Accurate prediction of protein structures and interactions using a 3-track network
Open accessThe DOI matches its Crossref record; 85 of the 85 references it cites were found in a registry or an index; 13 of the 36 sources its claims cite were checked for support.
Abstract
DeepMind presented remarkably accurate protein structure predictions at the CASP14 conference. We explored network architectures incorporating related ideas and obtained the best performance with a 3-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The 3-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables rapid solution of challenging X-ray crystallography and cryo-EM structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate models of protein-protein complexes from sequence information alone, short circuiting traditional approaches which require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research. One-Sentence Summary: Accurate protein structure modeling enables rapid solution of structure determination problems and provides insights into biological function.
Overview
CoChat Summary
TL;DR
RoseTTAFold is a 3-track neural network that integrates 1D sequence, 2D distance-map, and 3D coordinate information to predict protein structures and interactions. It approaches AlphaFold2 accuracy in CASP14, solves challenging crystallography and cryo-EM modeling problems, and generates protein-complex models directly from sequence.
Key findings
- In CASP14, the 3-track RoseTTAFold outperformed the top 2 server groups (Zhang-server and BAKER-ROSETTASERVER), the BAKER human group, and the authors’ 2-track models, though its performance was still not as good as AlphaFold2.
- On CAMEO, over 69 medium and hard targets released May 15–June 19, 2021, RoseTTAFold outperformed all other evaluated servers, including Robetta, IntFold6-TS, BestSingleTemplate, and SWISS-MODEL.
- RoseTTAFold models enabled molecular-replacement solutions for four previously unsolved crystallographic datasets; for the cryo-EM p101 Gβγ binding domain, the Cα-RMSD between predicted and refined structure was 3.0 Å over the beta-sheets.
- Among 693 domains from human disease-related proteins, over one-third had predicted lDDT > 0.8, corresponding to an average Cα-RMSD of 2.6 Å on CASP14 targets; many protein-complex models had TM-score > 0.8, with complex generation taking ~30 min on a 24G TITAN RTX GPU.
Methods
- The 3-track network transforms and integrates information at the 1D sequence/MSA level, the 2D distance/orientation-map level, and the 3D backbone-coordinate level, with information flowing between tracks; training used discontinuous sequence crops spanning 260 residues.
- Final 3D structures were generated either by feeding predicted distance/orientation distributions into pyRosetta for all-atom models or by an end-to-end SE(3)-equivariant layer producing backbone coordinates directly.
- Evaluation included CASP14 targets, CAMEO blind server assessment, four X-ray molecular-replacement datasets, a cryo-EM PI3Kγ complex, GPCRs, 693 human disease-related domains, and paired-sequence complexes of known structure, using accuracy comparisons, Cα-RMSD, lDDT, TM-score, and MR/cryo-EM fit.
Limitations
- Stated: hardware memory limitations constrained model size and training; performance remained below AlphaFold2; the end-to-end version was less accurate than the pyRosetta version due to incomplete optimization and neglect of side-chain information; the accuracy predictor did not consistently identify models better than the rapid single-pass method; the network was trained on monomeric proteins, not complexes; and current RoseTTAFold uses only the top 1000 sequences due to memory limits.
- Stated: the Perceiver-based extension for using all available sequences gave promising initial results, but more training is required for rigorous comparison.
- Evident: CASP14 and CAMEO evaluations cover limited target sets and time windows, and the end-to-end model does not produce side chains.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures




Verification
Claims the paper makes
3 Valid · 8 Not valid · 14 Not checked · 44 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The bi-annual Critical Assessment of Structure (CASP) meetings have demonstrated that deep learning methods such as AlphaFold (1, 2) and trRosetta (3), that extract information from the large database of known protein structures in the PDB, outperform more traditional approaches that explicitly model the folding process.”
§ Accurate prediction of protein structures and interactions using a 3-track neural network > Abstract
What the sources say
- Not matchedcites [1] Jumper John Evans Richard Pritzel Alexander Green Tim Figurnov Michael Tunyasuvunakool Kathryn Ronneberger Olaf Bates Russ Augustin Žídek Alex Bridgland Clemens Meyer Fourteenth Critical Assessment of Techniques for Protein Structure Prediction
Its entry has no title and year that we can search by.
- Partly supportedcites [2] Anishchenko I Chidyausiku TM Ovchinnikov S Pellock SJ Baker D De novo protein design by deep network hallucination bioRxiv 2020 2020.07.22.211482 PMC9293396 34853475 sourceExperimental — accuracy not yet measured
The Introduction (page not provided) says deep networks improved protein-structure prediction accuracy, notes an AlphaFold observation at CASP13, and reports that trRosetta outperformed other methods in benchmark tests. The Conclusion says the network learned from native protein structures and contrasts it with physically based approaches. The paper does not establish the claimed CASP comparison in which both AlphaFold and trRosetta outperform approaches that explicitly model folding. (LLM, conf=0.82)
- Not checked — no full text of the sourcecites [3] Yang J Anishchenko I Park H Peng Z Ovchinnikov S Baker D Improved protein structure prediction using predicted interresidue orientations Proc Natl Acad Sci U S A 2020 117 1496 1503 31896580 PMC6983395 source
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“The RoseTTAFold CERS1 model for residues 98 to 304 (Pfam TLC domain) (29) includes six TMH that traverse the membrane in an up and down arrangement (Fig. 3E).”
