Pith. sign in

REVIEW 3 major objections 5 minor 17 references

Learning to engineer protein flexibility

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper shows that protein inverse folding models can be steered toward increased flexibility by conditioning on per-residue flexibility predictions from a learned predictor.

desk verdict Useful flexibility predictors, but the design steering claim is only validated against its own surrogate. read the letter →

arxiv 2412.18275 v2 pith:6WM6MDVM submitted 2024-12-24 q-bio.BM

Petr Kouba , Joan Planas-Iglesias , Jiri Damborsky , Jiri Sedlar , Stanislav Mazurenko , Josef Sivic This is my paper · ORCID
classification q-bio.BM
keywords proteinflexibilityinversefoldingdesignmoleculardynamicsRMSFMPNNFlexpertdeeplearning
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

The paper aims to make protein flexibility an explicit, controllable input to generative protein design. It first benchmarks ways of quantifying flexibility and settles on molecular-dynamics-derived root mean square fluctuations as the learning target. It then trains fast predictors, Flexpert-Seq from sequence alone and Flexpert-3D from sequence plus backbone structure, using a pre-trained protein language model to overcome limited data. Finally, it introduces Flexpert-Design, which fine-tunes an inverse folding model to accept flexibility instructions and generate sequences with increased flexibility in specified regions. If correct, this would let protein engineers request flexible loops, tunnels, or other regions directly, without slow molecular dynamics simulations in the design loop.

What carries the argument

The load-bearing component is Flexpert-3D: a protein language model (ProtTrans) with LoRA fine-tuning and a linear regression head, augmented by a CNN adaptor that injects ANM-computed flexibility values into the embedding space, so the model learns to correct crude ANM estimates toward MD ground truth. Flexpert-Design then wraps this predictor around ProteinMPNN: flexibility instructions are added as zero-initialized node features, sequences are sampled with a straight-through Gumbel-Softmax estimator, and the sampled sequence is passed through Flexpert-3D so that a flexibility-matching loss can be backpropagated while sequence cross-entropy loss keeps the inverse folding ability intact.

What would settle it

Take a sample of Flexpert-Design-generated sequences from CATH4.3, run new atomistic molecular dynamics simulations on both native and designed proteins under identical conditions, and compute per-residue RMSF; if the MD-based median enrichment ratio in the engineered regions falls to roughly 1, the steering effect is an artifact of the predictor rather than real flexibility change.

Watch

Extended reading notes

Core claim

The central claim is that per-residue protein flexibility can be predicted quickly and then used as a conditioning signal for inverse folding. Flexpert-3D, which combines the sequence-based Flexpert-Seq with an Anisotropic Network Model corrected by a small convolutional adaptor, reaches a Pearson correlation of 0.83 to MD-derived flexibility on the ATLAS test set. Flexpert-Design takes a ProteinMPNN inverse folding model, adds flexibility instructions as input node features, and fine-tunes it with a loss that matches the flexibility of sampled sequences to the instructions using Flexpert-3D as the evaluator. On CATH4.3, the resulting model achieves a median flexibility enrichment ratio of 1.52 and 83% flexibility-increasing mutations, compared with 1.07 and 61% for the vanilla ProteinMPNN baseline.

Load-bearing premise

The entire steering result rests on the assumption that Flexpert-3D's flexibility predictions stay accurate for mutated, designed sequences rather than systematically overpredicting flexibility for any design change.

Editorial extensions

If this is right

  • Protein engineers can request increased flexibility in a designated contiguous region while keeping the backbone fixed, and the redesigned sequences show higher predicted flexibility in that region.
  • The flexibility predictors are fast enough to be embedded in iterative design pipelines, unlike the molecular dynamics simulations used to generate their training labels.
  • Fine-tuning with the flexibility loss retains sequence recovery within about one percentage point of the ProteinMPNN baseline, so the steering does not come at a large cost to inverse folding quality.
  • The same procedure is not effective for decreasing flexibility, so the method currently provides one-directional control over protein flexibility.
  • The flexibility signal is continuous and per-residue, which makes the training loop adaptable to other per-residue design objectives beyond flexibility.

