Pith. sign in

REVIEW 6 major objections 5 minor 44 references

Sequence-Only Prediction of Binding Affinity Changes: A Robust and Interpretable Model for Antibody Engineering

T0 review · 6 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read ProtAttBA predicts antibody–antigen binding affinity changes from protein sequences alone, without structural input.

desk verdict ProtAttBA is a clean sequence-only ΔΔG predictor with competitive numbers and a real robustness story, but it never states that mutations are restricted to the antibody chain, and if any antigen-side mutations are in S1131 or AB1101 the reported averages are partly for inputs that do not contain the mutation. read the letter →

arxiv 2505.20301 v1 pith:GYPWNJQA submitted 2025-05-14 q-bio.QM cs.LG

Chen Liu , Mingchen Li , Yang Tan , Wenrui Gou , Guisheng Fan , Bingxin Zhou This is my paper · ORCID
classification q-bio.QMcs.LG
keywords antibodyengineeringbindingaffinitychangepredictionsequence-onlymodelcross-attentionproteinlanguageΔΔGregressioninterpretabilityattentionweights
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 introduces ProtAttBA, a deep-learning model that predicts the change in antibody–antigen binding free energy ($\Delta\Delta G$) from protein sequences alone, with no structure input. Its central thesis is that sequence-only prediction can be competitive with structure-based methods in standard benchmarks and more dependable when complex structures are absent or predicted with low confidence. On three open benchmarks under cross-validation, sequence-identity splits, and mutation-depth splits, the best ProtAttBA variant reaches $R^2=0.69$ and Pearson correlation $0.84$ on a single-site benchmark, and it retains accuracy when structure-based baselines collapse after switching from crystal structures to predicted ones. If the claim holds, antibody engineers can rank mutant designs quickly from sequence and use the model's attention scores to choose which residues to test experimentally.

What carries the argument

The engine is a frozen pre-trained protein language model feeding a dual multi-head cross-attention block. In each block, one chain's representation is transformed into queries while the partner chain supplies keys and values, so residue-level information from the antigen can reshape the antibody representation and vice versa; rotary position embeddings preserve residue order. A 1D convolution with kernel size 1 gates the incoming residue features, and convolutional pooling compresses the attended matrices into chain-level vectors. The same machinery runs on wild-type and mutant pairs, and the final regression head concatenates the two pooled pairs before three fully connected layers. The cross-attention score matrices are also the interpretability output: the paper shows that high-attention residue pairs align with positions whose mutation disrupts hydrogen bonds.

What would settle it

Take any mutation in the benchmark data that changes the antigen rather than the antibody chain, and feed ProtAttBA the wild-type antigen sequence for both the wild-type and mutant complexes, as the architecture prescribes. If the model's predicted $\Delta\Delta G$ is unchanged or uncorrelated with the measured value for those cases, then the paper's sequence-only claim is limited to antibody-side mutations.

Watch

Extended reading notes

Core claim

ProtAttBA's core claim is that a quartet of sequences—wild-type antibody, wild-type antigen, mutant antibody, and again wild-type antigen—contains enough information to regress $\Delta\Delta G$ accurately, provided the chains are allowed to attend to each other. A frozen pretrained protein language model produces residue embeddings; a convolution gates local residue features; then symmetric multi-head cross-attention lets the antibody representation read from the antigen and vice versa, for both wild-type and mutant pairs. The pooled wild-type and mutant features are concatenated and passed through three fully connected layers to output $\Delta\Delta G$. On the single-site benchmark under cross-validation, the best variant reaches RMSE 1.31, $R^2=0.69$, PCC 0.84, and Spearman 0.75. This is competitive with the strongest structure-based method when that method receives crystal structures, and clearly better when that method is forced to use predicted structures instead. The paper's conclusion is that the bottleneck for practical binding-affinity prediction is often structural input quality—a bottleneck sequence-only models sidestep.

Load-bearing premise

The load-bearing assumption is that every mutation to be predicted lies on the antibody chain: the model is always given the wild-type antigen for both the wild-type and mutant complex, so an antigen-side mutation would be invisible to it and the prediction would be undefined.