§ Accurate prediction of protein structures and interactions using a 3-track neural network > Providing insights into biological function
What the sources say
- Could not verifycites [29] Cong Q Anishchenko I Ovchinnikov S Baker D Protein interaction networks revealed by proteome coevolution Science 2019 365 185 189 31296772 PMC6948103 sourceExperimental — accuracy not yet measured
The paper’s Abstract (page unavailable) describes coevolution-based protein interaction models in E. coli and M. tuberculosis, not a CERS1 model. Searches for CERS1, the Pfam TLC domain, RoseTTAFold, and six transmembrane helices found no passage in our copy addressing the cited domain identification or the stated membrane arrangement. (LLM, conf=0.76)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“in the case of SLP, only a distant model covering part of the N-terminal domain (38% of the sequence) was available in the PDB”
§ Accurate prediction of protein structures and interactions using a 3-track neural network > Enabling experimental protein structure determination
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“trRosetta failed to predict the correct fold with the same MSA input (fig. S6)”
§ Accurate prediction of protein structures and interactions using a 3-track neural network > Enabling experimental protein structure determination
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The 3-track models with attention operating at the 1D, 2D, and 3D levels and information flowing between the three levels were the best models we tested (Fig. 1B), clearly outperforming the top 2 server groups (Zhang-server and BAKER-ROSETTASERVER), BAKER human group (ranked second among all groups), and our 2-track attention models on CASP14 targets.”
§ Accurate prediction of protein structures and interactions using a 3-track neural network > Network architecture development
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
33 more claims
Create a free account to see all 69 claims and to verify the 2 references still open.
- It may state two separate claims as one. 24
- Probably not a substantial claim. 13
- The quote only partly states this claim. 2
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“RF has properties well suited for use in a protein design DDPM (Fig. 1a): it generates protein structures with high precision, operates on a rigid-frame representation of residues with rotational equivariance and has an architecture enabling conditioning on design specifications at the individual residue, inter-residue distance and orientation, and 3D coordinate levels.”
From De novo design of protein structure and function with RFdiffusion (2023) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“are the best and most accurate answer to the structure prediction problem to date”
From The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins (2024) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“DeepMind’s AlphaFold2 success in sequence-based protein structure prediction,4 as well as the RoseTTAFold70 open-source counterpart, and the publicly available AlphaFold2 prediction of all human proteins203 are major steps that benefit the scientific community.”
From AlphaFold, Artificial Intelligence (AI), and Allostery (2022) · cites this paper as [70]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The unprecedented accuracy of AlphaFold2,127 RosettaFold,128 and others129,130 offers, at least in principle, the opportunity to virtually screen any protein of interest with the structure-based approach, irrespective of its a priori ligandability (after AlphaFold2, the structural coverage of the human proteome increased from 48% to 76% of the total residues).131”
From Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence (2023) · cites this paper as [128]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The programs were also shown to accurately assess their confidence at the level of individual residues and interdomain contacts (27,28,62).”
From AI-based structure prediction empowers integrative structural analysis of human nuclear pores (2022) · cites this paper as [27]
Experimental — accuracy not yet measured
17 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability
Katoh, K., & Standley, D. M. (2013) · Molecular Biology and Evolution · doi:10.1093/molbev/mst010 · Cited by 46,209
Features and development of Coot
Emsley, P., Lohkamp, B., Scott, W. G., & Cowtan, K. (2010) · Acta Crystallographica Section D Biological Crystallography · doi:10.1107/s0907444910007493 · Cited by 29,915
Clustal W and Clustal X version 2.0
Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam, H., Valentin, F., Wallace, I. M., Wilm, A., Lopez, R., et al. (2007) · Bioinformatics · doi:10.1093/bioinformatics/btm404 · Cited by 28,358
Phaser crystallographic software
McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., & Read, R. J. (2007) · Journal of Applied Crystallography · doi:10.1107/s0021889807021206 · Cited by 20,612
Self-citationSWISS-MODEL: homology modelling of protein structures and complexes
Waterhouse, A., Bertoni, M., Bienert, S., Studer, G., Tauriello, G., Gumienny, R., Heer, F. T., de Beer, T. A. P., Rempfer, C., Bordoli, L., et al. (2018) · Nucleic Acids Research · doi:10.1093/nar/gky427 · Cited by 13,067 · cited 2×
“Fortunately,theContinuousAutomated ModelEvaluation(CAMEO)experiment( 7 ) tests structure prediction servers blindly on protein structures as they are submitted to the PDB. RoseTTAFold has been evaluated since 15 May 2021 on CAMEO; over the 69 medium and hard targets released during this time (15 May 2021 to 19 June 20…”
UCSF ChimeraX: Structure visualization for researchers, educators, and developers
Pettersen, E. F., Goddard, T. D., Huang, C. C., Meng, E. C., Couch, G. S., Croll, T. I., Morris, J. H., & Ferrin, T. E. (2021) · Protein Science · doi:10.1002/pro.3943 · Cited by 9,514
Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., et al. (2019) · Acta Crystallographica Section D Structural Biology · doi:10.1107/s2059798319011471 · Cited by 7,953
Self-citationUniProt: the Universal Protein knowledgebase
Apweiler, R. (2004) · Nucleic Acids Research · doi:10.1093/nar/gkh131 · Cited by 7,498
Towards automated crystallographic structure refinement with phenix.refine
Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H., & Adams, P. D. (2012) · Acta Crystallographica Section D Biological Crystallography · doi:10.1107/s0907444912001308 · Cited by 5,511
Self-citation
Research works printed but not linked: 2
- Bunkóczi G Read RJ phenix. ensembler: a tool for multiple superposition Comput Crystallogr Newsl 2011 2 8 9
- Vagin A Lebedev A ACTA CRYSTALLOGRAPHICA A-FOUNDATION AND ADVANCES INT UNION CRYSTALLOGRAPHY 2 ABBEY SQ, CHESTER, CH1 2HU ENGLAND 2015 71 S19
Cited by
Items 1,201–1,300 of 5,635
ModelCIF: An Extension of PDBx/mmCIF Data Representation for Computed Structure Models
Vallat, B., Tauriello, G., Bienert, S., Haas, J., Webb, B. M., Žídek, A., Zheng, W., Peisach, E., Piehl, D. W., Anischanka, I., et al. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.168021 · Cited by 26
Self-citationCryo-electron Microscopy Reveals the Structure of the Nuclear Pore Complex
Tai, L., Yin, G., Sun, F., & Zhu, Y. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.168051 · Cited by 31
The confluence of machine learning and multiscale simulations
Bhatia, H., Aydin, F., Carpenter, T. S., Lightstone, F. C., Bremer, P., Ingólfsson, H. I., Nissley, D. V., & Streitz, F. H. (2023) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2023.102569 · Cited by 29
Mutz, P., Resch, W., Faure, G., Senkevich, T. G., Koonin, E. V., & Moss, B. (2023) · mBio · doi:10.1128/mbio.00408-23 · Cited by 27
“…in by the new arti fi cial intelligence-based methods, namely, AlphaFold and RosettaFold, has dramatically expanded the opportunities for detecting homologous relationships among proteins via the comparison of protein structure models to experimentally solved structures or other models (4, 5).”
