Pith. sign in

REVIEW 3 major objections 5 minor 35 references

BindEnergyCraft: Casting Protein Structure Predictors as Energy-Based Models for Binder Design

T0 review · 3 major / 5 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 This paper claims that replacing the ipTM objective in a hallucination-based binder design pipeline with pTMEnergy, a log-sum-exp energy extracted from AlphaFold's predicted alignment error logits, improves in silico binder success rates…

desk verdict Useful in silico objective and screening score, but the JEM likelihood framing is overclaimed and should be toned down before publication. read the letter →

arxiv 2505.21241 v1 pith:C2M7W5GZ submitted 2025-05-27 cs.LG

Divya Nori , Anisha Parsan , Caroline Uhler , Wengong Jin This is my paper
classification cs.LG
keywords proteinbinderdesignenergy-basedmodelspTMEnergypredictedalignmenterrorhallucination-basedvirtualscreeningbindinglikelihoodRNAaptamers
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 sets out to show that the confidence outputs of a protein structure predictor can be read as an energy function over binder–target complexes, and that using this energy as an optimization objective improves computational binder design. The authors define pTMEnergy by taking a log-sum-exp over AlphaFold2's predicted alignment error (pAE) logits for each inter-chain residue pair, weighting the error bins by a TM-score-like kernel, and averaging across the interface. They embed this energy into the BindCraft hallucination pipeline as a direct replacement for the ipTM term, calling the result BECraft. Across eight targets, BECraft reports higher in silico success rates than BindCraft, RFDiffusion, and ESM3 under both Rosetta-based and folding-model constraints, while producing fewer relaxed complexes with atomic clashes. The same energy also ranks miniprotein binders and RNA aptamers better than ipTM and several physics-based or supervised baselines, which matters because it points toward likelihood-based, unsupervised design objectives.

What carries the argument

The load-bearing object is pTMEnergy (Eq. 8), an energy function assembled from the pAE logits of a folding model. For each pair of residues on different chains, the model outputs logits $\ell_{ijb}$ over distance-error bins; pTMEnergy applies the log-sum-exp operation to those logits after scaling each bin by the kernel $g(d_b) = 1/(1+(d_b/d_0(N))^2)$, then averages over the interface with a $-1/|I|$ factor. The kernel is the same weighting used in the pTM score, so it up-weights low alignment errors and thereby encodes the physical prior that accurate local packing is more likely. The averaging over all inter-chain pairs is what makes the gradients dense; the log-sum-exp is what gives the energy its likelihood-like interpretation, since any classifier's logits define an energy by $E(x) = -\log \sum_y \exp(f(x)[y])$.

What would settle it

One decisive check would be experimental: synthesize and assay BECraft and BindCraft designs for the same targets and compare hit rates; if pTMEnergy-optimized binders do not bind at least as often as ipTM-optimized ones, the central claim about better binder design fails. A cheaper computational check is to rank a held-out set of complexes with measured Kd values using pTMEnergy and see whether the energy orders affinities correctly.

Watch

Extended reading notes

Core claim

The paper's central claim is that per-residue predicted alignment error logits can be converted, through the standard log-sum-exp identity for classifier energies, into a continuous per-complex energy that behaves like a binding likelihood. The specific object is $E_{\mathrm{pTMEnergy}}(x) = -\frac{1}{|I|} \sum_{(i,j)\in I} \log \sum_b g(d_b) \exp(\ell_{ijb})$, where $\ell_{ijb}$ are the pAE bin logits for residue pair $(i,j)$, $g(d_b)$ damps high-error bins, and $I$ is the set of inter-chain pairs. Because every interface pair contributes to the sum, gradients reach nearly all target residues, in contrast to ipTM's hard maximum, which restricts signal to a single reference residue. In the paper's experiments, replacing ipTM with this energy in the BindCraft loss raises the share of designs passing independent Rosetta interface criteria and folding-model thresholds, lowers clash rates after relaxation, and yields the best unsupervised ranking scores in miniprotein and RNA aptamer screens.

Load-bearing premise

The argument depends on treating the many pairwise error-bin predictions as if they were one statistically sound classifier for the whole complex, with independent bins and a single normalization constant; if that joint-classifier story fails, pTMEnergy is a useful heuristic but not a true binding likelihood.

