{"id":"7f16908a-a601-4adc-b3a7-d5e4d9effb22","arxiv_id":"2412.10409","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A DDPM-derived force field combined with the string method produces minimum free energy paths for Müller-Brown and alanine dipeptide, with accuracy sensitive to the chosen noise level.","lead":"Diffusion models trained on molecular data can sketch the most likely paths between stable molecular states. The paper combines a denoising diffusion model with the string method and tests it on a simple energy landscape and a solvated peptide.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed accuracy rests on an empirically tuned noise level that is selected using the true path (Müller-Brown) and remains unvalidated for alanine dipeptide, so the paper does not yet establish a self-contained MFEP method.","rationale":"The reader's weakest-assumption analysis correctly identifies the noise level as the key uncontrolled hyperparameter. My stress test reaches the same conclusion through a more explicit mechanism: a fixed intermediate t corresponds to the score of the perturbed distribution q_t, so the string method converges to the MFEP of the smoothed potential, not the target. The empirical optimization of t against the true path on Müller-Brown (Fig. 2) means the reported 'accuracy' is not predictive for unknown systems. For alanine dipeptide, the lack of any reference path makes the claim of accurately recovering the transition path unsupported; the PII observation is suggestive but not quantitative. These are addressable with additional experiments (a reference MFEP and a parameter-selection rule), so the paper's conditional acceptance is appropriate. I do not see a fundamental inconsistency in the method itself, only insufficient validation, and therefore recommend keeping the reader's CONDITIONAL verdict (UNCHANGED).","tokens_in":3636,"tokens_out":5967,"duration_ms":63125,"concrete_test":"For the alanine dipeptide system, compute a reference MFEP on the same 15-dimensional backbone-atom coordinates using an established method (e.g., the string method with mean forces from umbrella sampling or adaptive biasing force). Then compare the DDPM-generated path at t=50 and at t=20/80 with the reference using both the maximum displacement between interpolated strings and the free-energy profile (PMF) along each path. If the t=50 path lies within one thermal fluctuation of the reference or the PMF barrier differs by less than k_BT, the concern is resolved. Additionally, on a two-dimensional test potential with known MFEP, select t without using the true path (e.g., smallest t for which the string converges within a fixed iteration budget) and check whether the resulting path matches the exact MFEP within tolerance; this tests whether the method is self-contained.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that DDPM-derived force fields accurately generate MFEPs. The force field is taken as −(1−ᾱ_t)^(−1/2) εθ(x,t) at a fixed noise level t. At intermediate t this is the score of a Gaussian-smoothed version of the equilibrium distribution, not the true mean force, so the string evolves on a biased landscape. For Müller-Brown, t≈30 is selected by minimizing the maximum displacement to the true string (Fig. 2), i.e., the ground truth is used to choose the key hyperparameter; this is circular as a validation of accuracy. For alanine dipeptide, t=50 is chosen without any reference MFEP; the only success indicator is the qualitative appearance of PII, which is known to be solvent-stabilized but also lies near the direct transition. The paper reports no error bars, no sensitivity analysis in t, and no free-energy barriers along the generated path. Absent an independent noise-level selection rule or a reference calculation, the claim that DDPM-derived force fields accurately generate transition pathways is not yet established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes combining denoising diffusion probabilistic models (DDPMs) with the string method to generate minimum free energy paths (MFEPs). The core idea is to use the DDPM score, scaled as −(1−ᾱ_t)^(−1/2) εθ(x,t), as an approximate mean force for evolving a discretized string. The method is tested on the analytical Müller-Brown potential and on solvated alanine dipeptide, with the authors reporting that at certain intermediate noise levels (t≈30 for Müller-Brown, t=50 for alanine dipeptide) the generated path recovers the expected transition pathway, including a PII state in alanine dipeptide that is attributed to implicit solvent effects. Data filtering is also applied: an energy threshold for Müller-Brown and a Ramachandran density filter for alanine dipeptide.","tokens_in":3894,"tokens_out":3892,"duration_ms":42805,"significance":"If the central claim holds, the method would offer a way to compute MFEPs without explicit free-energy calculations or enhanced-sampling reference runs, using generative model scores as surrogate forces. The paper provides a simple proof-of-concept, explicitly studies how accuracy depends on noise level for Müller-Brown, and makes code available. However, the validation strategy is partly circular because the key hyperparameters are tuned using the true path, and the alanine dipeptide result lacks any quantitative accuracy metric. The significance is therefore currently limited, but the approach is plausible and merits further rigorous validation.","major_comments":[{"comment":"The optimal noise level for Müller-Brown is selected by minimizing the maximum displacement to the true string images: the text says 'an empirically derived optimal noise level around t = 30, where the path deviation is minimized.' Because the ground truth