Editorial extensions

If this is right

  • A researcher with only antibody and antigen sequences can obtain useful $\Delta\Delta G$ estimates without building or predicting a complex structure.
  • On multi-site mutations, the model still performs well, so it can rank multi-residue variants, not just single point mutants.
  • Because all four tested pretrained language model backbones give similar results, the cross-attention interaction module, not the choice of embedding model, appears to carry the predictive signal.
  • When structure-based methods are given predicted rather than experimental structures, their accuracy drops; ProtAttBA's stability suggests it is the safer choice in screening pipelines that lack crystallography.
  • The attention weights can serve as a residue-level map that points to positions worth targeted mutagenesis, giving a mechanistic handle on the model's predictions.

Reading between the lines

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

  • Editorial extension: because the current input already uses the wild-type antigen twice, extending ProtAttBA to predict antigen-side mutations would require adding a mutant-antigen input; the symmetric cross-attention design could plausibly support this without architectural changes, but the paper does not test it.
  • Editorial extension: attention mass on residue pairs could be converted into a mutagenesis-prioritization score, and its agreement with alanine-scanning or deep-mutational-scanning data could be quantified experimentally.
  • Editorial extension: the evaluation design of feeding the same structure-based model crystal versus predicted structures suggests a general standard—any structure-aware binding model should report how much its accuracy drops when structures come from folding tools rather than crystals.
  • Editorial extension: if attention between antibody and antigen is the reason these sequence-only predictions work, then zero-shot protein language models that score mutants without antigen context may be leaving a large amount of signal on the table.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

6 major / 5 minor

Summary. The manuscript describes ProtAttBA, a sequence-only deep learning model that predicts binding-affinity changes (ΔΔG) for antibody-antigen complexes. The model encodes wild-type and mutant antibody/antigen sequences with a frozen protein language model (ESM2, ESM-1b, ProtBert, or Ankh), then applies 1D convolutions, dual multi-head cross-attention with rotary embeddings, and convolutional pooling before a regression head. The authors evaluate on AB645, S1131, and AB1101 under K-fold cross-validation, a 30% sequence-identity split, and a mutation-depth split, comparing with sequence- and structure-based baselines. They report that ProtAttBA is competitive or better than baselines, robust to structural uncertainty, and that attention weights highlight functionally relevant residues in two case studies.

Significance. Accurate sequence-only prediction of mutation effects would be practically valuable for antibody engineering, especially when structures are unavailable; the paper also provides a useful comparison of four protein language models and reports an attention ablation. Strengths include a clear architecture, code/data availability, and evaluation under multiple protocols. However, the headline performance claim is not fully supported by the reported tables: DDGPred-PDB outperforms ProtAttBA on S1131 in Table 1, and the cross-validation protocol may permit same-complex leakage. If the mutation-chain restriction is handled and the performance claims are revised to be accurate, the framework could still make a modest contribution.