Editorial extensions

If this is right

  • Any pipeline that already computes ipTM from pAE logits can swap in pTMEnergy at no extra model call, since it uses the same network outputs.
  • Hallucination-based design no longer needs to rely on a bounded heuristic score; the objective is a continuous energy with a probabilistic reading.
  • Dense interface gradients should make optimization less prone to getting stuck maximizing a single residue pair's alignment.
  • The same energy can rank candidate complexes without labeled binding data, reducing reliance on supervised affinity predictors.
  • The aptamer screening results suggest the energy transfers to protein–RNA interactions, so future predictors with nucleic-acid-aware confidence heads could use pTMEnergy for mixed-molecule design.

Reading between the lines

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

  • The paper does not demonstrate that pTMEnergy's normalization constant is constant across sequences; if it varies, the quantity is best understood as a heuristic objective whose empirical ranking power is what matters, not a normalized likelihood.
  • A natural testable extension would be calibrating the kernel $g(d_b)$ on measured binding affinities to see whether the energy's ranking improves with a learned weighting.
  • Because pTMEnergy only needs pairwise error logits, it should transfer to any structure predictor with a pAE-like head; whether it helps design, not just screen, RNA aptamers is an open question the paper leaves for future work.
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 proposes pTMEnergy, a scoring function obtained by applying the Joint Energy-based Model (JEM) LogSumExp trick to the per-residue-pair pAE logits of AlphaFold2, with a TM-score-like kernel weighting and an average over inter-chain pairs. This energy is used as a drop-in replacement for ipTM in the BindCraft hallucination pipeline, yielding BECraft. The authors report improved in silico design success rates on eight targets under both Rosetta-based and AlphaFold-confidence-based constraints, reduced atomic clashes, denser gradients than ipTM, and improved virtual screening performance for miniprotein and RNA aptamer binders.

Significance. If the empirical results hold, pTMEnergy is a practically useful dense gradient objective for binder hallucination and a competitive unsupervised scoring function for virtual screening. The paper's strengths include evaluation against multiple strong baselines (BindCraft variants, RFDiffusion+ProteinMPNN, ESM3), the use of independent Rosetta-based interface constraints in addition to folding-model filters, a clash analysis after relaxation, and screening tasks grounded in experimentally characterized binders. The method adds no computational overhead because it reuses pAE outputs. However, the central theoretical claim that pTMEnergy is a statistical likelihood under AlphaFold's learned distribution is not supported by the derivation in Section 3.2; as written, Eq. (8) defines a heuristic, TM-weighted scoring function. This does not necessarily invalidate the empirical contributions, but it requires a substantial reframing of the paper's motivation and claims.

major comments (3)
  1. [Section 3.2, Eq. (8)] The JEM derivation does not justify pTMEnergy as a likelihood. JEM applies to the logits of a single classifier over a label space; the pAE head is a set of per-residue-pair classifiers, and Eq. (7) is already a per-pair LogSumExp energy. Eq. (8) then replaces exp(ell_ijb) with g(d_b) exp(ell_ijb), which is exactly the LogSumExp of modified logits ell_ijb + log g(d_b). Those modified logits are not AlphaFold's confidence outputs, and the 1/|I| normalization changes the scaling that a true log-likelihood would have. Thus pTMEnergy is a hand-crafted, TM-score-weighted variant of the per-pair LogSumExp objective, not a likelihood under the model's learned distribution. This undermines the paper's central 'principled probabilistic interpretation' claim and should be either corrected with a real derivation or explicitly reframed as a heuristic energy with empirical justification.
  2. [Section 3.2, Eq. (8)] Even if the kernel weighting were removed, averaging per-pair energies over the interface assumes that the residue pairs are conditionally independent and share a single partition function. No such argument is provided. A proper marginal likelihood over a complex would have to account for dependencies between residue pairs and for a normalization constant that does not factor per pair. Without this, the phrase 'statistical energy function' and 'likelihood of a folded complex' overstate what Eq. (8) provides. The authors should either justify these assumptions or describe pTMEnergy as an empirical scoring function whose probabilistic interpretation is only heuristic.
  3. [Table 1] The folding-model constraint results are partly circular for BECraft and BindCraft because the success criteria (pLDDT > 0.8, ipTM > 0.5, pTM > 0.45, interface pAE < 0.4) are computed from the same AlphaFold2 confidence outputs that BECraft's objective is a function of, and the paper's own pTMEnergy is derived from pAE logits. The authors acknowledge this risk in the text, but the abstract and results sections still report success 'across both' constraint sets without this caveat. Since the Rosetta constraints are independent and do show improvements, the central empirical claim survives, but the presentation should clearly label the folding-model results as a self-consistency check and rest the headline success claim on the Rosetta-based evaluation.
minor comments (5)
  1. [Section 4.1] The table reports binomial standard errors but no pairwise significance tests; several comparisons, such as ALK under Rosetta constraints (BECraft 0.16 +/- 0.04 versus BindCraft 0.14 +/- 0.04), are within one standard error. Adding a paired test or explicit confidence intervals would make the 'consistently highest' claim more precise.
  2. [Algorithm 1] The pTMEnergy loss weight is set to 0.05 because that was the ipTM weight in BindCraft, but pTMEnergy has different units and scale (a negative log-sum over bins, averaged over pairs). The choice should be justified, or at least a sensitivity analysis should be reported.
  3. [Section 5] The limitations paragraph appropriately notes that results are purely computational and that the model inherits confidence-calibration failures, but it does not mention that Eq. (8) departs from the JEM derivation by inserting the g(d_b) kernel and averaging over pairs. This caveat should be added so readers are not misled about the probabilistic status of pTMEnergy.
  4. [Section 4.3] The gradient analysis is informative, but the text says gradients were 'computed but not used for optimization' at steps 0, 100, and 200 without specifying which design iteration or how many independent runs. Clarifying the protocol would improve reproducibility.
  5. [Section 4.4] The supervised baseline is described as a feed-forward network on ESM-2 3B embeddings; the table label 'Supervised Model' is vague. Reporting the architecture size and training details would help the comparison.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: pTMEnergy is a designed scoring objective; its empirical claims are supported by independent Rosetta constraints and ground-truth screening, and the JEM-likelihood wording is an overclaim rather than a circular dependency.