BackgroundMethodInfluentialDesigned Rubredoxin miniature in a fully artificial electron chain triggered by visible light
Chino, M., Costanzo, L. D., Leone, L., Gatta, S. L., Famulari, A., Chiesa, M., Lombardi, A., & Pavone, V. (2023) · Nature Communications · doi:10.1038/s41467-023-37941-8 · Cited by 29
“The combination of powerful computational tools 48,49 , and more recently machine learning 50,51 , together with the genome palette (e.g., directed evolution and phage/yeast display) 52,53 is signi fi cantly helping protein designers in increasing success rate, adapting the protein to host the…”
Li, F., Deng, H., & Renata, H. (2023) · Nature Synthesis · doi:10.1038/s44160-023-00358-8 · Cited by 26
“Looking to the future, emerging tools in computational modelling 63 , deep learning 64 and high-throughput screening 65 will accelerate the pace of enzyme discovery to provide us with a broader set of tools for applying chemoenzymatic methods in natural product medicinal chemistry.”
MethodExperimental phasing opportunities for macromolecular crystallography at very long wavelengths
Omari, K. E., Duman, R., Mykhaylyk, V. B., Orr, C. M., Latimer-Smith, M., Winter, G., Grama, V., Qu, F., Bountra, K., Kwong, H.-S., et al. (2023) · Communications Chemistry · doi:10.1038/s42004-023-01014-0 · Cited by 27
“More recently, the utilization of machine learning techniques by AlphaFold2 4 or RoseTTAFold 5 has revolutionized protein model prediction, already having a huge impact on biological research at various levels, for example by providing search models for molecular replacement (MR) to overcome the phase problem in macro…”
Enzymatic β-elimination in natural product O- and C-glycoside deglycosylation
Bitter, J., Pfeiffer, M., Borg, A. J. E., Kuhlmann, K., Pavkov‐Keller, T., Sánchez‐Murcia, P. A., & Nidetzky, B. (2023) · Nature Communications · doi:10.1038/s41467-023-42750-0 · Cited by 30
“The At OGE structure suggests convergent evolution of Mn 2 + active site for β -elimination of 3-keto-glycosides The X-raystructure of At OGE wasdetermined to 2.0Å resolution using molecular replacement based on a RoseTTAFold 52 calculated ab initio model of the enzyme.”
InfluentialSelf-citationBiocatalysis of CO2 and CH4: Key enzymes and challenges
Li, A., Cao, X., Fu, R., Guo, S., & Fei, Q. (2024) · Biotechnology Advances · doi:10.1016/j.biotechadv.2024.108347 · Cited by 40
Luu, R. K., Arevalo, S. E., Lu, W., Ni, B., Yang, Z., Shen, S. C.-Y., Berkovich, J., Hsu, Y.-C., Zan, S., & Buehler, M. J. (2024) · An MIT Exploration of Generative AI · doi:10.21428/e4baedd9.33bd7449 · Cited by 26
Engineering is evolution: a perspective on design processes to engineer biology
Castle, S. D., Stock, M., & Gorochowski, T. E. (2024) · Nature Communications · doi:10.1038/s41467-024-48000-1 · Cited by 31
The structural landscape and diversity of Pyricularia oryzae MAX effectors revisited
Lahfa, M., Barthe, P., de Guillen, K. M., Césari, S., Raji, M., Kroj, T., Naour—Vernet, M. L., Hoh, F., Gladieux, P., Roumestand, C., et al. (2024) · PLoS Pathogens · doi:10.1371/journal.ppat.1012176 · Cited by 26
Structure-based prediction of protein-nucleic acid binding using graph neural networks
Sagendorf, J. M., Mitra, R., Huang, J., Chen, X. S., & Rohs, R. (2024) · Biophysical Reviews · doi:10.1007/s12551-024-01201-w · Cited by 27
“In addition to experimentally determined structures, advances in protein structure prediction (Baek et al. 2021; Jumper et al. 2021; Ahdritz et al. 2024) offer vast amounts of accurately predicted protein structures that can be analyzed with computational methods.”
“Ready-to-use” immunosensor for the detection of small molecules with fast readout
Ding, Y., Cui, P., Chen, H., Li, J., Huang, L., González‐Sapienza, G., Hammock, B. D., Wang, M., & Hua, X. (2022) · Biosensors and Bioelectronics · doi:10.1016/j.bios.2022.113968 · Cited by 26
Jamasb, A. R., Viñas, R., Ma, E. J., Harris, C., Huang, K., Hall, D., Lió, P., & Blundell, T. L. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.07.15.204701 · Cited by 39
A Foundational Atlas of Autism Protein Interactions Reveals Molecular ConvergencePreprint
Wang, B., Vartak, R. S., Zaltsman, Y., Naing, Z. Z. C., Hennick, K. M., Polacco, B. J., Bashir, A., Eckhardt, M., Bouhaddou, M., Xu, J., et al. (2023) · SSRN Electronic Journal · doi:10.1101/2023.12.03.569805 · Cited by 31
Watkins, M. B., Wang, H., Burnim, A., & Ando, N. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2302531120 · Cited by 26
Spoendlin, F. C., Abanades, B., Raybould, M. I. J., Wong, W. K., Georges, G., & Deane, C. M. (2023) · doi:10.1101/2023.06.09.543890 · Cited by 26
Limits and potential of combined folding and docking
Pozzati, G., Zhu, W., Bassot, C., Lamb, J., Kundrotas, P., & Elofsson, A. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btab760 · Cited by 25
Sen, N., Anishchenko, I., Bordin, N., Sillitoe, I., Velankar, S., Baker, D., & Orengo, C. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac187 · Cited by 25
Self-citationHybridGCN for protein solubility prediction with adaptive weighting of multiple features
Chen, L., Wu, R., Zhou, F., Zhang, H., & Liu, J. K. (2023) · Journal of Cheminformatics · doi:10.1186/s13321-023-00788-8 · Cited by 25
“Deep learning models have achieved the state-of-the-art (SOTA) performance on various protein engineering tasks, including structure prediction [12, 13], protein design [14, 15], protein binder design [16], stability prediction [17, 18], and solubility prediction [9, 19].”