Reading between the lines

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

  • Inference: The steering result is measured by the same Flexpert-3D predictor that generated the training pseudolabels, so an independent check with new molecular dynamics simulations of designed sequences would be needed to confirm that the enrichment reflects true conformational flexibility rather than predictor bias.
  • Inference: The increased proportion of glycine and alanine in engineered segments is biochemically plausible but also points to a possible shortcut; testing whether the model still raises flexibility when these small residues are disallowed would clarify whether the signal is structural or residue-type-driven.
  • Inference: The same conditioning-and-loss loop could plausibly be applied to other inverse folding backbones, such as flow-matching or diffusion-based design models, and to other per-residue properties such as stability or solubility.
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

3 major / 5 minor

Summary. The paper addresses the problem of engineering protein flexibility in computational protein design. The authors first compare several flexibility quantification methods (MD RMSF, PDB B-factors, AlphaFold2/ESMFold pLDDT, GNM, ANM) on the ATLAS dataset, concluding that MD-derived RMSF is the most reliable learning target and that ANM/GNM are the strongest structure-based baselines. They then introduce two flexibility predictors: Flexpert-Seq, which uses a ProtTrans protein language model with LoRA fine-tuning and a linear regression head, and Flexpert-3D, which additionally incorporates ANM features through a CNN adaptor. Both predictors are evaluated on held-out ATLAS test proteins and on the mdCATH dataset across simulation temperatures; Flexpert-3D achieves a Pearson correlation of 0.83 with MD RMSF, outperforming ANM (0.76) and approaching an estimated upper bound of 0.88. Finally, the paper proposes Flexpert-Design, a method that fine-tunes ProteinMPNN to accept per-residue flexibility instructions, using Flexpert-3D to generate pseudolabels for native sequences (Eq. 1), a flexibility-matching loss (Eq. 2), and a flexibility enrichment ratio for evaluation (Eq. 6). On CATH4.3, Flexpert-Design is reported to achieve a median enrichment ratio of 1.52 and 83% flexibility-increasing mutations, compared to 1.07 and 61% for vanilla ProteinMPNN.

Significance. If the design claim holds, the paper makes a useful contribution by demonstrating that inverse folding models can be conditioned on per-residue flexibility instructions, which is a novel capability with potential impact on enzyme engineering and protein design. The predictor contribution is solid and independently meaningful: the comparison of flexibility quantification methods is informative, and Flexpert-Seq and Flexpert-3D are carefully evaluated on ATLAS and mdCATH, showing favorable correlations to MD against several baselines. The authors also ship code and trained weights, which supports reproducibility. However, the central claim about steering inverse folding toward increased physical flexibility is currently supported only by the same surrogate model used to create the training signal, and the paper's own appendices reveal a likely positive bias and structural-destabilization confound. These issues are load-bearing for the paper's main advertised capability, so the design claim requires additional independent validation or a substantially more cautious framing.