full rationale

The paper's main optimization objective, pTMEnergy (Eq. 8), is constructed directly from AlphaFold2 pAE logits and the pTM kernel g(d_b), with no free parameters fitted to the success metrics. BECraft's design gains are therefore not a fitted input relabeled as a prediction. The primary in silico evidence in Table 1 includes Rosetta-based constraints (shape complementarity, dSASA, H-bonds, hydrophobicity), which do not use the AlphaFold confidence outputs being optimized; the paper explicitly states it separates these because 'folding model metrics may advantage methods like BindCraft and BECraft, which optimize confidence scores during design' (Section 4.1). Tables 3 and 4 are retrospective virtual screening against experimentally curated positives (Bennett et al.; Huang et al.), so ranking performance is ground-truth based. The self-citations (DSMBind, FAFormer, RNAflow) appear only as baselines or related work and are not load-bearing premises. The genuine weakness is the probabilistic framing: Eq. 7 correctly applies JEM to per-pair pAE logits, but Eq. 8 replaces exp(l_ijb) with g(d_b) exp(l_ijb), which changes the effective logits; hence pTMEnergy is not exactly the JEM energy of AlphaFold's pAE classifier. That is a derivation overclaim, and the Limitations section concedes results are purely computational, but it is a correctness gap, not circular reasoning: none of the paper's empirical conclusions presuppose the validity of the likelihood interpretation. Thus no circular step meeting the quoted-evidence standard is present.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central method has no fitted parameters: bin centers, d0(N), and the loss weights come from AlphaFold and BindCraft rather than from fitting. The main load-bearing assumptions are the JEM-style mapping, pairwise independence, the ad hoc pTM kernel weighting, and the dropped partition function.

free parameters (2)
  • pTMEnergy loss weight alpha in BECraft loss = 0.05
    Inherited from BindCraft's ipTM weight; not optimized in this paper, and the design results depend on the full loss balance.
  • BindCraft auxiliary loss weights (pLDDT 0.1, pAE 0.4/0.1, contacts 1.0, radius of gyration 0.3) = 0.1, 0.4, 0.1, 1.0, 0.3
    Taken unchanged from BindCraft and used in BECraft; chosen by hand in the original pipeline, not fitted to the present data.
assumptions (4)
  • domain assumption JEM claim that classifier logits define an energy over inputs via logsumexp marginalization (Eq. 3-4).
    Invoked in Section 3.1 to reinterpret AlphaFold pAE logits as energies; requires treating per-pair logits as parts of one joint classifier over the complex.
  • domain assumption Conditional independence of residue-pair error-bin predictions.
    Needed so that the global energy in Eq. 8 is a sum of per-pair logsumexp terms; not stated or justified in the paper.
  • ad hoc to paper pTM kernel g(d_b) weighting improves the energy's physical plausibility.
    Introduced in Eq. 6-8 to up-weight low alignment errors; it is a heuristic added to the logsumexp, not derived from the structure predictor.
  • standard math The EBM partition function Z(theta) is constant across complexes and can be ignored for ranking and optimization.
    Standard in EBM practice for fixed model parameters; the paper does not discuss it but relies on it implicitly.

