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 301–400 of 5,634
John van der, O. O. S. T., & Patinios, C. (2023) · Trends in biotechnology · doi:10.1016/j.tibtech.2022.12.022 · Cited by 117
A Transformer-Based Ensemble Framework for the Prediction of Protein–Protein Interaction Sites
Mou, M., Pan, Z., Zhou, Z., Zheng, L., Zhang, H., Shi, S., Li, F., Sun, X., & Zhu, F. (2023) · Research · doi:10.34133/research.0240 · Cited by 117
Ali, R. S. A. E., Meng, J., Khan, M. E. I., & Jiang, X. (2024) · Artificial Intelligence Chemistry · doi:10.1016/j.aichem.2024.100049 · Cited by 123
A survey of geometric graph neural networks: data structures, models and applications
Han, J., Cen, J., Wu, L., Li, Z., Kong, X., Jiao, R., Yu, Z., Xu, T., Wu, F., Wang, Z., et al. (2025) · Frontiers of Computer Science · doi:10.1007/s11704-025-41426-w · Cited by 126
“Building upon these advancements, RoseTTAFold2 [10] extends the capabilities of both AlphaFold2 [128] and RoseTTAFold [9] by refining the attention mechanism and enhancing the three-track architecture, resulting in notable performance improvements.”
InfluentialStructure and dynamics in drug discovery
Wei, H., & McCammon, J. A. (2024) · npj Drug Discovery. · doi:10.1038/s44386-024-00001-2 · Cited by 129
AlphaFold Protein Structure Database 2025: a redesigned interface and updated structural coverage
Bertoni, D., Tsenkov, M. I., Magaña, P., Nair, S. S., Pidruchna, I., Afonso, M. Q. L., Midlik, A., Paramval, U., Lawal, D. K., Tanweer, A., et al. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkaf1226 · Cited by 137
“The field of life science has been transformed by protein structure prediction tools, such as AlphaFold [ 1 –3 ], ESMFold [ 4 ], and RoseTTAFold [ 5 ].”
AlphaFold3 takes a step toward decoding molecular behavior and biological computation
Roy, R., & Al‐Hashimi, H. M. (2024) · Nature Structural & Molecular Biology · doi:10.1038/s41594-024-01350-2 · Cited by 101
Effector Identification in Plant Pathogens
Lovelace, A. H., Dorhmi, S., Hulin, M. T., Li, Y., Mansfield, J. W., & Ma, W. (2023) · Phytopathology® · doi:10.1094/phyto-09-22-0337-kd · Cited by 90
Koh, H. Y., Zheng, Y., Yang, M., Arora, R., Webb, G. I., Pan, S., Li, L., & Church, G. M. (2025) · Nature Reviews Bioengineering · doi:10.1038/s44222-025-00349-8 · Cited by 147
Ranking Peptide Binders by Affinity with AlphaFold**
Chang, L., & Perez, A. (2023) · Angewandte Chemie International Edition · doi:10.1002/anie.202213362 · Cited by 100
Genomic language model predicts protein co-regulation and function
Hwang, Y., Cornman, A. L., Kellogg, E. H., Ovchinnikov, S., & Girguis, P. R. (2024) · Nature Communications · doi:10.1038/s41467-024-46947-9 · Cited by 99
Self-citationAlphaFold prediction of structural ensembles of disordered proteins
Brotzakis, Z. F., Zhang, S., Murtada, M. H., & Vendruscolo, M. (2025) · Nature Communications · doi:10.1038/s41467-025-56572-9 · Cited by 99
Retinal pigment epithelium 65 kDa protein (RPE65): An update
Kiser, P. D. (2022) · Progress in Retinal and Eye Research · doi:10.1016/j.preteyeres.2021.101013 · Cited by 91
Current challenges to virus discovery by meta-transcriptomics
Cobbin, J. C. A., Charon, J., Harvey, E., Holmes, E. C., & Mahar, J. E. (2021) · Current Opinion in Virology · doi:10.1016/j.coviro.2021.09.007 · Cited by 113
The science of the host–virus network
Albery, G. F., Becker, D. J., Brierley, L., Brook, C. E., Christofferson, R. C., Cohen, L., Dallas, T., Eskew, E. A., Fagre, A. C., Farrell, M. J., et al. (2021) · Nature Microbiology · doi:10.1038/s41564-021-00999-5 · Cited by 109
“In the near-term future, these datasets may increasingly be supported by machine learning tools that predict protein folding structures 73,74 .”