major comments (3)
  1. [Section 4.2, Eqs. (1)-(3) and Section 5.2, Eq. (6)] The evaluation of Flexpert-Design is circular: the pseudolabels used to construct training instructions (Eq. 1), the flexibility-matching loss (Eq. 2), and the enrichment metric (Eq. 6) all use the same Flexpert-3D model. The reported median enrichment ratio of 1.52 therefore demonstrates that the fine-tuned model produces sequences that Flexpert-3D scores as more flexible, not necessarily that the sequences are physically more flexible. This concern is amplified by the vanilla ProteinMPNN baseline, which also yields 61% flexibility-increasing mutations under the same metric, and by the authors' acknowledgment in Appendix K that Flexpert-3D 'might be slightly biased toward overpredicting flexibility since it predicts a flexibility increase even for the vanilla ProteinMPNN model.' To support the central claim in Section 1 that inverse folding can be steered toward increased protein flexibility, the paper needs independent validation of designed sequences, such as MD simulations on a subset of designs, or agreement with a structurally orthogonal flexibility measure (e.g., ANM fluctuations on the designed structures or experimental B-factors).
  2. [Section 5.2 and Appendix I, Table 9] The structure preservation analysis raises a serious confound. Flexpert-Design sequences have substantially higher Cα RMSD to the ground-truth backbone (3.36 Å vs 2.00 Å for ProteinMPNN) and a pronounced pLDDT drop in the engineered region (0.58 vs 0.81). This pattern is consistent with local unfolding or misfolding rather than native-like enhanced dynamics, which would confound the interpretation of the enrichment ratio. The authors should either stratify the evaluation by structural integrity (e.g., pLDDT or RMSD thresholds) to show that the flexibility increase is not an artifact of destabilization, or substantially temper the claim that the method 'engineers flexibility' in the sense of native dynamics rather than partially unfolding the engineered region.
  3. [Section 5.2, Table 4] The median enrichment ratios and proportions are reported without confidence intervals or significance tests. Given the small difference between the full Flexpert-Design model (1.52) and its without-loss variant (1.43), and the high baseline proportion for ProteinMPNN (61% flexibility-increasing mutations), it is unclear whether the differences are statistically meaningful. The authors should report bootstrap confidence intervals across CATH4.3 test proteins or paired per-protein tests, and should also discuss whether the improvement over the no-loss variant justifies the additional complexity of the flexibility-matching loss.
minor comments (5)
  1. [Abstract and Section 1 vs. Section 6] The abstract and introduction state that the method 'demonstrate[s] that inverse folding models can be steered toward' increased flexibility without the caveat 'as measured by our protein flexibility predictor,' which appears only in the conclusion. Please align these statements to avoid overclaiming in the abstract.
  2. [Section 3.2, Table 1] Please specify why 7 of the 1390 ATLAS proteins were skipped, since the text says 'some were skipped due to missing pieces of structure resulting in NaNs from ENMs' but does not quantify the number or give criteria.
  3. [Section 4.2, Eq. (3) and surrounding text] The notation 'LF lexpert' is inconsistent and appears with various spacing; please use a single consistent subscript, e.g., L_flex, throughout the paper.
  4. [Section 5.1] The term 'topology splitting' is used without definition or reference; please provide details in the experimental setup or point to an appendix that explains how topologies are split and how leakage is prevented.
  5. [Appendix J, Figure 8] The right panel of Figure 8, which shows the effect of segment length, does not have labeled axes in the text description; please clarify the x-axis and y-axis in the caption.

Circularity Check

1 steps flagged · score 7.0 of 10

The 1.52x flexibility enrichment is measured with the same Flexpert-3D model used to generate the training pseudolabels and to define the training loss, so the design claim is largely a self-consistency check of the surrogate.

  1. fitted input called prediction [Sections 4.2 and 5.2, Eqs. (1), (2), (4), (6); acknowledged in Appendix K]
    "First, we construct the set F native of flexibility pseudolabels obtained with Flexpert-3D as F native = {f native i | f native i = F3D(si, χi), (si, χi) ∈ Dbb, i ∈ D} ... we define the flexibility enrichment ratio rij: rij = F3D(ŝj, χj)i / f native_ij"

    Eq. (1) makes the training instructions Flexpert-3D outputs; Eq. (2) trains PF by regressing F3D(ŝ,χ) onto those instructions; Eq. (6) evaluates the design by dividing F3D(ŝ,χ) by the same native F3D value. The reported enrichment therefore measures how well the fine-tuned model implements the learned F3D-conditional mapping, not whether the designed protein is physically more flexible. Appendix K concedes that the predictor 'might be slightly biased toward overpredicting flexibility since it predicts a flexibility increase even for the vanilla ProteinMPNN model,' and Table 4 shows vanilla ProteinMPNN already gives enrichment 1.07 with 61% flexibility-increasing mutations. Hence the 1.52x median enrichment is partly self-consistency of the fitted surrogate.

