pith. sign in

arxiv: 2605.22133 · v3 · pith:4LFMY7YZnew · submitted 2026-05-21 · 🧬 q-bio.BM · cs.AI

Atom-level Protein Representation Learning Improves Protein Structure Prediction

Taewon Kim , Hyosoon Jang , Hyunjin Seo , Seonghwan Seo , Hyeongwoo Kim , Wonho Zhung , Mingyeong Shin , Wooyoun Kim
show 1 more author
Sungsoo Ahn
This is my paper

Pith reviewed 2026-06-30 16:33 UTC · model grok-4.3

classification 🧬 q-bio.BM cs.AI
keywords protein representation learningstructure predictionpretrainingVQ-VAEhomodimer co-foldingTriProRepRepSP benchmarkmulti-view encoding
0
0 comments X

The pith

TriProRep pretrains on three aligned protein views to improve structure prediction over sequence-only and prior models.

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

The paper proposes TriProRep, a pretraining approach that jointly encodes amino-acid identity, backbone geometry, and local full-atom geometry at the residue level using VQ-VAE tokenizers. It trains the model to recover original tokens after generator-based corruption of these views, forcing it to reject plausible but incorrect cross-view combinations. The authors introduce the RepSP benchmark to measure how well representations support three structure-related tasks: homodimer co-folding from separate apo chains, predicting residue-level interaction properties, and aligning representations to monomer structure prediction. TriProRep shows gains on these tasks while remaining competitive on standard benchmarks. A reader would care because improved representations could serve as stronger conditioning or alignment targets for downstream structure generation without retraining large models from scratch.

Core claim

TriProRep jointly models three aligned residue-level views—amino-acid identity, backbone geometry, and local full-atom geometry—discretely encoded via VQ-VAE tokenizers. By pretraining to recover original tokens from generator-corrupted views, the model learns to distinguish plausible but incorrect cross-view augmentations from the original protein. Across the RepSP tasks of homodimer co-folding from apo-chain representations, residue-level prediction of homodimer-derived interaction properties, and representation-aligned monomer structure prediction, TriProRep improves over sequence-only and prior structure-aware representation models.

What carries the argument

TriProRep, a structure-aware pretraining method that encodes three aligned residue-level views (amino-acid identity, backbone geometry, local full-atom geometry) via VQ-VAE tokenizers and recovers tokens from generator-corrupted multi-view inputs.

If this is right

  • TriProRep representations improve homodimer co-folding when starting from separate apo-chain inputs.
  • They yield better residue-level predictions of homodimer interaction properties.
  • They improve representation-aligned monomer structure prediction.
  • They maintain competitive results on conventional protein representation benchmarks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same three-view corruption objective could be tested on other multi-scale biological sequences such as RNA or glycans.
  • If the learned distinctions prove transferable, the representations might serve as fixed conditioning features in generative structure models to reduce compute.
  • The RepSP tasks could be expanded to include heterodimer cases or ligand-binding site recovery to probe broader utility.

Load-bearing premise

Recovering original tokens from generator-corrupted multi-view inputs teaches distinctions that transfer to better performance on structure prediction tasks.

What would settle it

A control model trained on the same three views but without the cross-view token-recovery objective would match or exceed TriProRep performance on the three RepSP tasks.

Figures

Figures reproduced from arXiv: 2605.22133 by Hyeongwoo Kim, Hyosoon Jang, Hyunjin Seo, Mingyeong Shin, Seonghwan Seo, Sungsoo Ahn, Taewon Kim, Wonho Zhung, Wooyoun Kim.