BackgroundCabral, L., Persinoti, G. F., Paixão, D. A. A., Martins, M. P., Morais, M. A. B., Chinaglia, M., Domingues, M. N., Sforça, M. L., Pirolla, R. A. S., Generoso, W. C., et al. (2022) · Nature Communications · doi:10.1038/s41467-022-28310-y · Cited by 109
“R: During the paper evaluation period, we updated the structural model of the GHXXX using RoseTTAFold, an improved deep learning-based modeling method (Baek et al., 2021).”
MethodGenetic regulatory networks of soybean seed size, oil and protein contents
Duan, Z., Li, Q., Wang, H., He, X., & Zhang, M. (2023) · Frontiers in Plant Science · doi:10.3389/fpls.2023.1160418 · Cited by 99
“Moreover, artificial intelligence (AI) technology or machine learning approach can make mining big data more efficient, for instance, omics data processing, protein structure construction, and pan-omics data integration (Baek et al., 2021; Jumper et al., 2021; Reel et al., 2021).”
BackgroundGenerative artificial intelligence for de novo protein design
Winnifrith, A., Outeiral, C., & Hie, B. (2024) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2024.102794 · Cited by 113
How accurately can one predict drug binding modes using AlphaFold models?
(2023) · doi:10.7554/elife.89386.2 · Cited by 98
“AF2 and other recent neural-network-based protein structure prediction methods use experimentally determined protein structures in two ways: first, to train the neural network and second, as templates for prediction of individual protein structures (Baek et al., 2021; Jumper et al., 2021).”
BackgroundMethodInfluentialGeometric deep learning for drug discovery
Liu, M., Li, C., Chen, R., Cao, D., & Zeng, X. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122498 · Cited by 107
Building Enzymes through Design and Evolution
Hossack, E. J., Hardy, F. J., & Green, A. P. (2023) · ACS Catalysis · doi:10.1021/acscatal.3c02746 · Cited by 111
Protein–RNA interaction prediction with deep learning: structure matters
Wei, J., Chen, S., Zong, L., Gao, X., & Li, Y. (2021) · Briefings in Bioinformatics · doi:10.1093/bib/bbab540 · Cited by 105
Protein folds vs. protein folding: Differing questions, different challenges
Chen, S.-J., Hassan, M., Jernigan, R. L., Jia, K., Kihara, D., Kloczkowski, A., Kotelnikov, S., Kozakov, D., Liang, J., Liwo, A., et al. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2214423119 · Cited by 93
Lostao, A., Lim, K., Pallarés, M. C., Ptak, A., & Marcuello, C. (2023) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2023.124089 · Cited by 101
Progress at protein structure prediction, as seen in CASP15
Elofsson, A. (2023) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2023.102594 · Cited by 92
Recent Advances in Deep Learning for Protein-Protein Interaction Analysis: A Comprehensive Review
Lee, M. (2023) · Molecules · doi:10.3390/molecules28135169 · Cited by 112
“[92] utilized a three-track neural network that integrates information at different dimensional levels for protein structure and interaction prediction, showing nearly comparable performance to DeepMind’s system in the 14th Critical Assessment of Structure Prediction (CASP14) conference.”
MethodAlQuraishi, M., & Sorger, P. K. (2021) · Nature Methods · doi:10.1038/s41592-021-01283-4 · Cited by 96
“33,51,70 For example, the knowledge that protein backbones are covalently bonded polypeptide chains with nearly fixed bond lengths and angles but sequence-dependent torsion angles 71 .”