major comments (6)
  1. [Section 2.2.1 and Fig. 1] The input is defined as wild-type antibody, wild-type antigen, mutated antibody, and wild-type antigen, so a mutation on the antigen chain is not represented in any input channel; for such a mutation the model output is invariant to the mutation. Since S1131 is derived from SKEMPI, which contains mutations on either interaction partner, and the text does not state that all evaluated mutations occur on the antibody chain, the reported results may include predictions for inputs that do not encode the labeled mutation. Please state the filtering rule, report the per-benchmark count of antibody-side versus antigen-side mutations, and either exclude antigen-side mutations or extend the architecture to encode them.
  2. [Section 3.2, Table 1] The text says ProtAttBA 'consistently outperforms all baseline methods,' but this is contradicted by the table. DDGPred-PDB achieves RMSE 0.95±0.13, R2 0.84±0.04, PCC 0.92±0.02, and Spearman 0.85±0.02 on S1131, while ProtAttBA-ESM2 achieves 1.31±0.09, 0.69±0.09, 0.84±0.05, and 0.75±0.06; DDGPred-PDB is also better on AB645's R2, PCC, and Spearman. The claim should be limited to what the table supports, and the comparison should include significance testing over folds.
  3. [Section 3.1] The K-fold cross-validation description does not state whether the splits are grouped by complex. Because AB645, S1131, and AB1101 contain multiple mutations per complex, random splitting can place mutations from the same complex in both training and test folds, which inflates correlation and R2. Please clarify whether folds are complex-aware; if they are not, report complex-grouped cross-validation results.
  4. [Section 3.2, Table 2] The sequence-identity and mutation-depth results are given as single numbers with no standard deviations or confidence intervals, so statements such as 'ProtAttBA maintained relatively stable and reliable performance' are not quantitatively supported. Please report repeated runs with different random seeds or otherwise provide uncertainty estimates for these splits.
  5. [Section 2.1] The paper calls S1131 an antibody-antigen benchmark, but S1131 from SKEMPI is a general protein-protein interaction mutation set. If S1131 includes non-antibody complexes, then the terms 'antibody' and 'antigen' are misnomers for those entries, and conclusions about antibody engineering from S1131 are not justified. Please clarify the composition of S1131 and restrict the claims or the dataset accordingly.
  6. [Section 3.3] The interpretability claim relies on two selected complexes and visual inspection of attention maps, with no quantitative measure of whether attention-highlighted residues coincide with known functional sites beyond the two examples. The conclusion that 'attention scores can identify critical residues' is therefore not established. Please add quantitative validation (e.g., overlap with known epitope residues across many complexes) or weaken the claim to a qualitative observation.
minor comments (5)
  1. [Abstract] In the abstract, 'Therefore, We propose' has an unnecessary capital 'W' in 'We'; this should be lowercase.
  2. [Equation (7)] The sentence 'Hl o denotes the lth column in Ho, i.e., the ith position of the protein' uses both l and i for the same index; the index notation should be made consistent.
  3. [Section 2.2.3] The text says the joint representation is 'summarized by summing the information from the wild-type and mutant complexes,' but the equation f = concat(f_wt_ab + f_wt_ag, f_mt_ab + f_mt_ag) is a concatenation of two sums; this wording should be clarified.
  4. [Table 1] Several result entries are typeset without spaces (e.g., 'ProtAttBA-ESM21.70±0.25'), making the table difficult to read; the LaTeX spacing should be fixed.
  5. [Section 3.1] The sentence fragment 'Using early stopping to avoid overfitting.' should be integrated into the preceding sentence.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: ProtAttBA's predictions are held-out supervised outputs, and the cited baselines and pLMs are external evidence, not self-citation chains.

full rationale

ProtAttBA is a supervised regressor: its reported numbers in Tables 1 and 2 are out-of-fold predictions on held-out folds and split protocols, so they are not fitted constants or renamed inputs. The prediction target ΔΔG comes from external benchmarks (SKEMPI and AB-bind), and the architecture equations (1)–(8) transform frozen pre-trained protein language model embeddings through learnable convolution, attention, and fully connected layers; no parameter is fit to the test labels and then reported as a prediction. Baseline comparisons either reproduce external methods or take sequence-baseline numbers from Jin et al. (2024), which is standard benchmarking rather than a self-citation chain. Some cited works include the same authors (e.g., Tan et al. 2024, Zhou et al. 2024, Li et al. 2025), but those citations support general contextual statements such as structures correlating with binding or AlphaFold2 being mainstream; they are not load-bearing uniqueness theorems, forced ansatz choices, or fitted inputs. The repeated 'wild-type antigen' in the embedding-module input description is a modeling-scope limitation for antigen-side mutations rather than a circular derivation: it does not make the reported predictions equal to a fitted input by construction. Overall, no circular step meeting the quoted-evidence standard was found.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The model itself contains thousands of trained weights, but those are estimated from benchmark training folds rather than introduced as physical constants. The explicit hyperparameters above are the hand-set values. The three listed axioms are the background assumptions the predictions depend on: the informativeness of frozen pLM embeddings, the trustworthiness of the experimental labels, and the antibody-only mutation assumption, which is implicit in the architecture.

free parameters (3)
  • learning rate = 3e-5
    AdamW learning rate selected for optimization, reported in Section 3.1.
  • conv1d kernel size = 1
    Kernel size of the spatial-weighting convolution in Eq. (2), a design choice.
  • attention head count = not reported
    The multi-head attention in Eq. (6) uses N heads, but the paper does not report N or hidden dimensions, leaving important hyperparameters unspecified.
assumptions (3)
  • domain assumption Frozen protein language model embeddings are sufficient to encode antibody-antigen interaction information needed for ΔΔG prediction.
    The model uses ESM2, ESM1b, ProtBert, or Ankh without fine-tuning (Section 2.2.1 and Section 3.1). If this fails, the whole method loses predictive power.
  • domain assumption The experimental ΔΔG labels in SKEMPI and AB-bind are reliable and comparable across complexes.
    All training and evaluation use these labels (Section 2.1). Systematic errors or inconsistent SPR conditions would propagate into every reported metric.
  • ad hoc to paper Mutations occur only on the antibody chain.
    The architecture always inputs the wild-type antigen as the mutant antigen (Section 2.2.1), so antigen mutations are invisible to the model. This is an unstated modeling restriction.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sequence-Only Prediction of Binding Affinity Changes: A Robust and Interpretable Model for Antibody Engineering." pith.science (2026). https://pith.science/paper/GYPWNJQA

