Pith. sign in

REVIEW 1 major objections 6 minor 102 references

The influence of multi-dimensionality and off-diagonal non-Markovian friction coupling on coarse-grained dynamics

T0 review · 1 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read For pentane's two coupled dihedrals, a 2D generalized Langevin equation beats 1D models mainly because the 2D potential confines both angles, not because of off-diagonal friction.

desk verdict Clean ablation shows the 2D potential, not off-diagonal friction, drives pentane MFPT improvement—but the Gaussian noise ansatz is the load-bearing premise. read the letter →

arxiv 2506.05966 v1 pith:YEEXTCAZ submitted 2025-06-06 physics.chem-ph cond-mat.stat-mech

Henrik Kiefer , Cihan Ayaz , Benjamin A. Dalton , Roland R. Netz This is my paper
classification physics.chem-phcond-mat.stat-mech
keywords generalizedLangevinequationmulti-dimensionalreactioncoordinatesnon-Markovianmemoryfrictioncouplingmatrixdihedralisomerizationdynamicsmeanfirst-passagetimeMarkovianembeddingcoarse-grained
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks when a coarse-grained model must keep two reaction coordinates together rather than treating each coordinate with its own one-dimensional generalized Langevin equation. For pentane in water, a 2D GLE with a matrix memory kernel reproduces the molecular-dynamics mean first-passage times much better than 1D GLEs, and the paper shows the improvement comes primarily from the coupled 2D potential, which sterically forbids combining the two cis states. Off-diagonal friction is real and negative, but its main visible effect is to reproduce cross-correlations between the dihedral angles, not the barrier-crossing rates. For alanine dipeptide, whose 2D potential is nearly separable, the 2D GLE offers no rate improvement over 1D GLEs. The methodological result is a tractable recipe: extract the full memory matrix from trajectories, fit it by matrix exponentials, and simulate the Markovian embedding.

What carries the argument

The central object is the memory kernel matrix $\hat{\Gamma}(t)$ in the multi-dimensional GLE, whose off-diagonal entries couple the frictional forces on different reaction coordinates. The paper extracts it from MD trajectories by iteratively solving a Volterra equation for the running integral $\hat{G}(t)$, fits it as a sum of matrix exponentials, and simulates the equivalent Markovian embedding. The decisive comparison isolates the influence of the multi-dimensional potential $U(\theta_1,\theta_2)$ and the off-diagonal entries $\Gamma_{12}$ and $M_{12}$ by switching each off separately in the simulations.

What would settle it

Run the 2D GLE simulation of pentane with the fitted matrix memory and mass but with the potential replaced by $U_1(\theta_1)+U_2(\theta_2)$: the paper predicts the MFPTs should remain close to the 1D-GLE values, so agreement with MD in that run would refute the claim that the coupled potential is the key factor.

Watch

Extended reading notes

Core claim

The paper's central claim is that for the two coupled dihedral angles of pentane in water, the full two-dimensional GLE with a matrix memory kernel reproduces the MD mean first-passage times significantly better than uncoupled 1D GLE models, and that this improvement is dominated by the coupled 2D potential of mean force rather than by the off-diagonal friction entries. The off-diagonal friction is nonetheless substantial, about 18% of the diagonal value and negative for pentane, and it is necessary to reproduce cross-correlations between the two dihedral coordinates. For alanine dipeptide, the off-diagonal friction is positive and roughly a third of the diagonal value, yet the 2D GLE gives no MFPT improvement over 1D GLEs because its 2D potential is nearly separable. The paper also shows that freezing pentane's inner carbons removes the off-diagonal friction, indicating an intramolecular origin for that coupling.

Load-bearing premise

The whole simulation protocol hinges on assuming that a Gaussian random force with variance $k_BT$ times the fitted memory kernel reproduces the actual random force acting on these nonlinear dihedral coordinates, and the paper notes this relation generally fails for nonlinear observables.

Editorial extensions

If this is right

  • For pentane, replacing the 2D potential by the decoupled sum $U_1+U_2$ removes most of the 2D-over-1D MFPT improvement, so coupled potential confinement is the controlling term for this molecule.
  • Off-diagonal friction and mass must be retained to reproduce the mean cross-displacement between the two coordinates; a diagonal-only 2D GLE fails at short times.
  • For near-separable 2D potentials such as alanine dipeptide, independent 1D GLEs are sufficient for MFPTs, and a 2D GLE is not advantageous for rates.
  • Multi-exponential matrix fits are sufficient for MFPT prediction; adding exponential-oscillatory terms to the memory kernel does not improve the rates.
  • Freezing the inner carbons of pentane makes off-diagonal friction negligible, indicating that the coupling arises from intramolecular motion rather than hydrodynamics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial extension: the sign and magnitude of the off-diagonal friction matrix could serve as a mechanistic fingerprint, with negative coupling tied to intramolecular motion (it vanishes when the inner carbons are frozen) and positive coupling in alanine dipeptide attributed to solvent-mediated hydrodynamic interactions; varying solvent viscosity should alter mainly the positive component.
  • Editorial extension: the paper's decomposition implies a cheap screening test for whether a multi-dimensional GLE is needed for rates: compute the difference between the 2D potential of mean force and the sum of the 1D potentials; if the difference is concentrated at sterically forbidden states, rates require the 2D potential, whereas strong off-diagonal friction alone will not.
  • Editorial extension: the failure of the averaged 1D coordinate suggests that dimensionality reduction can push non-Gaussianity into the random force; a testable follow-up is to measure the orthogonal-force non-Gaussianity for such 1D projections and check whether it predicts the MFPT error seen in the paper.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 6 minor