BackgroundAlphaFold, allosteric, and orthosteric drug discovery: Ways forward
Nussinov, R., Zhang, M., Liu, Y., & Jang, H. (2023) · Drug Discovery Today · doi:10.1016/j.drudis.2023.103551 · Cited by 96
PLMSearch: Protein language model powers accurate and fast sequence search for remote homology
Liu, W., Wang, Z., You, R., Xie, C., Wei, H., Xiong, Y., Yang, J., & Zhu, S. (2024) · Nature Communications · doi:10.1038/s41467-024-46808-5 · Cited by 96
Computational methods to predict protein aggregation
Navarro, S., & Ventura, S. (2022) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2022.102343 · Cited by 95
Targeting small GTPases: emerging grasps on previously untamable targets, pioneered by KRAS
Yin, G., Huang, J., Petela, J., Jiang, H., Zhang, Y., Gong, S., Wu, J., Liu, B., Shi, J., & Gao, Y. (2023) · Signal Transduction and Targeted Therapy · doi:10.1038/s41392-023-01441-4 · Cited by 82
“This issue might be improved with the recent developments of cryoEM for structural determinations and deep learning methods for structural predictions such as AlphaFold Multimer(529) and RoseTTA Fold.(530) Understanding the structural plasticity of switch regions in small GTPases is a unique resource for developing in…”
BackgroundUnlocking the potential of enzyme engineering via rational computational design strategies
Zhou, L., Tao, C., Shen, X., Sun, X., Wang, J., & Yuan, Q. (2024) · Biotechnology Advances · doi:10.1016/j.biotechadv.2024.108376 · Cited by 102
Self-citation(2023) · doi:10.1021/acs.jcim.2c01270 · Cited by 95
Structure of the nutrient-sensing hub GATOR2
Valenstein, M. L., Rogala, K. B., Lalgudi, P. V., Brignole, E. J., Gu, X., Saxton, R. A., Chantranupong, L., Kolibius, J., Quast, J.-P., & Sabatini, D. M. (2022) · Nature · doi:10.1038/s41586-022-04939-z · Cited by 86
Advancing Ligand Docking through Deep Learning: Challenges and Prospects in Virtual Screening
Zhang, X., Shen, C., Zhang, H., Kang, Y., Hsieh, C., & Hou, T. (2024) · Accounts of Chemical Research · doi:10.1021/acs.accounts.4c00093 · Cited by 94
Structure-based protein design with deep learning
Ovchinnikov, S., & Huang, P.-S. (2021) · Current Opinion in Chemical Biology · doi:10.1016/j.cbpa.2021.08.004 · Cited by 93
Self-citationThe European Bioinformatics Institute (EMBL-EBI) in 2021
Cantelli, G., Bateman, A., Brooksbank, C., Petrov, A. I., Malik-Sheriff, R. S., Ide-Smith, M., Hermjakob, H., Flicek, P., Apweiler, R., Birney, E., et al. (2022) · Nucleic Acids Research · doi:10.1093/nar/gkab1127 · Cited by 96
Protein‐protein interaction networks as miners of biological discovery
Wang, S., Wu, R., Lü, J., Jiang, Y., Huang, T., & Cai, Y. (2022) · PROTEOMICS · doi:10.1002/pmic.202100190 · Cited by 100
“The newest version of RoseTTAFold employed a three-track neu-ralnetworkthatintegrated1Dsequenceinformation,2Ddistancemap information, and 3D coordinate information [49].”
BackgroundMethodPasrija, P., Jha, P. K., Upadhyaya, P., Khan, M. S., & Chopra, M. (2022) · Current Topics in Medicinal Chemistry · doi:10.2174/1568026622666220701091339 · Cited by 115
NanoNet: Rapid and accurate end-to-end nanobody modeling by deep learning
Cohen, T., Halfon, M., & Schneidman-Duhovny, D. (2022) · Frontiers in Immunology · doi:10.3389/fimmu.2022.958584 · Cited by 88
“Most recent structure prediction methods, including AlphaFold2 and RosettaFold, use deep learning models for end-to-end learning, where the input is a sequence and the output is the 3D structure (15, 16, 20, 21).”
MethodHierarchical graph transformer with contrastive learning for protein function prediction
Gu, Z., Luo, X., Chen, J., Deng, M., & Lai, L. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad410 · Cited by 110
Rules and mechanisms governing G protein coupling selectivity of GPCRs
Masuho, I., Kise, R., Gainza, P., Von Moo, E., Li, X., Tany, R., Wakasugi-Masuho, H., Correia, B. E., & Martemyanov, K. A. (2023) · Cell Reports · doi:10.1016/j.celrep.2023.113173 · Cited by 87
Automated in vivo enzyme engineering accelerates biocatalyst optimization
Orsi, E., von Borzyskowski, L. S., Noack, S., Nikel, P. I., & Lindner, S. N. (2024) · Nature Communications · doi:10.1038/s41467-024-46574-4 · Cited by 97
“The fi ne-tuning of the deep learning neural network RoseTTAFold 54 on protein structure denoising tasks resulted in RFdiffusion, a generative model of protein backbones with outstanding performance on protein monomer design, enzymeactive sitescaffolding(Fig.”