@misc{pith2026250520301,
  author       = {Pith},
  title        = {Pith review of: Sequence-Only Prediction of Binding Affinity Changes: A Robust and Interpretable Model for Antibody Engineering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GYPWNJQA}},
  note         = {Machine review of arXiv:2505.20301}
}
read the original abstract

A pivotal area of research in antibody engineering is to find effective modifications that enhance antibody-antigen binding affinity. Traditional wet-lab experiments assess mutants in a costly and time-consuming manner. Emerging deep learning solutions offer an alternative by modeling antibody structures to predict binding affinity changes. However, they heavily depend on high-quality complex structures, which are frequently unavailable in practice. Therefore, we propose ProtAttBA, a deep learning model that predicts binding affinity changes based solely on the sequence information of antibody-antigen complexes. ProtAttBA employs a pre-training phase to learn protein sequence patterns, following a supervised training phase using labeled antibody-antigen complex data to train a cross-attention-based regressor for predicting binding affinity changes. We evaluated ProtAttBA on three open benchmarks under different conditions. Compared to both sequence- and structure-based prediction methods, our approach achieves competitive performance, demonstrating notable robustness, especially with uncertain complex structures. Notably, our method possesses interpretability from the attention mechanism. We show that the learned attention scores can identify critical residues with impacts on binding affinity. This work introduces a rapid and cost-effective computational tool for antibody engineering, with the potential to accelerate the development of novel therapeutic antibodies.

Figures

Figures reproduced from arXiv: 2505.20301 by the authors.

Figure 1
Figure 1. Overview of the ProtAttBA architecture. The model predicts changes in antigen–antibody binding affinity (∆∆G) by amino acid mutations. Given wild-type and mutant sequence pairs, ProtAttBA first encodes antibody and antigen sequences using a frozen pre-trained protein language model to generate contextualized residue embeddings {Hwt ab , Hwt ag , Hmt ab , Hmt ag }. The attention module then applies convolutional neur… view at source ↗
Figure 2
Figure 2. Ablative comparison of attention-based and MLP-based ProtAttBA by PCC (top) and R2 (bottom) on the prediction performance on the three benchmark datasets. (Dehouck et al., 2013), FoldX (Schymkowitz et al., 2005), and DDGPred (Shan et al., 2022). The performance comparison under the first cross-validation setup is presented in [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Protein structure visualization for interpretability analysis. Panels a, b, c, and d depict localized views of the antibody-antigen complex at the mutation site, before and after mutation, respectively. The antigen chain is highlighted in green. Panels e and f illustrate the attention weight matrices learned by the model, where cooler colors (tending towards blue) indicate regions where the model assigns higher impo… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 41 canonical work pages

  1. [1]

    X. Liu, Y. Luo, P. Li, S. Song, and J. Peng, ``Deep geometric representations for modeling effects of mutations on protein-protein binding affinity,'' PLOS Computational Biology, vol. 17, pp. 1--28, 08 2021

    work page 2021

  2. [2]

    G. Wang, X. Liu, K. Wang, Y. Gao, G. Li, D. Baptista-Hon, X. Yang, K. Xue, W. Tai, Z. Jiang, L. Cheng, M. Fok, J. Lau, S. Yang, L. Lu, P. Zhang, and K. Zhang, ``Deep-learning-enabled protein–protein interaction analysis for prediction of sars-cov-2 infectivity and variant evolution,'' Nature Medicine, vol. 29, pp. 1--12, 07 2023

    work page 2023

  3. [3]