Summary. The paper develops a multi-dimensional generalized Langevin equation (GLE) framework with a matrix memory kernel and a Markovian embedding scheme, and applies it to the two-dimensional dihedral dynamics of solvated pentane and alanine dipeptide. The authors extract the memory matrix from MD trajectories via a Volterra equation, fit it to a sum of matrix exponentials, and simulate the resulting GLE with off-diagonal friction and mass coupling. For pentane, they find that a full 2D GLE reproduces MD mean first-passage times (MFPTs) substantially better than a 1D GLE, and an ablation (Fig. 3d) indicates that the dominant factor is the multi-dimensional coupled potential rather than the off-diagonal friction or mass. For alanine dipeptide, the 2D GLE provides no MFPT advantage over separate 1D GLEs, although off-diagonal friction is still needed to reproduce cross-displacement correlations. The paper includes extensive validation: a model-system test of the extraction (Appendix B), tests of position-dependent mass (Appendix F), exponential-oscillatory memory fits (Appendix K), alternative 1D reaction coordinates (Appendix O), and a frozen-carbon freeze analysis (Appendix P).

Significance. If the central claim holds, the paper makes a useful contribution by delineating when multi-dimensional memory effects are kinetically relevant: for pentane, the off-diagonal friction is significant in magnitude but is not the main cause of the 2D-over-1D MFPT improvement, whereas the coupled potential is. For alanine dipeptide, the 2D coupling is irrelevant for individual MFPTs but essential for cross-correlations. The paper is careful and transparent: it ships a validated extraction scheme (Appendix B), tests several modeling choices (Appendices F, K, O, P), and compares out-of-sample MFPT predictions against MD. The main limitation, acknowledged by the authors, is that the Markovian embedding uses a Gaussian random force with the approximate fluctuation–dissipation relation of Eq. (8), which is generally not exact for non-linear observables. This limits but does not remove the value of the numerical conclusions, provided the Gaussian ansatz is adequate for the specific observables studied; the paper does not test this directly.

major comments (1)
  1. [Section 3.2, Eq. (8), Fig. 3(d)] The central attribution for pentane—that the multi-dimensional coupled potential, not off-diagonal friction, is the key factor reproducing MD MFPTs—rests entirely on Markovian embedding simulations whose noise is Gaussian with covariance given by the approximate fluctuation–dissipation relation in Eq. (8). The paper itself notes in Section 3.2 that Eq. (8) 'is generally not fulfilled for non-linear observables' and Appendix O demonstrates a concrete failure for the averaged dihedral coordinate x_avg. Because the 2D potential changes which configurations are sampled (it eliminates the cis-cis states), the orange and blue legs of the ablation in Fig. 3(d) are subject to potentially different noise-model errors. The observed slowdown in the orange curve could therefore be partly an artifact of the Gaussian ansatz being less inappropriate in the allowed region, rather than a genuine kinetic effect of the 2D confinement. I request a direct test of the validity of Eq. (8) for the 2D dihedral observables, for example by comparing the MD-derived random-force distribution with the Gaussian prediction, or by repeating the ablation with a non-Gaussian noise model from Ref. [69]. Without such a test, the quantitative MFPT agreement and the ranking in Fig. 3(d) could reflect error cancellation rather than the claimed physics.
minor comments (6)
  1. [Abstract] The phrase 'Unlike previous studies, our results highlight the critical role of different terms in the multi-dimensional GLE' is vague; it would be clearer to specify which terms (coupled potential versus off-diagonal friction) matter for which observables.
  2. [Section 3.1, Eq. (17)] The text in Section 3.2 states that the off-diagonal mass entry is 'a factor of around -0.05' of the diagonal entry, but the values in Eq. (17) give M12/M11 = -9.69e-7 / 2.56e-5 ≈ -0.038. Please correct the wording or clarify the rounding.
  3. [Appendix C, Eq. (C6)] The notation in Eq. (C6) is unclear: the left side ⟨y_kl_i(0) y_kn_i(0)⟩ uses three indices on y, and the right side is a double Kronecker delta with indices k,l,n, which seems inconsistent with the preceding definitions. Please rewrite this equation in terms of vector components, e.g., ⟨y_{i,k}(0) y_{i,l}(0)⟩.
  4. [Table 1 and Table 2] The off-diagonal memory times τ12_i are all reported as 0.001 ps, the smallest value in the tables. Please state whether this is a fitted value, a lower bound imposed by the optimizer, or a fixed parameter, and if fixed, whether the results are sensitive to that choice.
  5. [Figure 3 caption] The symbols '□k⁄=l' in the figure caption are rendered incorrectly; they should be Γ_{k≠l} (and M_{k≠l}). This appears to be a font/rendering issue, but the caption should be legible in the final version.
  6. [Data Availability Statement] The statement that code and input files are 'available from the corresponding author upon request' is less reproducible than depositing them in a public repository (e.g., Zenodo or GitHub). Given the paper's emphasis on validation, a permanent DOI would be preferable.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: memory kernels are inverted from MD correlation functions, MFPTs are simulated out-of-sample, and the main attribution to the 2D potential is an ablation; the Gaussian-FDR approximation is an acknowledged limitation, not a circular reduction.