BackgroundStructural Bioinformatics and Deep Learning of Metalloproteins: Recent Advances and Applications
Andreini, C., & Rosato, A. (2022) · International Journal of Molecular Sciences · doi:10.3390/ijms23147684 · Cited by 25
“…on 5 July 2022) A database of epitopes, drug binding site and metal binding sites in viral proteins of known 3D structure [71] InterMetalDB https://intermetaldb.biotech.uni.wroc.pl (accessedon5July2022) A database of MBSs occurring at macromolecular interfaces, built on biological assemblies [30]”
BackgroundReliability of AlphaFold2 Models in Virtual Drug Screening: A Focus on Selected Class A GPCRs
Alhumaid, N. K., & Tawfik, E. A. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms251810139 · Cited by 32
“Next-generation tools, including AlphaFold2 (AF2) [3], AlphaFill [4], RoseTTAFold [5], and many others, have established a significant breakthrough in structural biology.”
MethodHyperdimensional computing: A fast, robust, and interpretable paradigm for biological data
Stock, M., Criekinge, W. V., Boeckaerts, D., Taelman, S., Haeverbeke, M. V., Dewulf, P., & Baets, B. D. (2024) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1012426 · Cited by 28
“Prominently, the development and use of complex DL models such as AlphaFold [5], ESMFold [6] and RoseTTAFold [7] represent a paradigm shift in protein structure prediction.”
MethodMDRepo—an open data warehouse for community-contributed molecular dynamics simulations of proteins
Roy, A., Ward, E., Choi, I., Cosi, M., Edgin, T., Hughes, T., Islam, M. S., Khan, A. M., Kolekar, A., Rayl, M. L., et al. (2024) · Nucleic Acids Research · doi:10.1093/nar/gkae1109 · Cited by 31
A unified evolution-driven deep learning framework for virus variation driver prediction
Nie, Z., Liu, X., Chen, J., Wang, Z., Liu, Y., Si, H.-R., Dong, T., Xu, F., Song, G., Wang, Y., et al. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-024-00966-9 · Cited by 30
Machine learning to predict de novo protein–protein interactions
Gaínza, P., Bunker, R. D., Townson, S. A., & Castle, J. C. (2025) · Trends in biotechnology · doi:10.1016/j.tibtech.2025.04.013 · Cited by 27
AI meets physics in computational structure-based drug discovery for GPCRs
Michino, M., Vendôme, J., & Kufareva, I. (2025) · npj Drug Discovery. · doi:10.1038/s44386-025-00019-0 · Cited by 33
Synthetic lethality in cancer drug discovery: challenges and opportunities
Gonçalves, E., Ryan, C. J., & Adams, D. J. (2025) · Nature Reviews Drug Discovery · doi:10.1038/s41573-025-01273-7 · Cited by 34
“The break-through structural biology tools AlphaFold 181,182 and RoseTTAFold 183,184 have revolutionized our ability to rapidly predict protein structures from sequences with high accuracy, facilitating structure-based drug design.”
Jeyaraj, G., Rajendran, A. K., Sathishkumar, K., Almutairi, B. O., Vadivelu, A., Chokkakula, S., Tu, Y., & Xie, W. (2025) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2025.1688455 · Cited by 25
A Computational Community Blind Challenge on Pan-Coronavirus Drug Discovery Data
MacDermott-Opeskin, H., Scheen, J., Wognum, C., Horton, J. T., West, D., Payne, A. M., Castellanos, M. A., Colby, S., Griffen, E., Cousins, D. L., et al. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c02106 · Cited by 37
adabmDCA: adaptive Boltzmann machine learning for biological sequences
Muntoni, A. P., Pagnani, A., Weigt, M., & Zamponi, F. (2021) · BMC Bioinformatics · doi:10.1186/s12859-021-04441-9 · Cited by 36
“To compensate for this experimental gap, in silico protein and RNA domains modeling has shown an incredible predictive power in determining their structure [3, 4].”
BackgroundVaccines 2020: The era of the digital vaccine is here
Pizza, M., Pecetta, S., & Rappuoli, R. (2021) · Science Translational Medicine · doi:10.1126/scitranslmed.abm3249 · Cited by 29
In Mitosis You Are Not: The NIMA Family of Kinases in Aspergillus, Yeast, and Mammals
Bachus, S., Graves, D., Fulham, L., Akkerman, N., Stephanson, C., Shieh, J., & Pelka, P. (2022) · International Journal of Molecular Sciences · doi:10.3390/ijms23074041 · Cited by 28
“Amino acid sequences were input into Robetta’s RoseTTAFold [29] and Phyre2′s Intensive [30] algorithms, with the resulting structures visualized in the first quarter of 2022.”