    Zhang, Q

    J. Zhang, Q. Wu, Z. Liu, Q. Wang, J. Wu, Y. Hu, T. Bai, T. Xie, M. Huang, T. Wu et al., ``Spike-specific circulating t follicular helper cell and cross-neutralizing antibody responses in covid-19-convalescent individuals,'' Nature Microbiology, vol. 6, no. 1, pp. 51--58, 2021

    work page 2021

  4. [4]

    A. Beck, T. Wurch, C. Bailly, and N. Corvaia, ``Strategies and challenges for the next generation of therapeutic antibodies,'' Nature Reviews Immunology, vol. 10, no. 5, pp. 345--352, 2010

    work page 2010

  5. [5]

    E. M. Brustad and F. H. Arnold, ``Optimizing non-natural protein function with directed evolution,'' Current Opinion in Chemical Biology, vol. 15, no. 2, pp. 201--210, 2011

    work page 2011

  6. [6]

    Kouba, P

    P. Kouba, P. Kohout, F. Haddadi, A. Bushuiev, R. Samusevich, J. Sedlar, J. Damborsky, T. Pluskal, J. Sivic, and S. Mazurenko, ``Machine learning-guided protein engineering,'' ACS Catalysis, vol. 13, no. 21, pp. 13\,863--13\,895, 2023

    work page 2023

  7. [7]

    C. Hsu, R. Verkuil, J. Liu, Z. Lin, B. Hie, T. Sercu, A. Lerer, and A. Rives, ``Learning inverse folding from millions of predicted structures,'' in International Conference on Machine Learning. 1em plus 0.5em minus 0.4em PMLR, 2022, pp. 8946--8970

    work page 2022

  8. [8]

    M. Li, Y. Shi, S. Hu, S. Hu, P. Guo, W. Wan, L. Y. Zhang, S. Pan, J. Li, L. Sun et al., ``Mvsf-ab: Accurate antibody-antigen binding affinity prediction via multi-view sequence feature learning,'' Bioinformatics, p. btae579, 2024

    work page 2024

Show all 44 references
  1. [9]

    B. Zhou, L. Zheng, B. Wu, Y. Tan, O. Lv, K. Yi, G. Fan, and L. Hong, ``Protein engineering with lightweight graph denoising neural networks,'' Journal of Chemical Information and Modeling, 2024

    2024

  2. [10]

    B. Zhou, Y. Tan, Y. Hu, L. Zheng, B. Zhong, and L. Hong, ``Protein engineering in the deep learning era,'' mLife, vol. 3, no. 4, pp. 477--491, 2024

    2024

  3. [11]

    Cuturello, M

    F. Cuturello, M. Celoria, A. Ansuini, and A. Cazzaniga, ``Enhancing predictions of protein stability changes induced by single mutations using msa-based language models,'' Bioinformatics, vol. 40, no. 7, p. btae447, 2024

    2024

  4. [12]

    Schymkowitz, J

    J. Schymkowitz, J. Borg, F. Stricher, R. Nys, F. Rousseau, and L. Serrano, ``The foldx web server: an online force field,'' Nucleic Acids Research, vol. 33, no. suppl\_2, pp. W382--W388, 2005

    2005

  5. [13]

    Dehouck, J

    Y. Dehouck, J. M. Kwasigroch, M. Rooman, and D. Gilis, ``Beatmusic: prediction of changes in protein--protein binding affinity on mutations,'' Nucleic Acids Research, vol. 41, no. W1, pp. W333--W339, 2013

    2013

  6. [14]

    D. E. Pires and D. B. Ascher, ``mcsm-ab: a web server for predicting antibody--antigen affinity changes upon mutation with graph-based signatures,'' Nucleic Acids Research, vol. 44, no. W1, pp. W469--W473, 2016

    2016

  7. [15]

    M. Wang, Z. Cang, and G.-W. Wei, ``A topology-based network tree for the prediction of protein--protein binding affinity changes following mutation,'' Nature Machine Intelligence, vol. 2, no. 2, pp. 116--123, 2020

    2020

  8. [16]

    G. Yu, Q. Zhao, X. Bi, and J. Wang, ``Ddaffinity: predicting the changes in binding affinity of multiple point mutations using protein 3d structure,'' Bioinformatics, vol. 40, no. Supplement\_1, pp. i418--i427, 2024

    2024

  9. [17]