how reviews work

0 comments
Cite this review

Pith. "Pith review of BindEnergyCraft: Casting Protein Structure Predictors as Energy-Based Models for Binder Design." pith.science (2026). https://pith.science/paper/C2M7W5GZ

@misc{pith2026250521241,
  author       = {Pith},
  title        = {Pith review of: BindEnergyCraft: Casting Protein Structure Predictors as Energy-Based Models for Binder Design},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C2M7W5GZ}},
  note         = {Machine review of arXiv:2505.21241}
}
read the original abstract

Protein binder design has been transformed by hallucination-based methods that optimize structure prediction confidence metrics, such as the interface predicted TM-score (ipTM), via backpropagation. However, these metrics do not reflect the statistical likelihood of a binder-target complex under the learned distribution and yield sparse gradients for optimization. In this work, we propose a method to extract such likelihoods from structure predictors by reinterpreting their confidence outputs as an energy-based model (EBM). By leveraging the Joint Energy-based Modeling (JEM) framework, we introduce pTMEnergy, a statistical energy function derived from predicted inter-residue error distributions. We incorporate pTMEnergy into BindEnergyCraft (BECraft), a design pipeline that maintains the same optimization framework as BindCraft but replaces ipTM with our energy-based objective. BECraft outperforms BindCraft, RFDiffusion, and ESM3 across multiple challenging targets, achieving higher in silico binder success rates while reducing structural clashes. Furthermore, pTMEnergy establishes a new state-of-the-art in structure-based virtual screening tasks for miniprotein and RNA aptamer binders.

Figures

Figures reproduced from arXiv: 2505.21241 by the authors.

Figure 1
Figure 1. A: BindEnergyCraft (BECraft) optimizes binder sequences by backpropagating pTMEnergy, computed from pAE logits output by AlphaFold2-Multimer. B: Gradients from ipTM are sparse across interface residue pairs, since the maximum over target residue indices zeroes out the gradient from all but one position. pTMEnergy preserves gradients across the interface. We incorporate pTMEnergy into the BindCraft pipeline to create… view at source ↗
Figure 2
Figure 2. (A) A VirB8-binding complex designed using ipTM exhibits severe atomic clashes at the interface. (B) The corresponding pTMEnergy-optimized design for VirB8 forms a clash-free interface. (C) An ipTM-optimized binder for ALK with substantial steric overlap. (D) The pTMEnergy￾optimized design for ALK forms a well-packed, physically realistic interaction [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. (A–B): Magnitude of maximum gradient error bin for each binder–target residue pair (binder on y, target on x), normalized to [0, 1], for ipTM and pTMEnergy respectively on an IL2Ra design task. (C–D): Frequency with which each target residue ranks in the top-10 by gradient magnitude across all iterations, for ipTM and pTMEnergy respectively. Figure 3A and 3B show the maximum gradient bin magnitude for each binder–ta… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Distribution of predicted pTMEnergy scores. Scores are negated so that higher values [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

35 extracted references · 13 canonical work pages

  1. [1]

    Accurate structure prediction of biomolecular interactions with alphafold 3.Nature, pages 1–3, 2024

    Josh Abramson, Jonas Adler, Jack Dunger, Richard Evans, Tim Green, Alexander Pritzel, Olaf Ronneberger, Lindsay Willmore, Andrew J Ballard, Joshua Bambrick, et al. Accurate structure prediction of biomolecular interactions with alphafold 3.Nature, pages 1–3, 2024

    2024

  2. [2]

    The rosetta all-atom energy function for macromolecular modeling and design

    Rebecca F Alford, Andrew Leaver-Fay, Jeliazko R Jeliazkov, Matthew J O’Meara, Frank P DiMaio, Hahnbeom Park, Maxim V Shapovalov, P Douglas Renfrew, Vikram K Mulligan, Kalli Kappel, et al. The rosetta all-atom energy function for macromolecular modeling and design. Journal of chemical theory and computation, 13(6):3031–3048, 2017

    work page 2017

  3. [3]