full rationale

The central circular chain is concrete: native flexibility instructions are generated by Flexpert-3D (Eq. 1); the fine-tuning loss asks PF to produce sequences whose Flexpert-3D flexibility matches the input instruction (Eq. 2); and the success metric is the ratio of Flexpert-3D flexibility of the designed sequence to the Flexpert-3D native value (Eq. 6). Thus the design "prediction" is evaluated with the same fitted model whose outputs were used as training targets. This is not a mere self-citation issue; it is an evaluation-by-construction issue. The paper explicitly limits its conclusion to "increased flexibility as measured by our protein flexibility predictor" and discloses in Appendix K that Flexpert-3D "might be slightly biased toward overpredicting flexibility since it predicts a flexibility increase even for the vanilla ProteinMPNN model," consistent with the vanilla baseline's 1.07 enrichment and 61% flexibility-increasing mutations in Table 4. Appendix I adds that Flexpert-Design sequences have higher RMSD (3.36 Å vs 2.00 Å) and lower pLDDT in the engineered region (0.58 vs 0.81), so at least part of the predicted flexibility increase may reflect local structural degradation rather than native-like enhanced dynamics. These limitations do not invalidate the flexibility-prediction contribution, because Flexpert-Seq and Flexpert-3D are benchmarked against MD on held-out ATLAS proteins and on mdCATH; that part of the paper is self-contained. However, the design-engineering claim, which is the paper's headline result, reduces in large part to optimizing and then reading out the same surrogate, so the circularity score is high but not maximal.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The quantitative pipeline imports heavy machinery (ProtTrans, ProteinMPNN, ATLAS, ProDy) from prior work and treats MD RMSF as ground truth. The design contribution rests on hyperparameters theta, tau, and |S|, and on the self-referential use of Flexpert-3D for both generation and evaluation. No new physical entity is introduced.

free parameters (4)
  • loss mixing weight theta = 0.8
    Chosen by hand in Appendix H to balance sequence recovery against flexibility loss; central to Flexpert-Design training.
  • flexibility increment tau = 5 for increase, -10 for decrease
    Chosen for the main design experiments; Appendix J shows sensitivity and the main result uses tau=5.
  • engineered segment length |S| = 50 residues
    Chosen for the main design experiments; Figure 8 tests other lengths.
  • ENM cutoff distance = 16 Angstrom
    Used to build Hessian and Kirchhoff matrices for GNM and ANM baselines in Appendix B; a standard but hand-chosen distance.
assumptions (5)
  • domain assumption MD-derived RMSF from ATLAS is the gold standard for protein flexibility
    Section 3.1 states that MD simulations are regarded as the gold standard; this choice defines the learning target and the upper bound of 0.88.
  • domain assumption ProteinMPNN does not change the backbone structure when a sequence is predicted
    Explicitly stated in the Figure 2 caption and used to match predicted sequences with ground-truth backbones in Flexpert-3D evaluation.
  • domain assumption Topology-based splitting prevents leakage between ATLAS training and CATH4.3 test evaluations
    Section 5.1 assumes that excluding CATH4.3 test topologies from ATLAS training avoids leakage in the Flexpert-Design pipeline.
  • standard math Gumbel-Softmax straight-through estimator provides usable gradients through the discrete protein language model input
    Section 4.2 relies on this approximation; it is standard but an approximation.
  • domain assumption The ATLAS 100 ns, 300 K MD trajectories represent flexibility of full proteins
    The dataset is used as ground truth, and replica correlation 0.88 is treated as an upper bound rather than a validation of physical fidelity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning to engineer protein flexibility." pith.science (2026). https://pith.science/paper/6WM6MDVM

@misc{pith2026241218275,
  author       = {Pith},
  title        = {Pith review of: Learning to engineer protein flexibility},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6WM6MDVM}},
  note         = {Machine review of arXiv:2412.18275}
}
read the original abstract

Generative machine learning models are increasingly being used to design novel proteins for therapeutic and biotechnological applications. However, the current methods mostly focus on the design of proteins with a fixed backbone structure, which leads to their limited ability to account for protein flexibility, one of the crucial properties for protein function. Learning to engineer protein flexibility is problematic because the available data are scarce, heterogeneous, and costly to obtain using computational as well as experimental methods. Our contributions to address this problem are three-fold. First, we comprehensively compare methods for quantifying protein flexibility and identify data relevant to learning. Second, we design and train flexibility predictors utilizing sequential or both sequential and structural information on the input. We overcome the data scarcity issue by leveraging a pre-trained protein language model. Third, we introduce a method for fine-tuning a protein inverse folding model to steer it toward desired flexibility in specified regions. We demonstrate that our method Flexpert-Design enables guidance of inverse folding models toward increased flexibility. This opens up new possibilities for protein flexibility engineering and the development of proteins with enhanced biological activities.