full rationale

The derivation chain is self-contained rather than circular. The memory kernel matrix is obtained by solving the Volterra-type inversion in Eq. (10), derived in Appendix A from the GLE identity, and Appendix B validates the extraction against an independently generated model system with a known analytical kernel. The MFPT comparisons are genuine out-of-sample predictions: the friction and memory parameters are fitted to correlation functions and memory-kernel data, while mean first-passage times are produced by the Markovian embedding simulations and are not used as fit targets. The central pentane attribution to the multi-dimensional coupled potential rests on an ablation (Fig. 3d) performed within the same simulation model, so it does not reduce to the input by construction; the 2D potential is an equilibrium-derived input, but the MFPT slowdown it produces is a nontrivial dynamical consequence. Self-citations appear in the methodology (e.g., Refs. [18,28,67,68,69]), but the embedding equivalence is proved in Appendix C, the extraction is tested in Appendix B, and no load-bearing claim relies solely on a self-citation. The reviewer-flagged limitation is real and should be weighed: Section 3.2 explicitly states that Markovian embedding 'is based on the approximate relation in equation (8), which is generally not fulfilled for non-linear observables,' Section 5 repeats that the Gaussian random force is approximate, and Appendix O shows a related 1D Gaussian GLE fails for averaged and difference coordinates. These passages are limitations on model accuracy and on how strongly the pentane potential attribution can be interpreted, but they are not circular reductions: the fitted memory kernels, the Gaussian noise ansatz, and the MD MFPT targets remain distinct inputs and outputs. The score of 2 reflects minor, non-load-bearing self-citation and the acknowledged approximation caveat, not a finding of circularity.

Assumptions & free parameters 5 free parameters · 7 assumptions · 0 invented entities

The model itself is standard Mori-Zwanzig machinery; the paper contributes measurement and comparison, not new postulates. The burden comes from its own fitting parameters: roughly 50-60 memory-kernel parameters across the two systems (plus hand-set m = 5), all fitted to the data whose correlations are then partly re-predicted. No invented physical entities are introduced; the auxiliary embedding coordinates are standard bookkeeping.

free parameters (5)
  • Pentane 2D memory kernel fit parameters (gamma11_i, gamma12_i, tau11_i, tau12_i, i=1..5) = Table 1; totals gamma11 = 4.35e-4, gamma12 = -0.80e-4 in u nm^2 deg^-2 ps^-1
    Fitted to the extracted 2D memory kernel and its running integral; 20 parameters with symmetry constraints gamma11_i = gamma22_i, gamma12_i = gamma21_i, tau11_i = tau22_i, tau12_i = tau21_i. These parameters drive the 2D GLE simulations.
  • Alanine dipeptide 2D memory kernel fit parameters (gamma11_i, gamma22_i, gamma12_i, tau11_i, tau22_i, tau12_i, i=1..5) = Table 2; totals gamma11 = 4.67e-3, gamma22 = 5.66e-3, gamma12 = 1.35e-3 in u nm^2 deg^-2 ps^-1
    Fitted to the extracted kernel; 30 parameters. Off-diagonal memory times are pinned near 0.001 ps.
  • 1D memory kernel fit parameters for the 1D GLE baselines = Tables 3-6
    Fitted to the 1D-extracted kernels (Eq 15 or Eq K1) for the comparison simulations.
  • Number of exponential components m = 5 = m = 5
    Chosen by hand for all fits as the memory kernel Ansatz in Eq 13; not determined by a selection criterion.
  • Mass matrix entries (Eq 17 and Eq 19) = Pentane: M11 = 2.56e-5, M12 = -9.69e-7; alanine dipeptide: M11 = 1.28e-5, M12 = 4.31e-6, M22 = 1.00e-5 in u nm^2 deg^-2
    Extracted from velocity-velocity correlation averages (Eq 6), so they are measured model inputs rather than ad hoc choices, but they parameterize the simulated dynamics.
assumptions (7)
  • domain assumption The coarse-grained dynamics obey the hybrid-projection GLE (Eqs 2 and 7) with random force orthogonal to initial velocities, <F_R(t) ẋ^T(0)> = 0
    Adopted from Refs [28,29] and used to derive the Volterra extraction equation (A1); not re-derived in this paper.
  • domain assumption The fluctuation-dissipation relation <F_R(t) F_R^T(0)> = k_B T Gamma(t) (Eq 8) holds for the non-linear dihedral observables
    Underpins the Markovian embedding and the Gaussian-noise GLE simulations; the authors state in Section 3.2 that it is generally not fulfilled for non-linear observables, citing Refs [28,29,69].
  • ad hoc to paper The memory kernel is a sum of m = 5 matrix exponentials (Eq 13)
    Assumed functional form for fitting and embedding; neglects the observed oscillations in Gamma(t), acknowledged in Section 3.1. The 1D oscillatory variant (Eq K1) is tested and does not change MFPTs.
  • domain assumption Auxiliary variables follow overdamped dynamics with Gaussian delta-correlated noise and Gaussian initial conditions (Eqs 11-12 and C5)
    Standard Markovian embedding machinery required for the equivalence between Eqs 11 and 7.
  • domain assumption Correlation functions from finite MD trajectories (1 microsecond pentane, 250 ns alanine dipeptide) have converged sufficiently
    Small long-time deviations between G12 and G21 in appendix J are attributed by the authors to finite simulation length.
  • standard math Mass matrices, friction matrices gamma_i and memory time matrices tau_i are invertible, with off-diagonal friction magnitudes constrained below diagonal ones
    Invertibility is required for the embedding (appendix C) and for the fitting constraint that ensures well-defined matrices (appendix H).
  • standard math Trapezoidal-rule discretization of the Volterra equation (A7) gives a stable iterative extraction (Eq 10)
    Numerical discretization choice; validated against a known model kernel in appendix B.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The influence of multi-dimensionality and off-diagonal non-Markovian friction coupling on coarse-grained dynamics." pith.science (2026). https://pith.science/paper/YEEXTCAZ

@misc{pith2026250605966,
  author       = {Pith},
  title        = {Pith review of: The influence of multi-dimensionality and off-diagonal non-Markovian friction coupling on coarse-grained dynamics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YEEXTCAZ}},
  note         = {Machine review of arXiv:2506.05966}
}
read the original abstract