Figure 1
Figure 1. Figure 1: TRIPROREP. (a) Three-view tokenization. A protein is independently tokenized into amino-acid, backbone, and full-atom token sequences at the residue level. (b) ELECTRA-style discriminative pretraining. A small generator corrupts each of the three sequences, and a large discriminator predicts the original token at every position. The richer space of cross-token corruptions provides a stronger training signa… view at source ↗
Figure 2
Figure 2. Figure 2: REPSP. We define three structure-generative tasks that use protein representations as input: (task 1) homodimer structure prediction, (task 2) per-residue homodimer binding-property prediction via MLP probing, and (task 3) distillation into a monomer structure prediction model. identity. From the resulting cluster representatives, we select 400 validation and 1,000 test sequences, and use the remaining rep… view at source ↗
Figure 3
Figure 3. Figure 3: Scaling of flexible-docking. Predicted homodimer structures (chain A blue, chain B gold) overlaid on ground truth (gray) across encoder sizes (150M, 650M, 3B) for four test records. while the huge TRIPROREP model achieves the strongest performance on nearly all metrics, with ESM3 only marginally higher in LDDT. The gains are most pronounced on interface-level metrics, which depend not only on accurate mono… view at source ↗
Figure 4
Figure 4. Figure 4: Acceleration of monomer structure prediction via representation alignment on REPSP. We compare the no-REPA baseline against ESM2, SaProt, S-PLM, MIF-ST, and TRIPROREP as the alignment target. TRIPROREP provides strongest alignment target for structure prediction model. 5.2 Per-residue homodimer binding property prediction [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: Acceleration of monomer structure prediction via representation alignment on REPSP. We compare the no-REPA baseline against ESM2, SaProt, S-PLM, MIF-ST, and TRIPROREP as the alignment target. TRIPROREP provides strongest alignment target for structure prediction model [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Tokens vs. sidechain ro￾tamer. Density of codes in the χ1 simplex. (a) Backbone tokens. (b) Full-atom tokens. Hyperparameters. The tokenizer uses single width 256, pair width 128, and N = 6 Pairformer-style layers with 8 attention heads. The output embedding dimension is 256. The codebook contains V = 512 entries, uses EMA updates with decay 0.99, and uses entropy regularization with weight 0.1. Backbone a… view at source ↗
read the original abstract

Recent advances in generative modeling show that pretrained representations can improve generation as conditioning features or alignment targets. Motivated by this, we study protein representations for predicting structures beyond conventional function annotation. We propose TriProRep, a structure-aware pretraining method that jointly models three aligned residue-level views: amino-acid identity, backbone geometry, and local full-atom geometry, discretely encoded via VQ-VAE tokenizers. By pretraining to recover original tokens from generator-corrupted views, TriProRep learns to distinguish plausible but incorrect cross-view augmentations from the original protein. We further introduce RepSP, a benchmark for evaluating protein representations in structure-predictive settings. RepSP tests three uses of representations: homodimer co-folding from apo-chain representations, residue-level prediction of homodimer-derived interaction properties, and representation-aligned monomer structure prediction. Across these tasks, TriProRep improves over sequence-only and prior structure-aware representation models, while maintaining competitive performance on conventional benchmarks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes TriProRep, a structure-aware pretraining method that jointly models three aligned residue-level views (amino-acid identity, backbone geometry, local full-atom geometry) via VQ-VAE tokenizers and pretrains by recovering original tokens from generator-corrupted multi-view inputs. It introduces the RepSP benchmark consisting of three tasks (homodimer co-folding from apo representations, residue-level prediction of homodimer interaction properties, representation-aligned monomer structure prediction) and claims that TriProRep outperforms sequence-only and prior structure-aware models on these tasks while remaining competitive on conventional benchmarks.

Significance. If the claimed gains prove robust and specifically attributable to the cross-view pretraining mechanism, the work could advance the use of atom-level multi-view representations as conditioning or alignment targets for structure prediction, extending recent generative modeling ideas into representation learning for proteins.

major comments (2)
  1. [results section on RepSP experiments] The central claim that TriProRep improves on the three RepSP tasks rests on the assumption that recovering tokens from generator-corrupted multi-view inputs teaches the model to distinguish plausible but incorrect cross-view augmentations in a way that transfers to structure prediction. However, the manuscript provides no ablation studies that isolate the cross-view corruption component from the mere use of three aligned VQ-VAE views or from the discretization itself. This attribution is load-bearing for the strongest claim and remains untested.
  2. [abstract and results] The abstract asserts performance gains over baselines on RepSP tasks but supplies no quantitative metrics, specific baselines, error bars, statistical significance, or ablation details. Without these, the magnitude, reliability, and reproducibility of the reported improvements cannot be evaluated from the provided description.
minor comments (1)
  1. [methods] Notation for the three views and VQ-VAE tokenizers could be introduced more explicitly with consistent symbols to aid readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive comments. We address each major comment below, providing clarifications and committing to revisions that strengthen the manuscript without misrepresenting our results.

read point-by-point responses

  1. Referee: [results section on RepSP experiments] The central claim that TriProRep improves on the three RepSP tasks rests on the assumption that recovering tokens from generator-corrupted multi-view inputs teaches the model to distinguish plausible but incorrect cross-view augmentations in a way that transfers to structure prediction. However, the manuscript provides no ablation studies that isolate the cross-view corruption component from the mere use of three aligned VQ-VAE views or from the discretization itself. This attribution is load-bearing for the strongest claim and remains untested.