Figures

Figures reproduced from arXiv: 2412.18275 by the authors.

Figure 1
Figure 1. Overview of the architecture of our protein flex [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Overview of the Flexpert-Design method for fine-tuning of inverse folding models toward [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Comparison of the enrichment ratio distributions (left) and the amino acid type distribu [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: ATLAS datapoint 5jca S. Flexpert-Seq and Flexpert-3D seem to improve over the predic￾tion of ANMs mainly by correctly capturing the higher flexibility of the helix at the terminus. set, this appears as a reasonable generalization. We also notice that Flexpert-3D experi…
Figure 5
Figure 5. Figure 5: ATLAS datapoint 1c52 A. Anisotropic Network Models perform relatively poorly, which likely propagates to Flexpert-3D causing it to underperform (Pearson R = 0.77) with respect to Flexpert-Seq (Pearson R = 0.8) [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: ATLAS datapoint 6o2v A. This is a failure case for Flexpert-Seq (Pearson R = 0.28), which predicts one beta strand together with the terminal coil to be more flexible than what was observed in MD simulation (see RMSF subfigure). Anisotropic Network Models perform bette…
Figure 7
Figure 7. Figure 7: ATLAS datapoint 1egw B. This datapoint features relatively high variance in the flexibil￾ity, posing a challenge for Flexpert-Seq (Pearson R = 0.75). The Anisotropic Network Models work better (Pearson R=0.87) and presumably help Flexpert-3D to excel (Pearson R=0.92) …
Figure 8
Figure 8. Figure 8: The effect of the flexibility increasing parameter [PITH_FULL_IMAGE:figures/full_fig_p022_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 13 canonical work pages

  1. [5]

    Language models of protein sequences at the scale of evolution enable accurate structure prediction.bioRxiv, pp

    Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos San- tos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, and Alexander Rives. Language models of protein sequences at the scale of evolution enable accurate structure prediction.bioRxiv, pp. 2022.07.20.500902, July

    work page 2022

  2. [6]

    doi: 10.1038/s41592-022-01488-1

    ISSN 1548-7105. doi: 10.1038/s41592-022-01488-1. URL https://doi.org/ 10.1038/s41592-022-01488-1 . Gwendolyn Moise, Yalemi Morales, Victor Beaumont, Timothy Caradonna, J Patrick Loria, Sean J Johnson, and Alvan C Hengge. A YopH PTP1B chimera shows the importance of the WPD-loop sequence to the activity, structure, and dynamics of protein tyrosine phosphat...

    doi:10.1038/s41592-022-01488-1

  3. [7]

    Full data

    Table 7: Evaluation of Flexpert-3D and Flexpert-Seq predictors on MD simulations from the md- CATH dataset. Both predictors were evaluated on the full dataset (5398 proteins, columns “Full data”) as well as on its subset excluding topologies present in the ATLAS training set (4013 pro- teins, columns “Topo. filtered”). The reported numbers are Pearson cor...

    work page 2025

  4. [8]

    Large amplitude elastic motions in proteins from a single-parameter, atomic analysis

    14 Published as a conference paper at ICLR 2025 M M Tirion. Large amplitude elastic motions in proteins from a single-parameter, atomic analysis. Phys. Rev. Lett., 77(9):1905–1908, August

    work page 2025

  5. [9]

    We see that the Flexpert-Design generated sequences show lower recovery of the ground truth struc- ture compared to ProteinMPNN, which could suggest that Flexpert-Design produces less stable structures or that it introduces notable conformational changes. A particularly significant drop in 21 Published as a conference paper at ICLR 2025 Table 9: Evaluatio...

    work page 2025

  6. [10]

    Alexander Wlodawer, Wladek Minor, Zbigniew Dauter, and Mariusz Jaskolski

    URLhttps: //arxiv.org/abs/2312.00080. Alexander Wlodawer, Wladek Minor, Zbigniew Dauter, and Mariusz Jaskolski. Protein crystallog- raphy for non-crystallographers, or how to get the best (but not more) from published macro- molecular structures. FEBS J., 275(1):1–21, January

    arXiv

  7. [13]

    and DynaMut (Rodrigues et al., 2021). D T HE DIFFICULTY OF LEARNING FROM B- FACTORS COMPARED TO MD TRAJECTORIES We also investigated the potential of B-factors to be used as data for learning to predict protein flexibility. To this end, we retrained Flexpert-Seq using B-factors instead of the RMSFs from MD trajectories. Because the absolute values of B-fa...