MethodThe prospects and opportunities of protein structure prediction with AI
Tunyasuvunakool, K. (2022) · Nature Reviews Molecular Cell Biology · doi:10.1038/s41580-022-00488-5 · Cited by 27
Identification of Functional Spo0A Residues Critical for Sporulation in Clostridioides difficile
DiCandia, M. A., Edwards, A. N., Jones, J. B., Swaim, G. L., Mills, B. D., & McBride, S. M. (2022) · Journal of Molecular Biology · doi:10.1016/j.jmb.2022.167641 · Cited by 25
AlphaFold2 and CryoEM: Revisiting CryoEM modeling in near-atomic resolution density maps
Hryc, C. F., & Baker, M. L. (2022) · iScience · doi:10.1016/j.isci.2022.104496 · Cited by 24
“The success of AlphaFold2 and RoseTTAFold has undoubtedly inspired a fresh approach for model building in cryoEM density maps; however, the utility, limits, and expectations of such a combined approach must be first evaluated.”
InfluentialSmorodina, E., Tao, F., Qing, R., Jin, D., Yang, S., & Zhang, S. (2022) · QRB Discovery · doi:10.1017/qrd.2022.6 · Cited by 29
Structure and evolution of neuronal wiring receptors and ligands
Cortés, E. C., Pak, J. S., & Özkan, E. (2022) · Developmental Dynamics · doi:10.1002/dvdy.512 · Cited by 29
Centromere drive: model systems and experimental progress
Dudka, D., & Lampson, M. A. (2022) · Chromosome Research · doi:10.1007/s10577-022-09696-3 · Cited by 27
“One approach to probe the evolutionary landscapes is to examine recurrently changing residues or regions in the context of protein structure predictions at the level of individual proteins (Baek et al. 2021; Jumper et al. 2021) or larger complexes (Hum-phreys et al. 2021).”
BackgroundQian, S., Heller, W. T., Chen, W., Christianson, A. D., Do, C., Wang, Y., Lin, J., Huegle, T., Jiang, C., Boone, C., et al. (2022) · Review of Scientific Instruments · doi:10.1063/5.0090527 · Cited by 26
Enhancer-LSTMAtt: A Bi-LSTM and Attention-Based Deep Learning Method for Enhancer Recognition
Huang, G., Luo, W., Zhang, G., Zheng, P., Yao, Y., Lyu, J., Liu, Y., & Wei, D. (2022) · Biomolecules · doi:10.3390/biom12070995 · Cited by 28
Zhang, H., Zhang, T., Saravanan, K. M., Liao, L., Wu, H., Zhang, H., Zhang, H., Pan, Y. F., Wu, X., & Wei, Y. (2022) · Methods · doi:10.1016/j.ymeth.2022.07.009 · Cited by 25
Solution structure of recombinant Pvfp-5β reveals insights into mussel adhesion
Morando, M. A., Venturella, F., Sollazzo, M., Monaca, E., Sabbatella, R., Vetri, V., Passantino, R., Pastore, A., & Alfano, C. (2022) · Communications Biology · doi:10.1038/s42003-022-03699-w · Cited by 27
Deciphering the Folding Mechanism of Proteins G and L and Their Mutants
Chang, L., & Pérez, A. (2022) · Journal of the American Chemical Society · doi:10.1021/jacs.2c04488 · Cited by 26
An AI-Based Curling Game System for Winter Olympics
Shi, X., Wang, Q., Wang, C., Wang, R., Zheng, L., Qian, C., & Tang, W. (2022) · Research · doi:10.34133/2022/9805054 · Cited by 28
In vitro and in silico assessment of GPER-dependent neurocytotoxicity of emerging bisphenols
Wang, L., Huang, C., Li, L., Pang, Q., Wang, C., & Fan, R. (2022) · The Science of The Total Environment · doi:10.1016/j.scitotenv.2022.160762 · Cited by 25
Yin, P., Zhang, Y., Yang, L., & Feng, Y. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.167996 · Cited by 25
Bioinformatics approaches for unveiling virus-host interactions
Iuchi, H., Kawasaki, J., Kubo, K., Fukunaga, T., Hokao, K., Yokoyama, G., Ichinose, A., Suga, K., & Hamada, M. (2023) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2023.02.044 · Cited by 33
“Recently, AlphaFold [44], RoseTTAFold [45] and the alignment-free approach [46] achieved highly accurate protein structure prediction.”
Protein Oligomer Engineering: A New Frontier for Studying Protein Structure, Function, and Toxicity
Liu, C., & Luo, J. (2023) · Angewandte Chemie International Edition · doi:10.1002/anie.202216480 · Cited by 29
Benchmarking machine learning robustness in Covid-19 genome sequence classification
Ali, S., Sahoo, B., Zelikovsky, A., Chen, P., & Patterson, M. (2023) · Scientific Reports · doi:10.1038/s41598-023-31368-3 · Cited by 29
“Even closer to the subject of our paper — protein sequences – the authors of [22] show that methods, such as AlphaFold [23] and RoseTTAFold [24] which employ deep neural networks to predict protein conformation are not robust: producing drastically different protein structures as a result of very small biologically me…”
MethodCryoRes: Local Resolution Estimation of Cryo-EM Density Maps by Deep Learning
Dai, M., Dong, Z.-E., Xu, K., & Zhang, Q. C. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.168059 · Cited by 25
Sifri, C., Hoeg, L., Durocher, D., & Setiaputra, D. (2023) · EMBO Reports · doi:10.15252/embr.202356834 · Cited by 26
“The development of deep-learning based protein structure prediction systems such as AlphaFold2-Multimer and RoseTTAFold (Baek et al , 2021; preprint: Evans et al , 2021) made rapid, scalable, and highly accurate protein – protein structure prediction possible.”
Design and synthesis of bioinspired nanomaterials for biomedical application
Tang, K., Xue, J., Zhu, Y., & Wu, C. (2023) · Wiley Interdisciplinary Reviews Nanomedicine and Nanobiotechnology · doi:10.1002/wnan.1914 · Cited by 29
Intrinsic structural dynamics dictate enzymatic activity and inhibition
Shukla, V. K., Siemons, L., & Hansen, D. F. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2310910120 · Cited by 29
“In the era of computational structural predictions (54, 55), the prevalence of protein flexibility and heterogeneity has become increasingly clear.”