BackgroundSeong, K., & Krasileva, K. V. (2021) · Molecular Plant-Microbe Interactions® · doi:10.1094/mpmi-03-21-0071-r · Cited by 82
Machine learning: its challenges and opportunities in plant system biology
Hesami, M., Alizadeh, M., Jones, A. M. P., & Torkamaneh, D. (2022) · Applied Microbiology and Biotechnology · doi:10.1007/s00253-022-11963-6 · Cited by 103
“Moreover, the vast majority of protein functions are performed by interactions among different proteins that can be predicted through ML models (Senior et al. 2020; Wu et al. 2020; Baek et al. 2021).”
BackgroundInfluentialQiu, Y., & Wei, G.-W. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad289 · Cited by 88
“Two prominent methods are AlphaFold2 [24] and RosettaFold [144], which have demonstrated remarkable capabilities in predicting protein structures with atomic-level accuracy.”
An AI system to help scientists write expert-level empirical software
Aygün, E., Belyaeva, A., Comanici, G., Coram, M., Cui, H., Garrison, J., Kast, R. J. A., Kast, A., McLean, C. Y., Norgaard, P., et al. (2026) · Nature · doi:10.1038/s41586-026-10658-6 · Cited by 144
Shen, D.-Y., Pan, F., Yang, Z., Song, H., Zou, T., Xiong, J., Li, K., Li, P. S., Hu, N., & Xue, D. (2022) · Food Chemistry · doi:10.1016/j.foodchem.2022.132993 · Cited by 92
Kulmanov, M., & Hoehndorf, R. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac256 · Cited by 89
Ni, B., Kaplan, D. L., & Buehler, M. J. (2023) · Chem · doi:10.1016/j.chempr.2023.03.020 · Cited by 88
Design of stimulus-responsive two-state hinge proteins
Praetorius, F., Leung, P. J. Y., Tessmer, M. H., Broerman, A., Demakis, C., Dishman, A. F., Pillai, A., Idris, A., Juergens, D., Dauparas, J., et al. (2023) · Science · doi:10.1126/science.adg7731 · Cited by 90
“We used inpainting(33) with RoseTTAFold(34) to rigidly connect one end of hinge cs221 to a previously validated homotrimer(35, 36) and the other end of the hinge to a previously validated monomeric protein(37).”
MethodSelf-citationAlphaFold two years on: Validation and impact
Kovalevskiy, O., Mateos-Garcia, J., & Tunyasuvunakool, K. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2315002121 · Cited by 75
“Molecular replacement, a technique to solve phase problem in macromolecular crystallography, requires a search model that closely resembles the actual contents of the target crystal.”
Deep generative design of RNA aptamers using structural predictions
Wong, F. J., He, D., Krishnan, A., Hong, L., Wang, A. Z., Wang, J., Hu, Z., Omori, S., Li, A., Rao, J., et al. (2024) · Nature Computational Science · doi:10.1038/s43588-024-00720-6 · Cited by 110
“Deep learning methods for the accurate prediction of protein 3D structure—including AlphaFold and RoseTTAFold—are now available 14,15 .”
MethodSelf-citationArtificial intelligence challenges for predicting the impact of mutations on protein stability
Pucci, F., Schwersensky, M., & Rooman, M. (2022) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2021.11.001 · Cited by 85
The case for data science in experimental chemistry: examples and recommendations
Yano, J., Gaffney, K. J., Gregoire, J. M., Hung, L., Ourmazd, A., Schrier, J., Sethian, J. A., & Toma, F. M. (2022) · Nature Reviews Chemistry · doi:10.1038/s41570-022-00382-w · Cited by 98
“The accumulated data within the PDB has enabled the development of the recent AlphaFold 89 and RoseTTAFold 90 models for predicting the 3D structure of proteins solely on the basis of their amino acid sequences.”