    Authors: We agree that explicit ablations isolating the cross-view corruption mechanism would strengthen attribution of the RepSP gains. The current experiments compare the full TriProRep model against sequence-only and prior structure-aware baselines, but do not include variants that remove the generator corruption or restrict to single views. In the revised manuscript we will add these ablation studies, training and evaluating models with (i) no corruption and (ii) single-view inputs only, to quantify the specific contribution of recovering tokens from corrupted multi-view inputs. revision: yes

  2. Referee: [abstract and results] The abstract asserts performance gains over baselines on RepSP tasks but supplies no quantitative metrics, specific baselines, error bars, statistical significance, or ablation details. Without these, the magnitude, reliability, and reproducibility of the reported improvements cannot be evaluated from the provided description.

    Authors: We acknowledge that the abstract would be more informative with concrete numbers. Although the results section already reports quantitative metrics, baselines, error bars, and statistical details for the RepSP tasks, we will revise the abstract to include the key performance deltas, name the primary baselines, and reference the presence of error bars and significance tests. This change will allow readers to evaluate the claims directly from the abstract while remaining within length limits. revision: yes

Circularity Check

0 steps flagged

No circularity: standard pretraining objective with independent benchmark evaluation

full rationale

The paper defines TriProRep via a token-recovery pretraining objective on VQ-VAE encoded multi-view inputs and evaluates it on a newly introduced RepSP benchmark consisting of three downstream structure-prediction tasks. No derivation step equates a claimed prediction or result to a fitted parameter or self-citation by construction. The pretraining loss is a standard masked-token recovery objective; downstream gains are measured against external baselines on held-out tasks. No self-citation load-bearing steps, uniqueness theorems, or ansatz smuggling appear in the provided text. This is the expected non-finding for a representation-learning paper whose central claim rests on empirical transfer rather than algebraic identity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract alone supplies no concrete free parameters, axioms, or invented entities; full methods would be required to populate the ledger.

pith-pipeline@v0.9.1-grok · 5727 in / 994 out tokens · 45813 ms · 2026-06-30T16:33:15.457358+00:00 · methodology

Review history (3 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

40 extracted references · 3 canonical work pages

  1. [1]

    Douglas Renfrew, Tomasz Kosciolek, Julia Koehler Leman, Daniel Berenberg, Tommi Vatanen, Chris Chandler, Bryn C

    Vladimir Gligorijevi´c, P. Douglas Renfrew, Tomasz Kosciolek, Julia Koehler Leman, Daniel Berenberg, Tommi Vatanen, Chris Chandler, Bryn C. Taylor, Ian M. Fisk, Hera Vlamakis, Ramnik J. Xavier, Rob Knight, Kyunghyun Cho, and Richard Bonneau. Structure-based protein function prediction using graph convolutional networks.Nature Communications, 12(1):3168, 2021

    2021

  2. [2]

    Saprot: Protein language modeling with structure-aware vocabulary

    Jin Su, Chenchen Han, Yuyang Zhou, Junjie Shan, Xibin Zhou, and Fajie Yuan. Saprot: Protein language modeling with structure-aware vocabulary. InThe Twelfth International Conference on Learning Representations, 2024

    2024

  3. [3]

    Bilingual language model for protein sequence and structure.NAR Genomics and Bioinformatics, 6(4):lqae150, 12 2024

    Michael Heinzinger, Konstantin Weissenow, Joaquin Gomez Sanchez, Adrian Henkel, Milot Mirdita, Martin Steinegger, and Burkhard Rost. Bilingual language model for protein sequence and structure.NAR Genomics and Bioinformatics, 6(4):lqae150, 12 2024

    2024

  4. [4]

    Fast and accurate protein structure search with foldseek.Nature biotechnology, 42(2):243–246, 2024