is used to choose this key hyperparameter, the subsequent claim that the generated path is accurate is circular. Please provide an independent criterion for selecting t (for example, based on consistency of the predicted forces across t, or on a validation set not used for training), and report the accuracy metric over a range of t with error bars across the 10 independently trained networks; the current figure shows only averaged curves without uncertainty estimates.","section":"III.A, Fig. 2"},{"comment":"The force −(1−ᾱ_t)^(−1/2) εθ(x,t) used to evolve the string is, for finite t, the score of the noise-smoothed distribution q_t, not the gradient of the true potential of mean force ∇ log q_0. The paper itself states that the approximation holds 'at sufficiently low noise levels,' yet the chosen values t≈30 (with T=100) and t=50 (with T=1000) are not shown to be in this regime. The string method requires the true mean force for the MFEP to be well defined; evolving on a smoothed landscape generally biases the path. Please quantify the bias at the chosen t, for example by comparing the DDPM score against a numerical gradient of the true log-density for Müller-Brown, or by demonstrating that the generated path converges to the true MFEP as t approaches 1 and that the residual at t=30 is within the convergence tolerance.","section":"II, force-field equation"},{"comment":"The alanine dipeptide result provides no quantitative accuracy metric: the generated MFEP is only displayed on a Ramachandran plot, and the claim of 'implicitly capturing solvent effects' rests on the path passing through a metastable PII conformation. However, PII lies close to the direct transition region, so this observation alone does not discriminate between a correct solvent-aware path and an artifact of the chosen noise level or density filter. Please compare against an independent reference MFEP obtained from an explicit-solvent free-energy calculation (for example, the string method with umbrella sampling or metadynamics) and report a quantitative measure such as the average distance in Ramachandran space, the free-energy profile along the path, or committor probabilities. The choice t=50 and the density-filter threshold (retaining only configurations with probability density higher than average) also need independent justification and a sensitivity study.","section":"III.B"},{"comment":"The free-energy filtering threshold (−30) is an additional tuned parameter whose value is not independently justified; the paper acknowledges it 'should be monitored carefully,' but no sensitivity analysis is given. Because filtering changes the training distribution, it directly changes the learned score and hence the force field used by the string method. Please report the dependence of the generated path on the threshold (including unfiltered, moderately filtered, and strongly filtered datasets) and show that the stated accuracy is not an artifact of a particular threshold choice. If the threshold is set relative to the true path, that further strengthens the circularity concern raised in the first major comment.","section":"III.A, energy filtering"}],"minor_comments":[{"comment":"The axes of Figure 2 are labeled only with 'β', which is unclear; please use descriptive axis labels (for example, the horizontal axis as 'noise level t' and the vertical axis as 'maximum displacement' with units or dimensionless definition).","section":"Fig. 2"},{"comment":"The abstract states that the method works 'at some range of noise levels,' but no range is ever determined; please either specify the acceptable range from the experiments or revise the abstract to reflect that only isolated noise levels were tested.","section":"Abstract and Section III.A"},{"comment":"Figure 1 shows a path at noise level t=29, while the text reports the optimal level as t=30; please make the figure and text consistent or explain the difference.","section":"Fig. 1 and III.A"},{"comment":"The Code Availability section mentions a GitHub repository but provides no URL or DOI; please include a link or a permanent repository identifier so that the claimed code availability is verifiable.","section":"Code Availability"},{"comment":"The symbol t is used for the discrete diffusion timestep and also as the parameter in the force expression; please clarify whether t is the discrete index in the variance schedule β_1,...,β_T and specify the actual β schedule used in each experiment (for example, linear or cosine), since the chosen t values are otherwise hard to interpret.","section":"II, notation"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is an interesting proof-of-concept, but the validation is not yet self-contained: the central hyperparameters are chosen using the true path, and the molecular system lacks a quantitative reference. I would support publication after the authors provide an independent noise-level selection rule, quantify the smoothing bias, and add a comparison to a reference MFEP for alanine dipeptide. The paper is quite short and could benefit from additional experimental detail, but the scope is appropriate for a methods-focused journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe one thing to know: this is a genuine new combination, but the accuracy claim is not yet supported. The authors take the established result that a DDPM score approximates the mean force at low noise, feed it into the string method, and show paths on Müller-Brown and alanine dipeptide. That application is new relative to Arts et al., who used the score for dynamics rather than path optimization. The idea is clean and the presentation is straightforward.