Coarse-graining complex molecular systems to lower-dimensional reaction coordinates is a powerful approach for capturing their effective dynamics. The generalized Langevin equation (GLE) provides an exact framework for modeling coarse-grained dynamics, and is particularly useful when non-Markovian effects are significant. While one-dimensional GLE models are commonly used, many systems require multi-dimensional reaction coordinates to account for coupled dynamics. Here, we study the GLE formalism for multi-dimensional reaction coordinates, incorporating a memory matrix to quantify non-Markovian frictional coupling between coordinates, and a multi-dimensional potential. Using the GLE model, in conjunction with a multi-dimensional Markovian embedding scheme, we investigate different systems that are characterized by two-dimensional reaction coordinates, namely the dihedral dynamics of pentane and alanine dipeptide, obtained from molecular dynamics simulations in explicit water. We identify significant off-diagonal friction couplings arising from intramolecular and hydrodynamic interactions. Unlike previous studies, our results highlight the critical role of different terms in the multi-dimensional GLE in accurately capturing key dynamical properties, including mean first-passage times and mean-squared displacements, particularly in systems with coupled non-Markovian coordinates.

Figures

Figures reproduced from arXiv: 2506.05966 by the authors.

Figure 2
Figure 2. figure 2. We assume symmetric friction coefficient matrices, [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

102 extracted references · 79 canonical work pages

  1. [69]

    B. W. L. Margolis and K. R. Lyons. ndsplines: A Python Library for Tensor-Product B-Splines of Arbitrary Di- mension.Journal of Open Source Software, 4(42):1745, 2019

    work page 2019

  2. [1]

    N. G. van Kampen. Remarks on Non-Markov Processes. Brazilian Journal of Physics, 28:90–96, 1998

    1998

  3. [2]

    Huang.Introduction to Statistical Physics

    K. Huang.Introduction to Statistical Physics. Chapman and Hall/CRC, 2009

    2009

  4. [3]

    APPLICA TION OF THE MUL TI-DIMENSIONAL GLE TO THE DIHEDRAL ANGLE DYNAMICS IN PENT ANE 3.1. GLE parameters extraction We apply our GLE model to describe the isomerization dynamics of a solvated pentane molecule obtained from MD simulation trajectories (see appendix D for details). In water, the pentane molecule rotates around the plane defined by the three...

  5. [4]

    We choose the dihedral anglesϕ(t) andψ(t) (see figure 4)

    CONFORMA TIONAL MOTION OF ALANINE DIPEPTIDE As a second application, we investigate the dihedral angle dynamics of alanine dipeptide in water, which is a classical model system for coarse-grained dynamics [51, 73–80]. We choose the dihedral anglesϕ(t) andψ(t) (see figure 4). To explore the role of off-diagonal friction 9 and multi-dimensional landscapes, ...

  6. [5]

    Our GLE model effectively reproduces the dynamics observed in atomistic MD simulations of isomerization

    DISCUSSION AND CONCLUSIONS We use a multi-dimensional GLE framework to model the isomerization dynamics of solvated molecules, in con- junction with a Markovian embedding scheme that ac- counts for correlations between forces on reaction coor- dinates via off-diagonal friction coupling and confinement within a multi-dimensional potential landscape. Our GL...

    2020

  7. [6]

    H. A. Karimi-Varzaneh, N. F. A. van Der Vegt, F. M¨ uller- Plathe, and P. Carbone. How Good are Coarse-Grained Polymer Models? A Comparison for Atactic Polystyrene. ChemPhysChem, 13(15):3428–3439, 2012

    2012

  8. [7]

    Kardar.Statistical Physics of Particles

    M. Kardar.Statistical Physics of Particles. Cambridge University Press, 2007

    2007

Show all 102 references
  1. [8]

    Balian.From Microphysics to Macrophysics: Meth- ods and Applications of Statistical Physics

    R. Balian.From Microphysics to Macrophysics: Meth- ods and Applications of Statistical Physics. Volume II. Springer Science & Business Media, 2007

    2007

  2. [9]

    Hij´ on, P

    C. Hij´ on, P. Espa˜ nol, E. Vanden-Eijnden, and R. Delgado-Buscalioni. Mori-Zwanzig Formalism as a Practical Computational Tool.Faraday Discussions, 144:301–322, 2010

    2010

  3. [10]

    Bagchi and D

    B. Bagchi and D. W. Oxtoby. The Effect of Frequency Dependent Friction on Isomerization Dynamics in Solu- tion.The Journal of Chemical Physics, 78(5):2735–2741, 1983

    1983

  4. [11]

    S. A. Adelman. Generalized Langevin Theory for Many- Body Problems in Chemical Dynamics: Reactions in Liq- uids.The Journal of Chemical Physics, 73(7):3145–3158, 1980

    1980

  5. [12]

    J. E. Straub, M. Borkovec, and B. J. Berne. Calculation of Dynamic Friction on Intramolecular Degrees of Free- dom.Journal of Physical Chemistry, 91(19):4995–4998, 1987

    1987

  6. [13]

    Canales and G

    M. Canales and G. Sese. Generalized Langevin Dynamics Simulations of NaCl Electrolyte Solutions.The Journal of Chemical Physics, 109(14):6004–6011, 1998

    1998

  7. [14]

    R. Zwanzig. Memory Effects in Irreversible Thermody- namics.Physical Review, 124(4):983, 1961

    1961

  8. [15]

    S. S. Plotkin and P. G. Wolynes. Non-Markovian Config- urational Diffusion and Reaction Coordinates for Protein Folding.Physical Review Letters, 80(22):5015, 1998