    Michel Van Kempen, Stephanie S Kim, Charlotte Tumescheit, Milot Mirdita, Jeongjae Lee, Cameron LM Gilchrist, Johannes Söding, and Martin Steinegger. Fast and accurate protein structure search with foldseek.Nature biotechnology, 42(2):243–246, 2024

    2024

  5. [5]

    Simulating 500 million years of evolution with a language model.Science, 387(6736):850–858, 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, 387(6736):850–858, 2025

    2025

  6. [6]

    Masked inverse folding with sequence transfer for protein representation learning.Protein Engineering, Design and Selection, 36:gzad015, 2023

    Kevin K Yang, Niccolò Zanichelli, and Hugh Yeh. Masked inverse folding with sequence transfer for protein representation learning.Protein Engineering, Design and Selection, 36:gzad015, 2023

    2023

  7. [7]

    S-plm: structure-aware protein language model via contrastive learning between sequence and structure.Advanced Science, 12(5):2404212, 2025

    Duolin Wang, Mahdi Pourmirzaei, Usman L Abbas, Shuai Zeng, Negin Manshour, Farzaneh Esmaili, Biplab Poudel, Yuexu Jiang, Qing Shao, Jin Chen, et al. S-plm: structure-aware protein language model via contrastive learning between sequence and structure.Advanced Science, 12(5):2404212, 2025

    2025

  8. [8]

    Return of unconditional generation: A self- supervised representation generation method.Advances in Neural Information Processing Systems, 37:125441–125468, 2024

    Tianhong Li, Dina Katabi, and Kaiming He. Return of unconditional generation: A self- supervised representation generation method.Advances in Neural Information Processing Systems, 37:125441–125468, 2024

    2024

  9. [9]

    Ip-adapter: Text compatible image prompt adapter for text-to-image diffusion models

    Hu Ye, Jun Zhang, Sibo Liu, Xiao Han, and Wei Yang. Ip-adapter: Text compatible image prompt adapter for text-to-image diffusion models. 2023

    2023

  10. [10]

    Test-time conditioning with representation-aligned visual features.arXiv preprint arXiv:2602.03753, 2026

    Nicolas Sereyjol-Garros, Ellington Kirby, Victor Letzelter, Victor Besnier, and Nermin Samet. Test-time conditioning with representation-aligned visual features.arXiv preprint arXiv:2602.03753, 2026

    work page arXiv 2026

  11. [11]

    Representation alignment for generation: Training diffusion transformers is easier than you think

    Sihyun Yu, Sangkyung Kwak, Huiwon Jang, Jongheon Jeong, Jonathan Huang, Jinwoo Shin, and Saining Xie. Representation alignment for generation: Training diffusion transformers is easier than you think. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  12. [12]

    Repa-e: Unlocking vae for end-to-end tuning of latent diffusion transformers

    Xingjian Leng, Jaskirat Singh, Yunzhong Hou, Zhenchang Xing, Saining Xie, and Liang Zheng. Repa-e: Unlocking vae for end-to-end tuning of latent diffusion transformers. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 18262–18272, 2025

    2025

  13. [13]

    Neural discrete representation learning.Advances in neural information processing systems, 30, 2017

    Aaron Van Den Oord, Oriol Vinyals, et al. Neural discrete representation learning.Advances in neural information processing systems, 30, 2017

    2017

  14. [14]

    Protein structure tok- enization: Benchmarking and new recipe

    Xinyu Yuan, Zichen Wang, Marcus D Collins, and Huzefa Rangwala. Protein structure tok- enization: Benchmarking and new recipe. InInternational Conference on Machine Learning, pages 73645–73670. PMLR, 2025. 10

    2025

  15. [15]

    Le, and Christopher D

    Kevin Clark, Minh-Thang Luong, Quoc V . Le, and Christopher D. Manning. ELECTRA: Pre-training text encoders as discriminators rather than generators. InICLR, 2020

    2020

  16. [16]

    Alphafold database expands to proteome-scale quaternary structures.bioRxiv, pages 2026–03, 2026

    Yewon Han, Maxim I Tsenkov, Niccolo AE Venanzi, Damian Bertoni, Sooyoung Cha, Alejandro Chacon, Nick Dietrich, Boris Fomitchev, Yonathan Goldtzvik, Darren Hsu, et al. Alphafold database expands to proteome-scale quaternary structures.bioRxiv, pages 2026–03, 2026