\n\nWhat it does well: the score-to-force relation is externally grounded, the string update is standard, and the paper ships code. The Müller-Brown experiment tests many noise levels and reports a deviation curve, and the alanine result qualitatively recovers a PII state, which is plausibly solvent-stabilized. The citation pattern is fine; no self-citation or missing key references that I can see.\n\nThe soft spots are load-bearing. For Müller-Brown, the optimal noise level t ≈ 30 is selected by minimizing the maximum displacement between the converged string and the true string (Figure 2). That is tuning the key hyperparameter with the answer key, so the resulting accuracy is partly fitted, not predictive. The energy filter threshold is also chosen post hoc. There are no error bars on the deviation curve, only an average over ten networks. For alanine dipeptide, t = 50 is chosen without any independent criterion and there is no reference MFEP, no barrier calculation, and no numerical metric—just a Ramachandran plot. The claim about implicitly capturing solvent effects rests on the qualitative appearance of PII, with no comparison to a solvent-free calculation or a known path. The paper does not show that a user could pick the right noise level without already knowing the answer.\n\nI agree with the stress-test note. The central issue is not the math; it is the validation protocol. This is fixable: the authors could propose a noise-selection rule that does not use the true path, add a reference MFEP for alanine from umbrella sampling or metadynamics, report barriers along the path, and include sensitivity analysis in t and the filter threshold.\n\nWho is this for? Computational chemists working on generative-model surrogates for enhanced sampling. As a proof of concept it is worth reading, but it needs major revision before the accuracy claim is credible. I would send it to peer review rather than desk reject, with the expectation that the author does the additional validation. If that work lands, this could become a useful practical tool.\n\nRecommendation: engage, but require the missing validation before accepting.","headline":"The DDPM-score-plus-string combination is new and sensible, but the headline accuracy claim rests on a noise level tuned against the true path and an unbenchmarked alanine dipeptide result, so the paper is a promising proof-of-concept that needs major validation work.","tokens_in":4363,"tokens_out":2668,"would_cite":false,"duration_ms":29372,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A denoising diffusion model's predicted noise, interpreted as a mean-force gradient, can drive the string method to recover minimum free energy paths in molecular systems.","keywords":["denoising diffusion probabilistic models","string method","minimum free energy path","potential of mean force","Müller-Brown potential","alanine dipeptide","implicit solvation","transition pathways"],"falsifier":"For solvated alanine dipeptide, compute a reference minimum free energy path with umbrella sampling along the backbone dihedral angles, then run the DDPM-string method over a range of noise levels; the central claim fails if no single noise level reproduces the reference path and its polyproline II intermediate without prior knowledge of the true path.","tokens_in":3438,"feed_emoji":"🧪","tokens_out":11588,"duration_ms":101350,"temperature":0.7,"pith_summary":"The paper argues that denoising diffusion probabilistic models (DDPMs) can be turned into force fields for finding minimum free energy paths (MFEPs). It combines a DDPM's learned score, which at low noise approximates the gradient of the potential of mean force, with the string method, which evolves a discretized path until the force orthogonal to it vanishes. The method is tested on the two-dimensional Müller-Brown potential and on solvated alanine dipeptide. For each system, one intermediate noise level (around $t=30$ for Müller-Brown, $t=50$ for alanine dipeptide) produces a path that matches the reference transition path, and for alanine dipeptide the path passes through a water-stabilized polyproline II state, showing that solvent effects are captured implicitly. If this holds generally, transition path sampling becomes possible without explicit free energy calculations, at a fraction of the cost of full-dimensional sampling.","feed_headline":"Diffusion model force fields recover molecular transition paths","feed_subtitle":"A single noise level recovers both test paths and a solvent-stabilized intermediate","key_machinery":"The load-bearing identity is the score approximation $\\nabla_x \\log q(x) \\approx -\\frac{1}{\\sqrt{1-\\bar\\alpha_t}}\\epsilon_\\theta(x,t)$, which converts the network's predicted noise into an estimate of the mean-force gradient; feeding this force into the string method's evolution gives the working algorithm. The string is a discrete chain of images, each updated by one forward Euler step under the DDPM-derived force, then reparametrized by linear interpolation to maintain equal spacing, iterating until the maximum image displacement falls below a tolerance. The noise level $t$ plays the role of a smoothing parameter: low noise is most faithful to the sampled distribution but converges slowly, while intermediate noise smooths the landscape and accelerates convergence, and the paper chooses one intermediate level for each test system.","core_discovery":"On the paper's own terms, the central discovery is that a DDPM trained only on equilibrium configurations can serve as the force field for the string method, eliminating the need to compute the potential of mean force explicitly. At sufficiently low noise levels the DDPM's predicted noise is proportional to the gradient of the log-density, so