    Y. Tan, M. Li, B. Zhou, B. Zhong, L. Zheng, P. Tan, Z. Zhou, H. Yu, G. Fan, and L. Hong, ``Simple, efficient, and scalable structure-aware adapter boosts protein language models,'' Journal of Chemical Information and Modeling, vol. 64, no. 16, pp. 6338--6349, 2024

    2024

  10. [18]

    Huang, C

    J. Huang, C. Sun, M. Li, R. Tang, B. Xie, S. Wang, and J.-M. Wei, ``Structure-inclusive similarity based directed gnn: a method that can control information flow to predict drug--target binding affinity,'' Bioinformatics, vol. 40, no. 10, p. btae563, 2024

    2024

  11. [19]

    S. Shan, S. Luo, Z. Yang, J. Hong, Y. Su, F. Ding, L. Fu, C. Li, P. Chen, J. Ma et al., ``Deep learning guided optimization of human antibody against sars-cov-2 variants with broad neutralization,'' Proceedings of the National Academy of Sciences, vol. 119, no. 11, p. e2122954...

    2022

  12. [20]

    M. M. Rana and D. D. Nguyen, ``Geometric graph learning to predict changes in binding free energy and protein thermodynamic stability upon mutation,'' The Journal of Physical Chemistry Letters, vol. 14, no. 49, pp. 10\,870--10\,879, 2023

    2023

  13. [21]

    R. Jin, Q. Ye, J. Wang, Z. Cao, D. Jiang, T. Wang, Y. Kang, W. Xu, C.-Y. Hsieh, and T. Hou, ``Attabseq: an attention-based deep learning prediction method for antigen--antibody binding affinity changes based on protein sequences,'' Briefings in Bioinformatics, vol. 25, no. 4, ...

    2024

  14. [22]

    Y. Tan, R. Wang, B. Wu, L. Hong, and B. Zhou, ``Retrieval-enhanced mutation mastery: Augmenting zero-shot prediction of protein language model,'' arXiv:2410.21127, 2024

    2024 arXiv

  15. [23]

    Misra, J

    M. Misra, J. Jeffy, C. Liao, S. Pickthorn, K. Wagh, and A. Herschhorn, ``Hiresist: a database of hiv-1 resistance to broadly neutralizing antibodies,'' Bioinformatics, vol. 40, no. 3, p. btae103, 2024

    2024

  16. [24]

    Xiong, C

    P. Xiong, C. Zhang, W. Zheng, and Y. Zhang, ``Bindprofx: assessing mutation-induced binding affinity change by protein interface profiles with pseudo-counts,'' Journal of Molecular Biology, vol. 429, no. 3, pp. 426--434, 2017

    2017

  17. [25]

    I. H. Moal and J. Fern \'a ndez-Recio, ``Skempi: a structural kinetic and energetic database of mutant protein interactions and its use in empirical models,'' Bioinformatics, vol. 28, no. 20, pp. 2600--2607, 2012

    2012

  18. [26]

    Sirin, J

    S. Sirin, J. R. Apgar, E. M. Bennett, and A. E. Keating, ``Ab-bind: antibody binding mutational database for computational affinity predictions,'' Protein Science, vol. 25, no. 2, pp. 393--409, 2016

    2016

  19. [27]

    Y. Bai, J. Mei, A. L. Yuille, and C. Xie, ``Are transformers more robust than cnns?'' Advances in Neural Information Processing Systems, vol. 34, pp. 26\,831--26\,843, 2021

    2021

  20. [28]

    Elnaggar, M

    A. Elnaggar, M. Heinzinger, C. Dallago, G. Rehawi, Y. Wang, L. Jones, T. Gibbs, T. Feher, C. Angerer, M. Steinegger et al., ``Prottrans: Toward understanding the language of life through self-supervised learning,'' IEEE transactions on Pattern Analysis and Machine Intelligence...