    1998

  9. [16]

    Satija and D

    R. Satija and D. E. Makarov. Generalized Langevin Equation as a Model for Barrier Crossing Dynamics in Biomolecular Folding.The Journal of Physical Chem- istry B, 123(4):802–810, 2019

    2019

  10. [17]

    O. F. Lange and H. Grubm¨ uller. Collective Langevin Dynamics of Conformational Motions in Proteins.The Journal of Chemical Physics, 124(21):214903, 2006

    2006

  11. [18]

    C. Ayaz, L. Tepper, F. N. Br¨ unig, J. Kappler, J. O. Dal- drop, and R. R. Netz. Non-Markovian Modeling of Pro- tein Folding.Proceedings of the National Academy of Sciences, 118(31):e2023856118, 2021

    2021

  12. [19]

    H. Mori. Transport, Collective Motion, and Brownian Motion.Progress of Theoretical Physics, 33(3):423–455, 1965

    1965

  13. [20]

    Carof, R

    A. Carof, R. Vuilleumier, and B. Rotenberg. Two Algo- rithms to Compute Projected Correlation Functions in Molecular Dynamics Simulations.The Journal of Chem- ical Physics, 140(12):124103, 2014

    2014

  14. [21]

    A. J. Chorin, O. H. Hald, and R. Kupferman. Optimal Prediction with Memory.Physica D: Nonlinear Phenom- ena, 166(3-4):239–257, 2002

    2002

  15. [22]

    B. A. Dalton, C. Ayaz, H. Kiefer, A. Klimek, L. Tepper, and R. R. Netz. Fast Protein Folding is Governed by Memory-Dependent Friction.Proceedings of the National Academy of Sciences, 120(31):e2220068120, 2023

    2023

  16. [23]

    Klippenstein, M

    V. Klippenstein, M. Tripathy, G. Jung, F. Schmid, and N. F. A. van der Vegt. Introducing Memory in Coarse- Grained Molecular Simulations.The Journal of Physical Chemistry B, 125(19):4931–4954, 2021

    2021

  17. [24]

    F. N. Br¨ unig, O. Geburtig, A. von Canal, J. Kappler, and R. R. Netz. Time-Dependent Friction Effects on Vibrational Infrared Frequencies and Line Shapes of Liquid Water.The Journal of Physical Chemistry B, 126(7):1579–1589, 2022

    2022

  18. [25]

    F. N. Br¨ unig, J. O. Daldrop, and R. R. Netz. Pair- Reaction Dynamics in Water: Competition of Memory, Potential Shape, and Inertial Effects.The Journal of Physical Chemistry B, 126(49):10295–10304, 2022

    2022

  19. [26]

    B. J. Berne, M. E. Tuckerman, J. E. Straub, and A. L. R. Bug. Dynamic Friction on Rigid and Flexible Bonds.The Journal of Chemical Physics, 93(7):5084–5095, 1990

    1990

  20. [27]

    B. A. Dalton, H. Kiefer, and R. R. Netz. The Role of Memory-Dependent Friction and Solvent Viscosity in Iso- merization Kinetics in Viscogenic Media.Nature Com- munications, 15(1):3761, 2024

    2024

  21. [28]

    B. A. Dalton, A. Klimek, H. Kiefer, F. N. Br¨ unig, H. Co- linet, L. Tepper, A. Abbasi, and R. R. Netz. Memory and Friction: From the Nanoscale to the Macroscale.Annual Review of Physical Chemistry, 76:431–454, 2025

    2025

  22. [29]

    Kiefer, D

    H. Kiefer, D. Furtel, C. Ayaz, A. Klimek, J. O. Daldrop, and R. R. Netz. Predictability Analysis and Prediction of Discrete Weather and Financial Time-Series Data with a Hamiltonian-Based Filter-Projection Approach.arXiv preprint arXiv:2409.15026, 2024

    2024 arXiv

  23. [30]

    R. B. Best, G. Hummer, and W. A. Eaton. Native Con- tacts Determine Protein Folding Mechanisms in Atom- istic Simulations.Proceedings of the National Academy of Sciences, 110(44):17874–17879, 2013

    2013

  24. [31]

    Ceriotti, G

    M. Ceriotti, G. Bussi, and M. Parrinello. Colored-Noise Thermostats ` a la Carte.Journal of Chemical Theory and Computation, 6(4):1170–1180, 2010

    2010

  25. [32]

    C. Ayaz, L. Scalfi, B. A. Dalton, and R. R. Netz. Generalized Langevin Equation with a Nonlinear Po- tential of Mean Force and Nonlinear Memory Friction From a Hybrid Projection Scheme.Physical Review E, 105(5):054138, 2022

    2022

  26. [33]

    Vroylandt

    H. Vroylandt. On the Derivation of the Generalized Langevin Equation and the Fluctuation-Dissipation The- orem.Europhysics Letters, 140(6):62003, 2022

    2022

  27. [34]

    J. S. Langer. Statistical Theory of the Decay of Metastable States.Annals of Physics, 54(2):258–275, 1969

    1969

  28. [35]

    A. M. Berezhkovskii and D. E. Makarov. Single-Molecule Test for Markovianity of the Dynamics Along a Reaction Coordinate.The Journal of Physical Chemistry Letters, 9(9):2190–2195, 2018

    2018

  29. [36]

    Abrash, S

    S. Abrash, S. Repinec, and R. M. Hochstrasser. The Vis- cosity Dependence and Reaction Coordinate for Isomer- ization of Cis-Stilbene.The Journal of Chemical Physics, 93(2):1041–1053, 1990