BackgroundModule-Based Polyketide Synthase Engineering for de Novo Polyketide Biosynthesis
Nava, A. A., Roberts, J. B., Haushalter, R. W., Wang, Z., & Keasling, J. D. (2023) · ACS Synthetic Biology · doi:10.1021/acssynbio.3c00282 · Cited by 26
Chicowski, A. S., Bredow, M., Utiyama, A. S., Marcelino‐Guimarães, F. C., & Whitham, S. A. (2023) · Plant Biotechnology Journal · doi:10.1111/pbi.14206 · Cited by 29
“However, template-free structural modelling using RoseTTAFold (Baek et al ., 2021) or AlphaFold (Jumper et al ., 2021) can allow predictions to be made about how an effector executes its functions beyond sequence analysis.”
MethodSIGMA leverages protein structural information to predict the pathogenicity of missense variants
Zhao, H., Du, H., Zhao, S., Chen, Z., Li, Y., Xu, K., Liu, B., Cheng, X., Wen, W., Li, G., et al. (2024) · Cell Reports Methods · doi:10.1016/j.crmeth.2023.100687 · Cited by 24
Protein Design by Directed Evolution Guided by Large Language Models
Tran, T. P., & Hy, T. S. (2024) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2024.3439690 · Cited by 39
Singh, J., Paliwal, K. K., Litfin, T., Singh, J., & Zhou, Y. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac421 · Cited by 28
Efficient and scalable de novo protein design using a relaxed sequence space
Frank, C., Khoshouei, A., de Stigter, Y., Schiewitz, D., Feng, S., Ovchinnikov, S., & Dietz, H. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.02.24.529906 · Cited by 26
Self-citation(2022) · doi:10.1073/pnas.2203176119 · Cited by 25
“protein–protein complex structure predictions can now be generated with increasing accuracy for larger families (31, 48, 49), we reasoned that the sequence information from the larger FidL cluster (Fig.”
BackgroundMethodInfluentialSelf-citationLiu, J., Zhao, K., & Zhang, G. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbac507 · Cited by 25
Gene amplifications cause high-level resistance against albicidin in gram-negative bacteria
Saathoff, M., Kosol, S., Semmler, T., Tedin, K., Dimos, N., Kupke, J., Seidel, M., Ghazisaeedi, F., Jonske, M. C., Wolf, S. A., et al. (2023) · PLOS Biology · doi:10.1371/journal.pbio.3002186 · Cited by 25
“When generating homology models by RoseTTAFold (25) and Phyre2 (26), the E. coli transcription factor Rob (PDB-ID: 1d5y) (27) and GyrI (PDB-ID: 1jyha) (24) were identified as highest-ranking structural homologs (100 and 99.9% confidence, respectively; Fig.”
MethodInfluential(2022) · doi:10.1101/2022.08.12.503819 · Cited by 25
A universal framework for accurate and efficient geometric deep learning of molecular systems
Zhang, S., Liu, Y., & Xie, L. (2023) · Scientific Reports · doi:10.1038/s41598-023-46382-8 · Cited by 24
“For example, GNNs designed for modeling proteins may include operations that are specific to the structural characteristics of amino acids 19, 20 , which are not relevant for other types of molecules.”
BackgroundSimpkin, A. J., Thomas, J. M. H., Keegan, R. M., & Rigden, D. J. (2022) · Acta Crystallographica Section D Structural Biology · doi:10.1107/s2059798322003576 · Cited by 24
Kumar, H., & Kim, P. (2024) · Clinical and Translational Medicine · doi:10.1002/ctm2.1789 · Cited by 29
“17,18 The 3D structure of fusion proteins is crucial in the computer-aided drug design process.”
BackgroundAlphaFold and what is next: bridging functional, systems and structural biology
Szczepski, K., & Jaremko, Ł. (2025) · Expert Review of Proteomics · doi:10.1080/14789450.2025.2456046 · Cited by 29
“RosettaFold, developed by David Baker’s lab [16], employs a three-track neural network where information flows back and forth between the one-dimensional (1D) sequence level, the two-dimensional (2D) distance map, and the three-dimensional (3D) coordinates.”
Mahling, R. W., Hegyi, B., Cullen, E. R., Cho, T. M., Rodriques, A., Fossier, L., Yehya, M., Yang, L., Chen, B.-X., Katchman, A. N., et al. (2025) · Cell · doi:10.1016/j.cell.2025.07.038 · Cited by 29
Self-citationAbbasi, K., Razzaghi, P., Gharizadeh, A., Ghareyazi, A., Dehnad, A., Rabiee, H. R., & Mofrad, M. R. K. (2025) · Pharmacological Reviews · doi:10.1016/j.pharmr.2025.100095 · Cited by 30
Artificial intelligence in bioinformatics: a survey
Jiang, J., Li, Y., Cao, S., Shan, Y., Liu, Y., Fei, T., Yu, Y., Feng, Y., Li, Y., Li, Y., et al. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf576 · Cited by 33
“At the structural level, next-generation Transformer architectures and their hybrid algorithms allow rapid inference of protein and n ucleic acid 3D conformations and interaction interfaces directl y from sequence information [10, 18].”
FLOWR: flow matching for structure-aware de novo, interaction- and fragment-based ligand generation
Cremer, J., Irwin, R., Tibo, A., Janet, J. P., Olsson, S., & Clevert, D.-A. (2026) · Nature Computational Science · doi:10.1038/s43588-026-00998-8 · Cited by 36
“Among generative modeling techniques, diffusion models have emerged as particularly promising tools for de novo ligand design.”