    2026

  17. [17]

    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

  18. [18]

    Accurate structure prediction of biomolecular interactions with alphafold 3.Nature, 630(8016):493–500, 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, 630(8016):493–500, 2024

    2024

  19. [19]

    Boltz-1 democratizing biomolecular interaction modeling.BioRxiv, pages 2024–11, 2025

    Jeremy Wohlwend, Gabriele Corso, Saro Passaro, Noah Getz, Mateo Reveiz, Ken Leidal, Wojtek Swiderski, Liam Atkinson, Tally Portnoi, Itamar Chinn, et al. Boltz-1 democratizing biomolecular interaction modeling.BioRxiv, pages 2024–11, 2025

    2024

  20. [20]

    Boltz-2: Towards accurate and efficient binding affinity prediction.BioRxiv, 2025

    Saro Passaro, Gabriele Corso, Jeremy Wohlwend, Mateo Reveiz, Stephan Thaler, Vignesh Ram Somnath, Noah Getz, Tally Portnoi, Julien Roy, Hannes Stark, et al. Boltz-2: Towards accurate and efficient binding affinity prediction.BioRxiv, 2025

    2025

  21. [21]

    Susskind, and Miguel Ángel Bautista

    Yuyang Wang, Jiarui Lu, Navdeep Jaitly, Joshua M. Susskind, and Miguel Ángel Bautista. Simplefold: Folding proteins is simpler than you think. InThe Fourteenth International Conference on Learning Representations, 2026

    2026

  22. [22]

    Evolutionary-scale prediction of atomic- level protein structure with a language model.Science, 379(6637):1123–1130, 2023

    Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Nikita Smetanin, Robert Verkuil, Ori Kabeli, Yaniv Shmueli, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Salvatore Candido, and Alexander Rives. Evolutionary-scale prediction of atomic- level protein structure with a language model.Science, 379(6637):1123–1130, 2023

    2023

  23. [23]

    Mc-bert: Efficient language pre-training via a meta controller.arXiv preprint arXiv:2006.05744, 2020

    Zhenhui Xu, Linyuan Gong, Guolin Ke, Di He, Shuxin Zheng, Liwei Wang, Jiang Bian, and Tie-Yan Liu. Mc-bert: Efficient language pre-training via a meta controller.arXiv preprint arXiv:2006.05744, 2020

    work page arXiv 2006

  24. [24]

    Alphafold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models.Nucleic acids research, 50(D1):D439–D444, 2022

    Mihaly Varadi, Stephen Anyango, Mandar Deshpande, Sreenath Nair, Cindy Natassia, Galabina Yordanova, David Yuan, Oana Stroe, Gemma Wood, Agata Laydon, et al. Alphafold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models.Nucleic acids research, 50(D1):D439–D444, 2022

    2022

  25. [25]

    Uniprot: the universal protein knowledgebase in 2023, 2023

    The UniProt Consortium. Uniprot: the universal protein knowledgebase in 2023, 2023

    2023

  26. [26]

    Mgnify: the microbiome analysis resource in 2020.Nucleic acids research, 48(D1):D570–D578, 2020

    Alex L Mitchell, Alexandre Almeida, Martin Beracochea, Miguel Boland, Josephine Burgin, Guy Cochrane, Michael R Crusoe, Varsha Kale, Simon C Potter, Lorna J Richardson, et al. Mgnify: the microbiome analysis resource in 2020.Nucleic acids research, 48(D1):D570–D578, 2020

    2020

  27. [27]

    Alphafold protein structure database in 2024: providing structure coverage for over 214 million protein sequences.Nucleic Acids Research, 52(D1):D368–D375, 01 2024

    Mihaly Varadi, Damian Bertoni, Paulyna Magana, Urmila Paramval, Ivanna Pidruchna, Malarvizhi Radhakrishnan, Maxim Tsenkov, Sreenath Nair, Milot Mirdita, Jingi Yeo, Oleg Kovalevskiy, Kathryn Tunyasuvunakool, Agata Laydon, Augustin Žídek, Hamish Tomlin- son, Dhavanthi Hariharan, Josh Abrahamson, Tim Green, John Jumper, Ewan Birney, Martin Steinegger, Demis ...