    1990

  30. [37]

    Bagchi.Molecular Relaxation in Liquids

    B. Bagchi.Molecular Relaxation in Liquids. Oxford Uni- versity Press, 2012

    2012

  31. [38]

    Mazur and D

    P. Mazur and D. Bedeaux. When and Why is the Ran- dom Force in Brownian Motion a Gaussian Process.Bio- physical Chemistry, 41(1):41–49, 1991

    1991

  32. [39]

    R. F. Grote and J. T. Hynes. Reactive Modes in Condensed Phase Reactions.The Journal of Chemical Physics, 74(8):4465–4475, 1981. 22

    1981

  33. [40]

    W. F. van Gunsteren, H. J. C. Berendsen, and J. A. C. Rullmann. Stochastic Dynamics for Molecules with Con- straints: Brownian Dynamics of n-Alkanes.Molecular Physics, 44(1):69–95, 1981

    1981

  34. [41]

    Acharya and B

    S. Acharya and B. Bagchi. Diffusion in a Two- Dimensional Energy Landscape in the Presence of Dy- namical Correlations and Validity of Random Walk Model.Physical Review E, 107(2):024127, 2023

    2023

  35. [42]

    J. S. Langer. Theory of the Condensation Point.Annals of Physics, 281(1-2):941–990, 2000

    2000

  36. [43]

    Vroylandt and P

    H. Vroylandt and P. Monmarch´ e. Position-Dependent Memory Kernel in Generalized Langevin Equations: Theory and Numerical Estimation.The Journal of Chemical Physics, 156(24):244105, 2022

    2022

  37. [44]

    van der Zwan and J

    G. van der Zwan and J. T. Hynes. Reactive Paths in the Diffusion Limit.The Journal of Chemical Physics, 77(3):1295–1301, 1982

    1982

  38. [45]

    J. D. Bryngelson and P. G. Wolynes. Intermediates and Barrier Crossing in a Random Energy Model (with Ap- plications to Protein Folding).The Journal of Physical Chemistry, 93(19):6902–6915, 1989

    1989

  39. [46]

    J. M. Deutch and I. Oppenheim. Molecular Theory of Brownian Motion for Several Particles.The Journal of Chemical Physics, 54(8):3547–3555, 1971

    1971

  40. [47]

    D. J. Kraft, R. Wittkowski, B. ten Hagen, K. V. Edmond, D. J. Pine, and H. L¨ owen. Brownian Motion and the Hydrodynamic Friction Tensor for Colloidal Particles of Complex Shape.Physical Review E, 88(5):050301, 2013

    2013

  41. [48]

    Acharya, S

    S. Acharya, S. Mondal, S. Mukherjee, and B. Bagchi. Rate of Insulin Dimer Dissociation: Interplay Between Memory Effects and Higher Dimensionality.The Journal of Physical Chemistry B, 125(34):9678–9691, 2021

    2021

  42. [49]

    Acharya and B

    S. Acharya and B. Bagchi. Non-Markovian Rate Theory on a Multidimensional Reaction Surface: Complex Inter- play Between Enhanced Configuration Space and Mem- ory.The Journal of Chemical Physics, 156(13):134101, 2022

    2022

  43. [50]

    Ansari, C

    A. Ansari, C. M. Jones, E. R. Henry, J. Hofrichter, and W. A. Eaton. The Role of Solvent Viscosity in the Dy- namics of Protein Conformational Changes.Science, 256(5065):1796–1798, 1992

    1992

  44. [51]

    D. L. Ermak and J. A. McCammon. Brownian Dynamics with Hydrodynamic Interactions.The Journal of Chem- ical Physics, 69(4):1352–1360, 1978

    1978

  45. [52]

    Reichert and H

    M. Reichert and H. Stark. Hydrodynamic Coupling of Two Rotating Spheres Trapped in Harmonic Potentials. Physical Review E, 69(3):031407, 2004

    2004

  46. [53]

    Jung and F

    G. Jung and F. Schmid. Frequency-Dependent Hydrody- namic Interaction Between Two Solid Spheres.Physics of Fluids, 29(12):126101, 2017

    2017

  47. [54]

    Schaudinnus, B

    N. Schaudinnus, B. Bastian, R. Hegger, and G. Stock. Multidimensional Langevin Modeling of Nonoverdamped Dynamics.Physical Review Letters, 115(5):050602, 2015

    2015

  48. [55]

    De Sancho, A

    D. De Sancho, A. Sirur, and R. B. Best. Molecular Ori- gins of Internal Friction Effects on Protein-Folding Rates. Nature Communications, 5(1):1–10, 2014

    2014

  49. [56]

    J. O. Daldrop, J. Kappler, F. N. Br¨ unig, and R. R. Netz. Butane Dihedral Angle Dynamics in Water is Dom- inated by Internal Friction.Proceedings of the National Academy of Sciences, 115(20):5169–5174, 2018

    2018

  50. [57]

    Hegger and G

    R. Hegger and G. Stock. Multidimensional Langevin Modeling of Biomolecular Dynamics.The Journal of Chemical Physics, 130(3):034106, 2009

    2009

  51. [58]

    Z. She, P. Ge, and H. Lei. Data-Driven Construction of Stochastic Reduced Dynamics Encoded with Non- Markovian Features.The Journal of Chemical Physics, 158(3):034102, 2023

    2023

  52. [59]

    Lickert and G

    B. Lickert and G. Stock. Modeling Non-Markovian Data Using Markov State and Langevin Models.The Journal of Chemical Physics, 153(24):244112, 2020

    2020

  53. [60]

    Zwanzig.Nonequilibrium Statistical Mechanics

    R. Zwanzig.Nonequilibrium Statistical Mechanics. Ox- ford University Press, 2001

    2001

  54. [61]