    work page 2021

  8. [15]

    effectively normalize

    Therefore, we selected θ = 0.8 as the parameter that still sufficiently maintains the original performance of the vanilla ProteinMPNN. When balancing the LF lexpertloss, we also tried to “effectively normalize” the LF lexand LSeq losses to have better control with the parameter θ. This we did by normalization of the gradients in the backward pass and then...

    work page 2025

Show all 17 references
  1. [17]

    22 Published as a conference paper at ICLR 2025 Table 10: Comparing the ability of flexibility engineering models to decrease the flexibility of engi- neered segments. The presented median enrichment ratiosM ed(rij) and the proportion of flexibility decreasing mutations were o...

    2025

  2. [2011]

    package to compute the ENMs and the corresponding RMSF values. Firstly, we used ProDy to build the Hessian matrix, based on the α-carbon coordinates of our input structure and the cutoff of 16 Angstrom, then the non-trivial modes of the Hessian were calculated, and consequentl...

    2025

  3. [2014]

    Zhangyang Gao, Cheng Tan, and Stan Z. Li. Pifold: Toward effective and efficient protein inverse folding. In International Conference on Learning Representations , 2023a. URL https:// openreview.net/forum?id=oMsN9TYwJ0j. Zhangyang Gao, Cheng Tan, and Stan Z. Li. Knowledge-desi...

    2023 arXiv

  4. [2018]

    Backbone resonance assignments of an artificially engineered TEM-1/PSE-4 class a β- lactamase chimera

    13 Published as a conference paper at ICLR 2025 S´ebastien Morin, Christopher M Clouthier, Sophie Gobeil, Joelle N Pelletier, and St ´ephane M Gagn´e. Backbone resonance assignments of an artificially engineered TEM-1/PSE-4 class a β- lactamase chimera. Biomol. NMR Assign., 4(...

    2025

  5. [2019]

    Eric Jang, Shixiang Gu, and Ben Poole

    URL https://proceedings.neurips.cc/paper_files/ paper/2019/file/f3a4ff4839c56a5f460c88cce3666a2b-Paper.pdf. Eric Jang, Shixiang Gu, and Ben Poole. Categorical reparameterization with gumbel-softmax. In5th International Conference on Learning Representations, ICLR 2017, Toulon,...

    2019

  6. [2021]

    As a product of these measurements, atom-to-atom distance restraints can be inferred

    15 Published as a conference paper at ICLR 2025 APPENDIX A E XPERIMENTAL MEASUREMENT OF PROTEIN FLEXIBILITY BY NMR AND HDX-MS In contrast to crystallographic methods directly obtaining the B-factors (temperature factors), NMR reads chemical shift perturbations, in other words,...

    2025

  7. [2022]

    ProtTrans: Toward understanding the language of life through self-supervised learning

    11 Published as a conference paper at ICLR 2025 Ahmed Elnaggar, Michael Heinzinger, Christian Dallago, Ghalia Rehawi, Yu Wang, Llion Jones, Tom Gibbs, Tamas Feher, Christoph Angerer, Martin Steinegger, Debsindhu Bhowmik, and Burkhard Rost. ProtTrans: Toward understanding the l...

    2025

  8. [2023]

    doi: 10.1093/nar/gkad1084

    ISSN 0305-1048. doi: 10.1093/nar/gkad1084. URL https://doi.org/10.1093/nar/gkad1084. Chuanrui Wang, Bozitao Zhong, Zuobai Zhang, Narendra Chaudhary, Sanchit Misra, and Jian Tang. Pdb-struct: A comprehensive benchmark for structure-based protein design,

    doi

  9. [2024]

    URL https://proceedings.mlr.press/ v235/jing24a.html. 12 Published as a conference paper at ICLR 2025 John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin ˇZ´ıdek, Anna Potapenko, Alex Bridg...

    2025

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.