Deepak, R. N. V. K., Verma, R. K., Hartono, Y. D., Yew, W. S., & Fan, H. (2021) · Pharmaceuticals · doi:10.3390/ph15010012 · Cited by 25
Korhonen, P. K., Kinkar, L., Young, N. D., Cai, H., Lightowlers, M. W., Gauci, C. G., Jabbar, A., Chang, B. C. H., Wang, T., Hofmann, A., et al. (2022) · Communications Biology · doi:10.1038/s42003-022-03125-1 · Cited by 28
Structures of synthetic helical filaments and tubes based on peptide and peptido-mimetic polymers
Miller, J. G., Hughes, S. A., Modlin, C., & Conticello, V. P. (2022) · Quarterly Reviews of Biophysics · doi:10.1017/s0033583522000014 · Cited by 24
Dominguez, M. J., McCord, J. J., & Sutton, R. B. (2022) · PLOS ONE · doi:10.1371/journal.pone.0270188 · Cited by 24
“The six ferlin C2B models from RoseTTAFold superimpose on each other with an RMSD less than 1.0 Å (Fig.”
MethodInfluentialRed Rubiscos and opportunities for engineering green plants
Oh, Z. G., Askey, B. C., & Gunn, L. H. (2022) · Journal of Experimental Botany · doi:10.1093/jxb/erac349 · Cited by 25
Sichel, S. R., Bratton, B. P., & Salama, N. R. (2022) · eLife · doi:10.7554/elife.80111 · Cited by 27
Sugiki, S., Niide, T., Toya, Y., & Shimizu, H. (2022) · ACS Synthetic Biology · doi:10.1021/acssynbio.2c00315 · Cited by 29
“(C) MaeB model structure created by RoseTTAFold (2) through the Robetta server (https://robetta.bakerlab.org/).”
MethodInfluentialSmall Peptides in the Detection of Mycotoxins and Their Potential Applications in Mycotoxin Removal
Zhao, Z., Zhang, Z., Zhang, H., & Liang, Z. (2022) · Toxins · doi:10.3390/toxins14110795 · Cited by 24
Squitti, R., Catalli, C., Gigante, L., Marianetti, M., Rosari, M., Mariani, S., Bucossi, S., Mastromoro, G., Ventriglia, M., Simonelli, I., et al. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms24076377 · Cited by 26
“To predict the three-dimensional protein structure, this server uses homologous sequences or homology modeling [63].”
MethodProtein structure dynamics by crosslinking mass spectrometry
Chen, Z. A., & Rappsilber, J. (2023) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2023.102599 · Cited by 24
Recent Advances in Computer-Aided Structure-Based Drug Design on Ion Channels
Pliushcheuskaya, P., & Künze, G. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms24119226 · Cited by 29
“However, with the advance in the artificial intelligence field, these methods were superseded by methods like AlphaFold [66], RoseTTAFold [124], or OmegaFold [125], which can accurately predict protein three-dimensional structures in the absence of template information and also provide a likelihood…”
Machine learning modeling of RNA structures: methods, challenges and future perspectives
Wu, K., Zou, J., & Chang, H. Y. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad210 · Cited by 27
Deep transfer learning for inter-chain contact predictions of transmembrane protein complexes
Lin, P., Yan, Y., Tao, H., & Huang, S. (2023) · Nature Communications · doi:10.1038/s41467-023-40426-3 · Cited by 29
“For the structure feature of monomer structures, we applied the radial basis function to convert the intra-protein distance of monomer as input features 55 .”
MethodGao, W., Li, T., Zhou, H., Ju, J., & Yin, H. (2023) · Journal of Biological Chemistry · doi:10.1016/j.jbc.2023.105573 · Cited by 24
“To obtain a structural impression of Mt LPMO9G, we used RoseTTAFold to generate the three-dimensional structures of Mt LPMO9G and Mt LPMO9G-CD, respectively.”
MethodInfluentialLi, M., Zhang, T., Li, C., Gao, W., Liu, Z., & Miao, M. (2024) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2024.129316 · Cited by 29
Zhang, Z., Cai, Y., Zhang, B., Zheng, W., Freddolino, L., Zhang, G., & Zhou, X. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae113 · Cited by 23
Establishing cell suitability for high-level production of licorice triterpenoids in yeast
Sun, W., Wan, S., Liu, C., Wang, R., Zhang, H., Qin, L., Wang, R., Lv, B., & Li, C. (2024) · Acta Pharmaceutica Sinica B · doi:10.1016/j.apsb.2024.04.032 · Cited by 24
“To investigate the potential influence of substrate tunnel microenvironment on substrate binding and catalytic activity, we generated the 3D model of Uni25647 using AlphaFold2 30 .”
Genetic control over biogenic crystal morphogenesis in zebrafish
Deis, R., Lerer-Goldshtein, T., Baiko, O., Eyal, Z., Brenman-Begin, D., Goldsmith, M., Kaufmann, S., Heinig, U., Dong, Y., Lushchekina, S., et al. (2025) · Nature Chemical Biology · doi:10.1038/s41589-024-01722-1 · Cited by 24
“To elucidate the observed differences in catalytic activities of D. rerio PNPases, we compared their sequences and computationally predicted structures using several algorithms 47–51 , including Alpha-Fold2 (refs.”
Application Evaluation and Performance-Directed Improvement of the Native and Engineered Biosensors
Li, M., Chen, Z., & Huo, Y. (2024) · ACS Sensors · doi:10.1021/acssensors.4c01072 · Cited by 24
Assessing and enhancing foldability in designed proteins
Listov, D., Lipsh‐Sokolik, R., Rosset, S., Yang, C., Correia, B. E., & Fleishman, S. J. (2022) · Protein Science · doi:10.1002/pro.4400 · Cited by 24
Ni, B., Glaser, B., & Taheri-Mousavi, S. M. (2025) · npj Computational Materials · doi:10.1038/s41524-025-01768-2 · Cited by 23
Rational exploration of fold atlas for human solute carrier proteins
Xie, T., Chi, X., Huang, B., Ye, F., Zhou, Q., & Huang, J. (2022) · Structure · doi:10.1016/j.str.2022.05.015 · Cited by 23
Rapid and accurate prediction of protein homo-oligomer symmetry using Seq2Symm
Kshirsagar, M., Meller, A., Humphreys, I. R., Sledzieski, S., Xu, Y., Dodhia, R., Horvitz, E., Berger, B., Bowman, G. R., Ferres, J. L., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-57148-3 · Cited by 23
“However,usingAlphaFold 11 orRoseTTAFold 10 forabinitio oligomeric state prediction poses signi fi cant computational challenges, as it requires running inference for each potential number of chains to score various copy number models, and is generally limited to proteins with high-quality MSAs.”