    H. S. Lee, S.-H. Ahn, and E. F. Darve. The Multi- Dimensional Generalized Langevin Equation for Confor- mational Motion of Proteins.The Journal of Chemical Physics, 150(17):174113, 2019

    2019

  55. [62]

    Darve and A

    E. Darve and A. Pohorille. Calculating Free Energies Using Average Force.The Journal of Chemical Physics, 115(20):9169–9183, 2001

    2001

  56. [63]

    Lyu and H

    L. Lyu and H. Lei. Construction of Coarse-Grained Molecular Dynamics with Many-Body Non-Markovian Memory.Physical Review Letters, 131(17):177301, 2023

    2023

  57. [64]

    Xie and W

    P. Xie and W. E. Coarse-Graining Conformational Dynamics with Multidimensional Generalized Langevin Equation: How, When, and Why.Journal of Chemical Theory and Computation, 20(18):7708–7715, 2024

    2024

  58. [65]

    P. Xie, R. Car, and W. E. Ab Initio Generalized Langevin Equation.Proceedings of the National Academy of Sci- ences, 121(14):e2308668121, 2024

    2024

  59. [66]

    Z. Li, H. S. Lee, E. Darve, and G. E. Karniadakis. Com- puting the Non-Markovian Coarse-Grained Interactions Derived from the Mori-Zwanzig Formalism in Molecular Systems: Application to Polymer Melts.The Journal of Chemical Physics, 146(1):014104, 2017

    2017

  60. [67]

    B. J. Berne and G. D. Harp. On the Calculation of Time Correlation Functions.Advances in Chemical Physics, 17:63–227, 1970

    1970

  61. [68]

    Kowalik, J

    B. Kowalik, J. O. Daldrop, J. Kappler, J. C. F. Schulz, A. Schlaich, and R. R. Netz. Memory-Kernel Ex- traction for Different Molecular Solutes in Solvents of Varying Viscosity in Confinement.Physical Review E, 100(1):012126, 2019

    2019

  62. [70]

    B. G. Mitterwallner, L. Lavacchi, and R. R. Netz. Neg- ative Friction Memory Induces Persistent Motion.The European Physical Journal E, 43(10):1–11, 2020

    2020

  63. [71]

    Kappler, V

    J. Kappler, V. B. Hinrichsen, and R. R. Netz. Non- Markovian Barrier Crossing with Two-Time-Scale Mem- ory is Dominated by the Faster Memory Component.The European Physical Journal E, 42(9):1–16, 2019

    2019

  64. [72]

    Lavacchi, J

    L. Lavacchi, J. Kappler, and R. R. Netz. Barrier Crossing in the Presence of Multi-Exponential Memory Functions with Unequal Friction Amplitudes and Memory Times. Europhysics Letters, 131(4):40004, 2020

    2020

  65. [73]

    Kiefer, B

    H. Kiefer, B. J. A. H´ ery, L. Tepper, B. A. Dalton, C. Ayaz, and R. R. Netz. Analysis and Simulation of Generalized Langevin Equations with Non-Gaussian Or- thogonal Forces.arXiv preprint arXiv:2505.15665, 2025

    2025 arXiv

  66. [74]

    Hummer and I

    G. Hummer and I. G. Kevrekidis. Coarse Molecular Dy- namics of a Peptide Fragment: Free Energy, Kinetics, and Long-Time Dynamics Computations.The Journal of Chemical Physics, 118(23):10762–10773, 2003

    2003

  67. [75]

    Klimek and R

    A. Klimek and R. R. Netz. Optimal Non-Markovian Composite Search Algorithms for Spatially Correlated Targets.Europhysics Letters, 139(3):32003, 2022. 23

    2022

  68. [76]

    Kappler, J

    J. Kappler, J. O. Daldrop, F. N. Br¨ unig, M. D. Boehle, and R. R. Netz. Memory-Induced Acceleration and Slow- down of Barrier Crossing.The Journal of Chemical Physics, 148(1):014903, 2018

    2018

  69. [77]

    P. E. Smith, B. M. Pettitt, and M. Karplus. Stochastic Dynamics Simulations of the Alanine Dipeptide Using a Solvent-Modified Potential Energy Surface.The Journal of Physical Chemistry, 97(26):6907–6913, 1993

    1993

  70. [78]

    X. Wu, B. R. Brooks, and E. Vanden-Eijnden. Self- Guided Langevin Dynamics via Generalized Langevin Equation.Journal of Computational Chemistry, 37(6):595–601, 2016

    2016

  71. [79]

    Prada-Gracia, J

    D. Prada-Gracia, J. G´ omez-Garde˜ nes, P. Echenique, and F. Falo. Exploring the Free Energy Landscape: From Dynamics to Networks and Back.PLOS Computational Biology, 5(6):e1000415, 2009

    2009

  72. [80]

    Stamati, C

    H. Stamati, C. Clementi, and L. E. Kavraki. Appli- cation of Nonlinear Dimensionality Reduction to Char- acterize the Conformational Landscape of Small Pep- tides.Proteins: Structure, Function, and Bioinformatics, 78(2):223–235, 2010

    2010

  73. [81]

    Leimkuhler and C

    B. Leimkuhler and C. Matthews. Robust and Efficient Configurational Molecular Sampling via Langevin Dynamics.The Journal of Chemical Physics, 138(17):174102, 2013

    2013

  74. [82]

    Jung and G

    B. Jung and G. Jung. Dynamic Coarse-Graining of Linear and Non-Linear Systems: Mori-Zwanzig Formal- ism and Beyond.The Journal of Chemical Physics, 159(8):084110, 2023