    2021

  21. [29]

    M. Li, Y. Tan, X. Ma, B. Zhong, H. Yu, Z. Zhou, W. Ouyang, B. Zhou, P. Tan, and L. Hong, ``Pro SST : Protein language modeling with quantized structure and disentangled attention,'' in The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024

    2024

  22. [30]

    X. Xu, C. Xu, W. He, L. Wei, H. Li, J. Zhou, R. Zhang, Y. Wang, Y. Xiong, and X. Gao, ``Helm-gpt: de novo macrocyclic peptide design using generative pre-trained transformer,'' Bioinformatics, p. btae364, 2024

    2024

  23. [31]

    Y. Xiao, E. Sun, Y. Jin, Q. Wang, and W. Wang, ``Proteingpt: Multimodal llm for protein property prediction and structure understanding,'' arXiv preprint arXiv:2408.11363, 2024

    2024 arXiv

  24. [32]

    Rives, J

    A. Rives, J. Meier, T. Sercu, S. Goyal, Z. Lin, J. Liu, D. Guo, M. Ott, C. L. Zitnick, J. Ma et al., ``Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences,'' Proceedings of the National Academy of Sciences, vol. 118, no....

    2021

  25. [33]

    Z. Lin, H. Akin, R. Rao, B. Hie, Z. Zhu, W. Lu, N. Smetanin, R. Verkuil, O. Kabeli, Y. Shmueli et al., ``Evolutionary-scale prediction of atomic-level protein structure with a language model,'' Science, vol. 379, no. 6637, pp. 1123--1130, 2023

    2023

  26. [34]

    Elnaggar, H

    A. Elnaggar, H. Essam, W. Salah-Eldin, W. Moustafa, M. Elkerdawy, C. Rochereau, and B. Rost, ``Ankh: Optimized protein language model unlocks general-purpose modelling,'' arXiv preprint arXiv:2301.06568, 2023

    2023 arXiv

  27. [35]

    J. Su, M. Ahmed, Y. Lu, S. Pan, W. Bo, and Y. Liu, ``Roformer: Enhanced transformer with rotary position embedding,'' Neurocomputing, vol. 568, p. 127063, 2024

    2024

  28. [36]

    Loshchilov and F

    I. Loshchilov and F. Hutter, ``Decoupled weight decay regularization,'' in International Conference on Learning Representations, 2019

    2019

  29. [37]

    Y. Yao, X. Du, Y. Diao, and H. Zhu, ``An integration of deep learning with feature embedding for protein--protein interaction prediction,'' PeerJ, vol. 7, p. e7126, 2019

    2019

  30. [38]

    Tsukiyama, M

    S. Tsukiyama, M. M. Hasan, S. Fujii, and H. Kurata, ``Lstm-phv: prediction of human-virus protein--protein interactions by lstm with word2vec,'' Briefings in Bioinformatics, vol. 22, no. 6, p. bbab228, 2021

    2021

  31. [39]

    M. Chen, C. J.-T. Ju, G. Zhou, X. Chen, T. Zhang, K.-W. Chang, C. Zaniolo, and W. Wang, ``Multifaceted protein--protein interaction prediction based on siamese residual rcnn,'' Bioinformatics, vol. 35, no. 14, pp. i305--i314, 2019

    2019

  32. [40]

    X. Yang, S. Yang, X. Lian, S. Wuchty, and Z. Zhang, ``Transfer learning via multi-scale convolutional neural layers for human--virus protein--protein interaction prediction,'' Bioinformatics, vol. 37, no. 24, pp. 4771--4778, 2021

    2021

  33. [41]

    Jumper, R

    J. Jumper, R. Evans, A. Pritzel, T. Green, M. Figurnov, O. Ronneberger, K. Tunyasuvunakool, R. Bates, A. Z \' dek, A. Potapenko et al., ``Highly accurate protein structure prediction with alphafold,'' Nature, vol. 596, no. 7873, pp. 583--589, 2021

    2021

  34. [42]

    Townshend, R

    R. Townshend, R. Bedi, P. Suriana, and R. Dror, ``End-to-end learning on 3d protein structure for interface prediction,'' Advances in Neural Information Processing Systems, vol. 32, 2019

    2019

  35. [43]

    B. Zhou, L. Zheng, B. Wu, K. Yi, B. Zhong, Y. Tan, Q. Liu, P. Li \`o , and L. Hong, ``A conditional protein diffusion model generates artificial programmable endonuclease sequences with enhanced activity,'' Cell Discovery, vol. 10, no. 1, p. 95, 2024

    2024

  36. [44]

    S. Li, Y. Tan, S. Ke, L. Hong, and B. Zhou, ``Immunogenicity prediction with dual attention enables vaccine target selection,'' in The Thirteenth International Conference on Learning Representations, 2025

    2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.