it directly supplies $\\nabla W$; the string method then evolves a chain of images until the component of this force perpendicular to the path is zero. The paper reports that for the analytical Müller-Brown landscape the best path is obtained at an intermediate noise level around $t=30$, and for solvated alanine dipeptide at $t=50$. For alanine dipeptide, despite training on only the coordinates of five backbone atoms, the generated MFEP passes through the water-stabilized polyproline II conformation, indicating that the implicitly modeled solvent environment shapes the transition path.","pith_inferences":["A testable extension is to make the noise level spatially adaptive, using low noise near the minima and higher noise near the barrier, because the paper's single global $t$ applies one smoothing to the whole path.","The reliance on an empirically chosen noise level is the main practical bottleneck; a selection rule based on the curvature of the landscape or on convergence statistics of the string would make the method predictive rather than retrospective.","If the alanine dipeptide result is reproduced, retraining the same network on vacuum data should remove the polyproline II intermediate, offering a direct check that solvation is what the implicit model learns.","The diffusion-model force could be paired with other path search tools, such as nudged elastic band or transition interface sampling, beyond the string method used here."],"forward_implications":["A transition path can be generated without explicit free-energy calculations, provided an equilibrium training set is available.","The alanine dipeptide example indicates that solvent effects can be captured implicitly, so path generation can use a few backbone atoms rather than the full solvated system.","Because the method works at intermediate noise levels, it can trade fine-grained accuracy for faster convergence, with the optimal level set empirically per system.","For the Müller-Brown potential, the maximum-displacement metric shows the generated path is close to the true string at the optimal noise level, suggesting the approximation is quantitatively useful."],"supporting_citations":[{"why":"Supplies the DDPM parametrization, noise schedule, and training loss that the paper uses to learn the score.","marker":"[1]"},{"why":"Defines the diffusion and denoising process that the method inherits for its noise levels.","marker":"[2]"},{"why":"Establishes the key premise that low-noise DDPM scores approximate mean-force gradients, enabling the dynamics used here.","marker":"[3]"},{"why":"Provides the string method that evolves a discretized path to satisfy the minimum-free-energy-path condition.","marker":"[4]"},{"why":"Supplies the molecular dynamics tool used to generate the solvated alanine dipeptide training dataset.","marker":"[5]"},{"why":"Identifies the water-stabilized polyproline II conformation that the generated alanine dipeptide path passes through.","marker":"[6]"}],"fun_headline_variants":["Diffusion model force fields replace explicit PMF","Denoising diffusion recovers transition paths without PMF","Implicit solvent captured by diffusion-model paths","String method meets diffusion models for free energy paths"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that one intermediate noise level, chosen without comparing to a known true path, approximates the true mean-force gradient well enough across the whole transition region to drive the string to the correct minimum free energy path.","fun_headline_variants_meta":{"raw":{"variants":["Diffusion model force fields replace explicit PMF","Denoising diffusion recovers transition paths without PMF","Implicit solvent captured by diffusion-model paths","String method meets diffusion models for free energy paths"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1426,"prompt_tokens":827,"completion_tokens":599,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":443,"completion_tokens_details":{"reasoning_tokens":539}},"tokens_in":443,"tokens_out":599,"duration_ms":6829,"temperature":1.0,"reasoning_tokens":539,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T20:40:21.329446+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For solvated alanine dipeptide, compute a reference minimum free energy path with umbrella sampling along the backbone dihedral angles, then run the DDPM-string method over a range of noise levels; the central claim fails if no single noise level reproduces the reference path and its polyproline II intermediate without prior knowledge of the true path.","supporting_citations":[{"cited_title":"Generating Minimum Free Energy Paths With Denoising Diffusion Probabilistic Models","cited_arxiv_id":"2412.10409","evidence_quote":"Supplies the DDPM parametrization, noise schedule, and training loss that the paper uses to learn the score."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the diffusion and denoising process that the method inherits for its noise levels."},{"cited_title":"Sohl-Dickstein, E","cited_arxiv_id":null,"evidence_quote":"Establishes the key premise that low-noise DDPM scores approximate mean-force gradients, enabling the dynamics used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the string method that evolves a discretized path to satisfy the minimum-free-energy-path condition."},{"cited_title":"Weinan, W","cited_arxiv_id":null,"evidence_quote":"Supplies the molecular dynamics tool used to generate the solvated alanine dipeptide training dataset."},{"cited_title":"Van Der Spoel, E","cited_arxiv_id":null,"evidence_quote":"Identifies the water-stabilized polyproline II conformation that the generated alanine dipeptide path passes through."}],"review_version":1}