    Accurate prediction of protein–nucleic acid complexes using rosettafoldna.Nature methods, 21(1):117–121, 2024

    Minkyung Baek, Ryan McHugh, Ivan Anishchenko, Hanlun Jiang, David Baker, and Frank DiMaio. Accurate prediction of protein–nucleic acid complexes using rosettafoldna.Nature methods, 21(1):117–121, 2024

    work page 2024

  4. [4]

    Improving de novo protein binder design with deep learning.Nature Communications, 14(1):2625, 2023

    Nathaniel R Bennett, Brian Coventry, Inna Goreshnik, Buwei Huang, Aza Allen, Dionne Vafeados, Ying Po Peng, Justas Dauparas, Minkyung Baek, Lance Stewart, et al. Improving de novo protein binder design with deep learning.Nature Communications, 14(1):2625, 2023

    2023

  5. [5]

    Pyrosetta: a script-based interface for implementing molecular modeling algorithms using rosetta.Bioinformatics, 26(5):689–691, 2010

    Sidhartha Chaudhury, Sergey Lyskov, and Jeffrey J Gray. Pyrosetta: a script-based interface for implementing molecular modeling algorithms using rosetta.Bioinformatics, 26(5):689–691, 2010

    2010

  6. [6]

    Pepmlm: target sequence-conditioned generation of therapeutic peptide binders via span masked language modeling.ArXiv, pages arXiv–2310, 2024

    Tianlai Chen, Madeleine Dumas, Rio Watson, Sophia Vincoff, Christina Peng, Lin Zhao, Lauren Hong, Sarah Pertsemlidis, Mayumi Shaepers-Cheu, Tian Zi Wang, et al. Pepmlm: target sequence-conditioned generation of therapeutic peptide binders via span masked language modeling.ArXiv, pages arXiv–2310, 2024

    work page 2024

  7. [7]

    Massively parallel de novo protein design for targeted therapeutics.Nature, 550(7674):74–79, 2017

    Aaron Chevalier, Daniel-Adriano Silva, Gabriel J Rocklin, Derrick R Hicks, Renan Vergara, Patience Murapa, Steffen M Bernard, Lu Zhang, Kwok-Ho Lam, Guorui Yao, et al. Massively parallel de novo protein design for targeted therapeutics.Nature, 550(7674):74–79, 2017

    work page 2017

  8. [8]

    Boltzde- sign1: Inverting all-atom structure prediction model for generalized biomolecular binder design

    Yehlin Cho, Martin Pacesa, Zhidian Zhang, Bruno Correia, and Sergey Ovchinnikov. Boltzde- sign1: Inverting all-atom structure prediction model for generalized biomolecular binder design. bioRxiv, pages 2025–04, 2025

    work page 2025

Show all 35 references
  1. [9]

    Robust deep learning–based protein sequence design using proteinmpnn.Science, 378(6615):49–56, 2022

    Justas Dauparas, Ivan Anishchenko, Nathaniel Bennett, Hua Bai, Robert J Ragotte, Lukas F Milles, Basile IM Wicky, Alexis Courbet, Rob J de Haas, Neville Bethel, et al. Robust deep learning–based protein sequence design using proteinmpnn.Science, 378(6615):49–56, 2022

    2022

  2. [10]

    Protein complex prediction with alphafold-multimer.biorxiv, pages 2021–10, 2021

    Richard Evans, Michael O’Neill, Alexander Pritzel, Natasha Antropova, Andrew Senior, Tim Green, Augustin Žídek, Russ Bates, Sam Blackwell, Jason Yim, et al. Protein complex prediction with alphafold-multimer.biorxiv, pages 2021–10, 2021

    2021

  3. [11]

    Se (3)-transformers: 3d roto-translation equivariant attention networks.Advances in neural information processing systems, 33:1970–1981, 2020

    Fabian Fuchs, Daniel Worrall, V olker Fischer, and Max Welling. Se (3)-transformers: 3d roto-translation equivariant attention networks.Advances in neural information processing systems, 33:1970–1981, 2020

    1970

  4. [12]

    Computational protein design—the next generation tool to expand synthetic biology applications.Current opinion in biotechnology, 52: 145–152, 2018

    Pablo Gainza-Cirauqui and Bruno Emanuel Correia. Computational protein design—the next generation tool to expand synthetic biology applications.Current opinion in biotechnology, 52: 145–152, 2018. 10