Self-citationKulke, M., Kurtz, E., Boren, D. M., Olson, D. M., Koenig, A. M., Hoffmann-Benning, S., & Vermaas, J. V. (2023) · Plant Science · doi:10.1016/j.plantsci.2023.111900 · Cited by 18
Hafez, M., Elarab, S. M. E., Tohamy, H. G., & El‐Far, A. H. (2024) · Scientific Reports · doi:10.1038/s41598-024-62780-y · Cited by 24
“While the 3D structure of the interleukin-6 receptor (IL-6R) was produced by the Robetta server 40 .”
MethodExploring the conformational landscape of protein kinases
Gough, N. R., & Kalodimos, C. G. (2024) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2024.102890 · Cited by 25
PDBench: evaluating computational methods for protein-sequence design
Castorina, L. V., Petrenas, R., Subr, K., & Wood, C. W. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad027 · Cited by 21
Metadata
- Authors
- Baek, M., DiMaio, F., Anishchenko, I., Dauparas, J., Ovchinnikov, S., Lee, G. R., Wang, J., Cong, Q., Kinch, L. N., Schaeffer, R. D., Millán, C., Park, H., Adams, C., Glassman, C. R., DeGiovanni, A., Pereira, J. H., Rodrigues, A. V., van Dijk, A. A., Ebrecht, A. C., Opperman, D. J., Sagmeister, T., Buhlheller, C., Pavkov-Keller, T., Rathinaswamy, M. K., Dalwadi, U., Yip, C. K., Burke, J. E., Garcia, K. C., Grishin, N. V., Adams, P. D., Read, R. J., & Baker, D.
- Affiliations
- University of Washington (US)
- Harvard University (US)
- The University of Texas Southwestern Medical Center (US)
- Howard Hughes Medical Institute (US)
- University of Cambridge (GB)
- Stanford Medicine (US)
- Lawrence Berkeley National Laboratory (US)
- North-West University (ZA)
- University of the Free State (ZA)
- University of Graz (AT)
- University of Victoria (CA)
- University of British Columbia (CA)
- Corresponding author
- Baker, David
- ORCID
- Baek, Minkyung
- DiMaio, Frank
- Anishchenko, Ivan
- Dauparas, Justas
- Ovchinnikov, Sergey
- Lee, Gyu Rie
- Wang, Jue
- Cong, Qian
- Kinch, Lisa N.
- Schaeffer, R. Dustin
- Millán, Claudia
- Park, Hahnbeom
- Adams, Carson
- Glassman, Caleb R.
- Pereira, Jose H.
- Rodrigues, Andria V.
- van Dijk, Alberdina A.
- Ebrecht, Ana C.
- Opperman, Diederik J.
- Sagmeister, Theo
- Buhlheller, Christoph
- Rathinaswamy, Manoj K
- Dalwadi, Udit
- Yip, Calvin K
- Burke, John E
- Grishin, Nick V.
- Adams, Paul D.
- Read, Randy J.
- Baker, David
- Venue
- Science
- Publisher
- openRxiv
- Published
- 2021-06-15
- Volume
- 373
- Issue
- 6557
- Pages
- 871-876
- ISSN
- 0036-8075 (ISSN-L)
- PMC
- PMC7612213
- OpenAlex
- W3186179742
- Semantic Scholar
- 51362860fb8ad08255660303550ed4bc345535fc
- Cited by
- 5,549
- References
- 85
- Citations by year
- 2026: 912 · 2025: 1108 · 2024: 1165 · 2023: 1144 · 2022: 982 · 2021: 227
- FWCI
- 247.22
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Science (publishedVersion, cc-by)
- bioRxiv (Cold Spring Harbor Laboratory) (submittedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion, cc-by)
- eScholarship (California Digital Library) (submittedVersion, cc-by)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion, cc-by)
- DESY Publication Database (PUBDB) (Deutsches Elektronen-Synchrotron) (submittedVersion)
- Apollo (University of Cambridge) (submittedVersion, cc-by)
- Apollo (cc-by)
- Funders
- National Science Foundation (GM127390, 1937533)
- Welch Foundation (I‐1505)
- Washington Research Foundation
- Microsoft
- TU Graz, Internationale Beziehungen und Mobilitätsprogramme
- Wellcome Trust (Z/17/Z, 209407/Z/17/Z, 209407)
- G. Harold and Leila Y. Mathers Foundation
- Diamond Light Source
- Open Philanthropy Project
- Global Challenges Research Fund (ST/R002754/1)
- Schmidt Family Foundation
- European Synchrotron Radiation Facility
- Austrian Science Fund (P 29432, DOC 50)
- National Institutes of Health (DP5OD026389, RO1-AI51321)
- Canadian Institutes of Health Research (168998, 168907)
- Science and Technology Facilities Council (ST/R002754/1)
- Deutsches Elektronen-Synchrotron
- National Institute of General Medical Sciences (P01GM063210, R35GM127390, R01GM123089)
- National Institute of Allergy and Infectious Diseases (R01AI051321)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Structural Biology › Enzyme Structure and Function
- Keywords
- protein structure prediction, protein-protein complex prediction, distance maps, sequence-based structure prediction, CASP14, protein function prediction
- MeSH terms
- Deep Learning, Amino Acid Sequence, Computer Simulation, Membrane Proteins, Models, Molecular, Protein Conformation, Proteins, Neural Networks, Computer, Protein Folding, Crystallography, X-Ray, Cryoelectron Microscopy, Protein Subunits, Databases, Protein, Receptors, G-Protein-Coupled, Multiprotein Complexes, Sphingosine N-Acyltransferase, ADAM Proteins
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.