    2023

  75. [83]

    Kmiecik, D

    S. Kmiecik, D. Gront, M. Kolinski, L. Wieteska, A. E. Dawid, and A. Kolinski. Coarse-Grained Pro- tein Models and Their Applications.Chemical Reviews, 116(14):7898–7936, 2016

    2016

  76. [84]

    Mardt, L

    A. Mardt, L. Pasquali, H. Wu, and F. No´ e. V AMPnets for Deep Learning of Molecular Kinetics.Nature Com- munications, 9(1):5, 2018

    2018

  77. [85]

    C. Ayaz, L. Tepper, and R. R. Netz. Self-Consistent Markovian Embedding of Generalized Langevin Equa- tions with Configuration-Dependent Mass and a Non- linear Friction Kernel.Turkish Journal of Physics, 46(6):194–205, 2022

    2022

  78. [86]

    R. O. Rosenberg, B. J. Berne, and D. Chandler. Iso- merization Dynamics in Liquids by Molecular Dynamics. Chemical Physics Letters, 75(1):162–168, 1980

    1980

  79. [87]

    N. Wolf, V. Klippenstein, and N. F. A. van der Vegt. Cross-Correlations in the Fluctuation–Dissipation Rela- tion Influence Barrier-Crossing Dynamics.The Journal of Chemical Physics, 162(5):054113, 2025

    2025

  80. [88]

    Durlofsky, J

    L. Durlofsky, J. F. Brady, and G. Bossis. Dynamic Sim- ulation of Hydrodynamically Interacting Particles.Jour- nal of Fluid Mechanics, 180:21–49, 1987

    1987

  81. [89]

    Happel and H

    J. Happel and H. Brenner.Low Reynolds Number Hy- drodynamics: With Special Applications to Particulate Media, volume 1. Springer Science & Business Media, 2012

    2012

  82. [90]

    H. J. C. Berendsen, J. R. Grigera, and T. P. Straatsma. The Missing Term in Effective Pair Potentials.Journal of Physical Chemistry, 91(24):6269–6271, 1987

    1987

  83. [91]

    Pronk, S

    S. Pronk, S. P´ all, R. Schulz, P. Larsson, P. Bjelkmar, R. Apostolov, M. R. Shirts, J. C. Smith, P. M. Kas- son, D. van der Spoel, et al. GROMACS 4.5: A High- Throughput and Highly Parallel Open Source Molecular Simulation Toolkit.Bioinformatics, 29(7):845–854, 2013

    2013

  84. [92]

    Oostenbrink, A

    C. Oostenbrink, A. Villa, A. E. Mark, and W. F. van Gunsteren. A Biomolecular Force Field Based on the Free Enthalpy of Hydration and Solvation: The GROMOS Force-Field Parameter Sets 53A5 and 53A6.Journal of Computational Chemistry, 25(13):1656–1676, 2004

    2004

  85. [93]

    and a cut-off of 1 nm. Additionally, we simulate a solvated alanine dipeptide molecule in a cubic (1.2 nm) 3 water box , using the Am- ber99 force field for the peptide [94] and the TIP3P water model [95] for the 1134 water molecules. Bond lengths and angles are fixed by the L...

  86. [94]

    Ryckaert, G

    J.-P. Ryckaert, G. Ciccotti, and H. J. C. Berendsen. Nu- merical Integration of the Cartesian Equations of Motion of a System with Constraints: Molecular Dynamics of n- Alkanes.Journal of Computational Physics, 23(3):327– 341, 1977

    1977

  87. [95]

    H. J. C. Berendsen, J. P. M. Postma, W. F. van Gun- steren, A. DiNola, and J. R. Haak. Molecular Dynam- ics with Coupling to an External Bath.The Journal of Chemical Physics, 81(8):3684–3690, 1984

    1984

  88. [96]

    Bussi, D

    G. Bussi, D. Donadio, and M. Parrinello. Canonical Sam- pling Through Velocity Rescaling.The Journal of Chem- ical Physics, 126(1):014101, 2007

    2007

  89. [97]

    Darden, D

    T. Darden, D. York, and L. Pedersen. Particle Mesh Ewald: An N log(N) Method for Ewald Sums in Large Systems.The Journal of Chemical Physics, 98(12):10089–10092, 1993

    1993

  90. [98]

    Salomon-Ferrer, D

    R. Salomon-Ferrer, D. A. Case, and R. C. Walker. An Overview of the Amber Biomolecular Simulation Pack- age.Wiley Interdisciplinary Reviews: Computational Molecular Science, 3(2):198–210, 2013

    2013

  91. [99]

    W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. W. Impey, and M. L. Klein. Comparison of Simple Potential Functions for Simulating Liquid Water.The Journal of Chemical Physics, 79(2):926–935, 1983

    1983

  92. [100]

    B. Hess, H. Bekker, H. J. C. Berendsen, and J. G. E. M. Fraaije. LINCS: A Linear Constraint Solver for Molec- ular Simulations.Journal of Computational Chemistry, 18(12):1463–1472, 1997

    1997

  93. [101]

    P. de Buyl. tidynamics: A Tiny Package to Compute the Dynamics of Stochastic and Molecular Simulations. Journal of Open Source Software, 3(28):877, 2018

    2018

  94. [102]

    Virtanen, R

    P. Virtanen, R. Gommers, T. E. Oliphant, M. Haber- land, T. Reddy, D. Cournapeau, E. Burovski, P. Peter- son, W. Weckesser, J. Bright, et al. SciPy 1.0: Funda- mental Algorithms for Scientific Computing in Python. Nature Methods, 17(3):261–272, 2020

    2020

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.