author author R. L. \ Harrison ,\ title title Introduction to Monte Carlo simulation , \ 10.1063/1.3295638 journal journal AIP Conference Proceedings \ volume 1204 ,\ pages 17--21 ( year 2009 ) NoStop

    work page doi:10.1063/1.3295638 2009

  44. [44]

    author author G Bhanot ,\ title title The Metropolis algorithm , \ 10.1088/0034-4885/51/3/003 journal journal Reports on Progress in Physics \ volume 51 ,\ pages 429 ( year 1988 ) NoStop

    work page doi:10.1088/0034-4885/51/3/003 1988

  45. [45]

    Karplus \ and\ author G

    author author M. Karplus \ and\ author G. A. \ Petsko ,\ title title Molecular dynamics simulations in biology , \ 10.1038/347631a0 journal journal Nature \ volume 347 ,\ pages 631--639 ( year 1990 ) NoStop

    work page doi:10.1038/347631a0 1990

  46. [46]

    Bussi \ and\ author M

    author author G. Bussi \ and\ author M. Parrinello ,\ title title Accurate sampling using Langevin dynamics , \ 10.1103/PhysRevE.75.056707 journal journal Physical Review E \ volume 75 ,\ pages 56707 ( year 2007 ) NoStop

    work page doi:10.1103/physreve.75.056707 2007

  47. [47]

    author author E. A. \ Codling , author M. J. \ Plank , \ and\ author S. Benhamou ,\ title title Random walk models in biology , \ 10.1098/rsif.2008.0014 journal journal Journal of the Royal Society Interface \ volume 5 ,\ pages 813--834 ( year 2008 ) NoStop

    work page doi:10.1098/rsif.2008.0014 2008

  48. [48]

    Karmakar , author A

    author author T. Karmakar , author A. R. \ Finney , author M. Salvalaglio , author A. O. \ Yazaydin , \ and\ author C. Perego ,\ title title Non-Equilibrium Modeling of Concentration-Driven processes with Constant Chemical Potential Molecular Dynamics Simulations , \ 10.1021/acs.accounts.2c00811 journal journal Accounts of Chemical Research \ volume 56 ,\...

    work page doi:10.1021/acs.accounts.2c00811 2023

  49. [49]

    author author W. Janke ,\ title title Multicanonical Monte Carlo simulations , \ 10.1016/S0378-4371(98)00014-4 journal journal Physica A: Statistical Mechanics and its Applications \ volume 254 ,\ pages 164--178 ( year 1998 ) NoStop

    work page doi:10.1016/s0378-4371(98)00014-4 1998

  50. [50]

    Chen , author H

    author author H. Chen , author H. Fu , author C. Chipot , author X. Shao , \ and\ author W. Cai ,\ title title Overcoming Free-Energy Barriers with a Seamless Combination of a Biasing Force and a Collective Variable-Independent Boost Potential , \ 10.1021/acs.jctc.1c00103 journal journal Journal of Chemical Theory and Computation \ volume 17 ,\ pages 3886...

    work page doi:10.1021/acs.jctc.1c00103 2021

  51. [51]

    author author A. Slav \' i k ,\ title title Generalized differential equations: Differentiability of solutions with respect to initial conditions and parameters , \ 10.1016/j.jmaa.2013.01.027 journal journal Journal of Mathematical Analysis and Applications \ volume 402 ,\ pages 261--274 ( year 2013 ) NoStop

    work page doi:10.1016/j.jmaa.2013.01.027 2013

  52. [52]

    Generative Modeling by Estimating Gradients of the Data Distribution

    author author Y. Song \ and\ author S. Ermon ,\ title title Generative modeling by estimating gradients of the data distribution , \ @noop journal journal Advances in Neural Information Processing Systems \ volume 32 ( year 2019 ) ,\ http://arxiv.org/abs/1907.05600 arXiv:1907.05600 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  53. [53]

    Score-Based Generative Modeling through Stochastic Differential Equations

    author author Y. Song , author J. Sohl-Dickstein , author D. P. \ Kingma , author A. Kumar , author S. Ermon , \ and\ author B. Poole ,\ title title Score-Based Generative Modeling Through Stochastic Differential Equations , \ @noop journal journal ICLR 2021 - 9th International Conference on Learning Representations \ ,\ pages 1--36 ( year 2021 ) ,\ http:...

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  54. [54]

    Wu \ and\ author S

    author author F. Wu \ and\ author S. Z. \ Li ,\ title title DIFFMD: A Geometric Diffusion Model for Molecular Dynamics Simulations , \ 10.1609/aaai.v37i4.25663 journal journal Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023 \ volume 37 ,\ pages 5321--5329 ( year 2023 ) ,\ http://arxiv.org/abs/2204.08672 arXiv:2204.08672 NoStop

    work page doi:10.1609/aaai.v37i4.25663 2023

  55. [55]

    Petersen , author G

    author author M. Petersen , author G. Roig , \ and\ author R. Covino ,\ title title DynamicsDiffusion: Generating and Rare Event Sampling of Molecular Dynamic Trajectories Using Diffusion Models , \ @noop journal journal NeurIPS 2021 AI for Science Workshop \ ,\ pages 1--19 ( year 2021 ) NoStop

    work page 2021

  56. [56]

    Denoising Diffusion Probabilistic Models

    author author J. Ho , author A. Jain , \ and\ author P. Abbeel ,\ title title Denoising diffusion probabilistic models , \ @noop journal journal Advances in Neural Information Processing Systems \ volume 2020-Decem ,\ pages 1--25 ( year 2020 ) ,\ http://arxiv.org/abs/2006.11239 arXiv:2006.11239 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  57. [57]

    Improved Denoising Diffusion Probabilistic Models

    author author A. Nichol \ and\ author P. Dhariwal ,\ title title Improved Denoising Diffusion Probabilistic Models , \ in\ @noop booktitle Proceedings of Machine Learning Research ,\ Vol.\ volume 139 \ ( year 2021 )\ pp.\ pages 8162--8171 ,\ http://arxiv.org/abs/2102.09672 arXiv:2102.09672 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  58. [58]

    Density estimation using Real NVP

    author author L. Dinh , author J. Sohl-Dickstein , \ and\ author S. Bengio ,\ title title Density estimation using real NVP , \ @noop journal journal 5th International Conference on Learning Representations, ICLR 2017 - Conference Track Proceedings \ ( year 2017 ) ,\ http://arxiv.org/abs/1605.08803 arXiv:1605.08803 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  59. [59]

    Boltzmann Generators -- Sampling Equilibrium States of Many-Body Systems with Deep Learning

    author author F. No \' e , author S. Olsson , author J. K \" o hler , \ and\ author H. Wu ,\ title title Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning , \ 10.1126/science.aaw1147 journal journal Science \ volume 365 ( year 2019 ),\ 10.1126/science.aaw1147 ,\ http://arxiv.org/abs/1812.01729 arXiv:1812.01729 NoStop

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1126/science.aaw1147 2019

  60. [60]

    merlin.mbs apsrev4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked

    FUNCTION id.bst "merlin.mbs apsrev4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked" ENTRY address archive archivePrefix author bookaddress booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key language month note number organization pages primaryClass publisher school SLACcitation series title translati...

    work page 2010

  61. [61]

    merlin.mbs apsrmp4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked

    FUNCTION id.bst "merlin.mbs apsrmp4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked" ENTRY address archive archivePrefix author bookaddress booktitle chapter collaboration doi edition editor eid eprint howpublished institution isbn issn journal key language month note number organization pages primaryClass publisher school SLACcitation series title translati...

    work page 2010