    2024

  28. [28]

    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

  29. [29]

    The interpretation of protein structures: estimation of static accessibility.Journal of molecular biology, 55(3):379–IN4, 1971

    Byungkook Lee and Frederic M Richards. The interpretation of protein structures: estimation of static accessibility.Journal of molecular biology, 55(3):379–IN4, 1971. 11

    1971

  30. [30]

    Environment and exposure to solvent of protein atoms

    Andrew Shrake and John A Rupley. Environment and exposure to solvent of protein atoms. lysozyme and insulin.Journal of molecular biology, 79(2):351–371, 1973

    1973

  31. [31]

    Plip: fully automated protein–ligand interaction profiler.Nucleic acids research, 43(W1):W443–W447, 2015

    Sebastian Salentin, Sven Schreiber, V Joachim Haupt, Melissa F Adasme, and Michael Schroeder. Plip: fully automated protein–ligand interaction profiler.Nucleic acids research, 43(W1):W443–W447, 2015

    2015

  32. [32]

    Arpeggio: a web server for calculating and visualising interatomic interactions in protein structures.Journal of molecular biology, 429(3):365–371, 2017

    Harry C Jubb, Alicia P Higueruelo, Bernardo Ochoa-Montaño, Will R Pitt, David B Ascher, and Tom L Blundell. Arpeggio: a web server for calculating and visualising interatomic interactions in protein structures.Journal of molecular biology, 429(3):365–371, 2017

    2017

  33. [33]

    Structure-based protein function prediction using graph convolutional networks.Nature communications, 12(1):3168, 2021

    Vladimir Gligorijevi´c, P Douglas Renfrew, Tomasz Kosciolek, Julia Koehler Leman, Daniel Berenberg, Tommi Vatanen, Chris Chandler, Bryn C Taylor, Ian M Fisk, Hera Vlamakis, et al. Structure-based protein function prediction using graph convolutional networks.Nature communications, 12(1):3168, 2021

    2021

  34. [34]

    Stephen K Burley, Charmi Bhikadiya, Chunxiao Bi, Sebastian Bittrich, Li Chen, Gregg V Crichlow, Cole H Christie, Kenneth Dalenberg, Luigi Di Costanzo, Jose M Duarte, et al. Rcsb protein data bank: powerful new tools for exploring 3d structures of biological macromolecules for basic and applied research and education in fundamental biology, biomedicine, bi...

    2021

  35. [35]

    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

  36. [36]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. InProceedings of the IEEE/CVF international conference on computer vision, pages 4195–4205, 2023

    2023

  37. [37]

    Peter J. A. Cock, Tiago Antao, Jeffrey T. Chang, Brad A. Chapman, Cymon J. Cox, Andrew Dalke, Iddo Friedberg, Thomas Hamelryck, Frank Kauff, Bartek Wilczynski, and Michiel J. L. de Hoon. Biopython: freely available python tools for computational molecular biology and bioinformatics.Bioinformatics, 25(11):1422–1423, 2009

    2009

  38. [38]

    A simple definition of structural regions in proteins and its use in analyzing interface evolution.Journal of molecular biology, 403(4):660–670, 2010

    Emmanuel D Levy. A simple definition of structural regions in proteins and its use in analyzing interface evolution.Journal of molecular biology, 403(4):660–670, 2010

    2010

  39. [39]

    Learning to design protein-protein interactions with enhanced generalization

    Anton Bushuiev, Roman Bushuiev, Petr Kouba, Anatolii Filkin, Marketa Gabrielova, Michal Gabriel, Jiri Sedlar, Tomas Pluskal, Jiri Damborsky, Stanislav Mazurenko, and Josef Sivic. Learning to design protein-protein interactions with enhanced generalization. InThe Twelfth International Conference on Learning Representations, 2024

    2024

  40. [40]

    Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features.Biopolymers, 22(12):2577–2637, 1983

    Wolfgang Kabsch and Christian Sander. Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features.Biopolymers, 22(12):2577–2637, 1983. 12 A Details in TRIPROREP A.1 Full-atom tokenization For each residue, the tokenizer computes heavy-atom geometry features from Atom37 coordinates. These features include (i) ...

    work page arXiv 1983