    2018

  5. [13]

    Your classifier is secretly an energy based model and you should treat it like one.arXiv preprint arXiv:1912.03263, 2019

    Will Grathwohl, Kuan-Chieh Wang, Jörn-Henrik Jacobsen, David Duvenaud, Mohammad Norouzi, and Kevin Swersky. Your classifier is secretly an energy based model and you should treat it like one.arXiv preprint arXiv:1912.03263, 2019

    1912 arXiv

  6. [14]

    Simulating 500 million years of evolution with a language model.Science, page eads0018, 2025

    Thomas Hayes, Roshan Rao, Halil Akin, Nicholas J Sofroniew, Deniz Oktay, Zeming Lin, Robert Verkuil, Vincent Q Tran, Jonathan Deaton, Marius Wiggert, et al. Simulating 500 million years of evolution with a language model.Science, page eads0018, 2025

    2025

  7. [15]

    Protein-nucleic acid complex modeling with frame averaging transformer.arXiv preprint arXiv:2406.09586, 2024

    Tinglin Huang, Zhenqiao Song, Rex Ying, and Wengong Jin. Protein-nucleic acid complex modeling with frame averaging transformer.arXiv preprint arXiv:2406.09586, 2024

    2024 arXiv

  8. [16]

    Dsmbind: Se (3) denoising score matching for unsupervised binding energy prediction and nanobody design.bioRxiv, pages 2023–12, 2023

    Wengong Jin, Xun Chen, Amrita Vetticaden, Siranush Sarzikova, Raktima Raychowdhury, Caroline Uhler, and Nir Hacohen. Dsmbind: Se (3) denoising score matching for unsupervised binding energy prediction and nanobody design.bioRxiv, pages 2023–12, 2023

    2023

  9. [17]

    Learn- ing from protein structure with geometric vector perceptrons.arXiv preprint arXiv:2009.01411, 2020

    Bowen Jing, Stephan Eismann, Patricia Suriana, Raphael JL Townshend, and Ron Dror. Learn- ing from protein structure with geometric vector perceptrons.arXiv preprint arXiv:2009.01411, 2020

    2009 arXiv

  10. [18]

    Highly accurate protein structure prediction with alphafold.nature, 596(7873):583–589, 2021

    John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ron- neberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, et al. Highly accurate protein structure prediction with alphafold.nature, 596(7873):583–589, 2021

    2021

  11. [19]

    Full-atom peptide design based on multi-modal flow matching.arXiv preprint arXiv:2406.00735, 2024

    Jiahan Li, Chaoran Cheng, Zuofan Wu, Ruihan Guo, Shitong Luo, Zhizhou Ren, Jian Peng, and Jianzhu Ma. Full-atom peptide design based on multi-modal flow matching.arXiv preprint arXiv:2406.00735, 2024

    2024 arXiv

  12. [20]

    Equiformer: Equivariant graph attention transformer for 3d atomistic graphs.arXiv preprint arXiv:2206.11990, 2022

    Yi-Lun Liao and Tess Smidt. Equiformer: Equivariant graph attention transformer for 3d atomistic graphs.arXiv preprint arXiv:2206.11990, 2022

    2022 arXiv

  13. [21]

    Ppflow: Target-aware peptide design with torsional flow matching.bioRxiv, pages 2024–03, 2024

    Haitao Lin, Odin Zhang, Huifeng Zhao, Dejun Jiang, Lirong Wu, Zicheng Liu, Yufei Huang, and Stan Z Li. Ppflow: Target-aware peptide design with torsional flow matching.bioRxiv, pages 2024–03, 2024

    2024

  14. [22]

    Language models of protein sequences at the scale of evolution enable accurate structure prediction.BioRxiv, 2022: 500902, 2022

    Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, et al. Language models of protein sequences at the scale of evolution enable accurate structure prediction.BioRxiv, 2022: 500902, 2022

    2022

  15. [23]

    Language models enable zero-shot prediction of the effects of mutations on protein function.Advances in neural information processing systems, 34:29287–29303, 2021

    Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu, and Alex Rives. Language models enable zero-shot prediction of the effects of mutations on protein function.Advances in neural information processing systems, 34:29287–29303, 2021

    2021

  16. [24]

    Bill R Miller III, T Dwight McGee Jr, Jason M Swails, Nadine Homeyer, Holger Gohlke, and Adrian E Roitberg. Mmpbsa. py: an efficient program for end-state free energy calculations. Journal of chemical theory and computation, 8(9):3314–3321, 2012

    2012

  17. [25]

    Rnaflow: Rna structure & sequence design via inverse folding- based flow matching.arXiv preprint arXiv:2405.18768, 2024

    Divya Nori and Wengong Jin. Rnaflow: Rna structure & sequence design via inverse folding- based flow matching.arXiv preprint arXiv:2405.18768, 2024

    2024 arXiv

  18. [26]

    Bindcraft: one-shot design of functional protein binders.bioRxiv, pages 2024–09, 2024

    Martin Pacesa, Lennart Nickel, Christian Schellhaas, Joseph Schmidt, Ekaterina Pyatova, Lucas Kissling, Patrick Barendse, Jagrity Choudhury, Srajan Kapoor, Ana Alcaraz-Serna, et al. Bindcraft: one-shot design of functional protein binders.bioRxiv, pages 2024–09, 2024

    2024

  19. [27]

    Frame averaging for invariant and equivariant network design.arXiv preprint arXiv:2110.03336, 2021

    Omri Puny, Matan Atzmon, Heli Ben-Hamu, Ishan Misra, Aditya Grover, Edward J Smith, and Yaron Lipman. Frame averaging for invariant and equivariant network design.arXiv preprint arXiv:2110.03336, 2021

    2021 arXiv

  20. [28]

    E (n) equivariant graph neural networks

    Vıctor Garcia Satorras, Emiel Hoogeboom, and Max Welling. E (n) equivariant graph neural networks. InInternational conference on machine learning, pages 9323–9332. PMLR, 2021

    2021

  21. [29]

    The foldx web server: an online force field.Nucleic acids research, 33(suppl_2): W382–W388, 2005

    Joost Schymkowitz, Jesper Borg, Francois Stricher, Robby Nys, Frederic Rousseau, and Luis Serrano. The foldx web server: an online force field.Nucleic acids research, 33(suppl_2): W382–W388, 2005. 11

    2005

  22. [30]

    Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets.Nature biotechnology, 35(11):1026–1028, 2017

    Martin Steinegger and Johannes Söding. Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets.Nature biotechnology, 35(11):1026–1028, 2017

    2017

  23. [31]

    Attention is all you need.Advances in Neural Information Processing Systems, 2017

    A Vaswani. Attention is all you need.Advances in Neural Information Processing Systems, 2017

    2017

  24. [32]

    De novo design of protein structure and function with rfdiffusion.Nature, 620(7976):1089–1100, 2023

    Joseph L Watson, David Juergens, Nathaniel R Bennett, Brian L Trippe, Jason Yim, Helen E Eisenach, Woody Ahern, Andrew J Borst, Robert J Ragotte, Lukas F Milles, et al. De novo design of protein structure and function with rfdiffusion.Nature, 620(7976):1089–1100, 2023

    2023

  25. [33]

    Boltz-1: Democratizing biomolecular interaction modeling.bioRxiv, pages 2024–11, 2024

    Jeremy Wohlwend, Gabriele Corso, Saro Passaro, Mateo Reveiz, Ken Leidal, Wojtek Swiderski, Tally Portnoi, Itamar Chinn, Jacob Silterra, Tommi Jaakkola, et al. Boltz-1: Democratizing biomolecular interaction modeling.bioRxiv, pages 2024–11, 2024

    2024

  26. [34]

    Design of high-affinity binders to immune modulating receptors for cancer immunotherapy.Nature Communications, 16(1):2001, 2025

    Wei Yang, Derrick R Hicks, Agnidipta Ghosh, Tristin A Schwartze, Brian Conventry, Inna Goreshnik, Aza Allen, Samer F Halabiya, Chan Johng Kim, Cynthia S Hinck, et al. Design of high-affinity binders to immune modulating receptors for cancer immunotherapy.Nature Communications,...

    2001

  27. [35]

    De novo design of high-affinity protein binders with alphaproteo.arXiv preprint arXiv:2409.08022, 2024

    Vinicius Zambaldi, David La, Alexander E Chu, Harshnira Patani, Amy E Danson, Tristan OC Kwan, Thomas Frerix, Rosalia G Schneider, David Saxton, Ashok Thillaisundaram, et al. De novo design of high-affinity protein binders with alphaproteo.arXiv preprint arXiv:2409.08022, 2024...

    2024 arXiv

Pith tools

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