MethodG-Quadruplex-Binding Proteins: Promising Targets for Drug Design
Shu, H., Zhang, R., Xiao, K., Yang, J., & Sun, X. S. (2022) · Biomolecules · doi:10.3390/biom12050648 · Cited by 99
Targeting SLC transporters: small molecules as modulators and therapeutic opportunities
Schlessinger, A., Zatorski, N., Hutchinson, K., & Colas, C. (2023) · Trends in Biochemical Sciences · doi:10.1016/j.tibs.2023.05.011 · Cited by 94
Metal3D: a general deep learning framework for accurate metal ion location prediction in proteins
Dürr, S. L., Levy, A., & Rothlisberger, U. (2023) · Nature Communications · doi:10.1038/s41467-023-37870-6 · Cited by 88
“…based deep learning approaches have been used in the fi eld of protein research for a variety of applications such as protein structure prediction 27,28 , prediction of identity of masked residues 29 – 31 , functional site prediction 32,33 , for ranking of docking poses 34,35 , prediction of the…”
Ma, B., Fan, Y., Zhang, D., Wei, Y., Jian, Y., Liu, D., Wang, Z., Gao, Y., Ma, J., Chen, Y., et al. (2022) · Advanced Science · doi:10.1002/advs.202201859 · Cited by 96
Chemistry Nobel goes to developers of AlphaFold AI that predicts protein structures
Callaway, E. (2024) · Nature · doi:10.1038/d41586-024-03214-7 · Cited by 88
AlphaFold2 Can Predict Single-Mutation Effects
McBride, J. M., Polev, K., Abdirasulov, A., Reinharz, V., Grzybowski, B. A., & Tlusty, T. (2023) · Physical Review Letters · doi:10.1103/physrevlett.131.218401 · Cited by 87
Predicting protein-protein interactions in the human proteome
Zhang, J., Humphreys, I. R., Pei, J., Kim, J., Choi, C., Yuan, R., Durham, J., Liu, S., Choi, H.-J., Baek, M., et al. (2025) · Science · doi:10.1126/science.adt1630 · Cited by 87
“Coevolution-ary analysis (residues at PPI interfaces coevolve to preserve their interactions) has been combined with 3D structure prediction methods, such as AlphaFold (AF) ( 11 ) and Ro-seTTAFold (RF) ( 12 ), to identify PPIs on a proteome-wide scale in bacteria and yeast ( 13 , 14 ).”
Self-citationHigh-Resolution Single-Molecule Magnetic Tweezers
Choi, H.-K., Kim, H. G., Shon, M. J., & Yoon, T. (2022) · Annual Review of Biochemistry · doi:10.1146/annurev-biochem-032620-104637 · Cited by 97
Restoring protein glycosylation with GlycoShape
Ives, C. M., Singh, O., D’Andrea, S., Fogarty, C. A., Harbison, A. M., Satheesan, A., Tropea, B., & Fadda, E. (2024) · Nature Methods · doi:10.1038/s41592-024-02464-7 · Cited by 101
“Ultimately, the results in this work confirm an argument some of us presented earlier 59 as well as recent work from the developers of RoseTTAFold 82,83 , that protein 3D structures from ML can be readily glycosylated and also used to predict glycosylation site occupancy, where this depends on 3D…”
ResultTowards a Definition of Generative Artificial Intelligence
Ronge, R., Maier, M., & Rathgeber, B. (2025) · Philosophy & Technology · doi:10.1007/s13347-025-00863-y · Cited by 101
“While AlphaFold is not the only such approach to the prediction of tertiary structure of proteins (see, e.g., Baek et al., 2021; Callaway, 2022), it is of particular interest for our purposes because of its latest two versions: While AlphaFold 2 predicts the threedimensional structure of proteins…”
Arnold, A., McLellan, S., & Stokes, J. M. (2025) · npj Antimicrobials and Resistance · doi:10.1038/s44259-025-00085-4 · Cited by 108
AlphaFold2 can predict single-mutation effects
McBride, J. M., Polev, K., Abdirasulov, A., Reinharz, V., Grzybowski, B. A., & Tlusty, T. (2022) · doi:10.1101/2022.04.14.488301 · Cited by 87
BFVD—a large repository of predicted viral protein structures
Kim, R. S., Levy Karin, E., Mirdita, M., Chikhi, R., & Steinegger, M. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkae1119 · Cited by 86
Democratizing protein language models with parameter-efficient fine-tuning
Sledzieski, S., Kshirsagar, M., Baek, M., Dodhia, R., Lavista Ferres, J., & Berger, B. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2405840121 · Cited by 86
Self-citationEnhancing alphafold-multimer-based protein complex structure prediction with MULTICOM in CASP15
Liu, J., Guo, Z., Wu, T., Roy, R. S., Quadir, F., Chen, C., & Cheng, J. (2023) · Communications Biology · doi:10.1038/s42003-023-05525-3 · Cited by 86
“Based on UniProt accession numbers, sequence alignments from the subunit MSAs are concatenated if the difference between their UniProt accession numbers is smaller than 10 as in RosettaFold 37 .”
Integration of Mass Spectrometry Data for Structural Biology
Britt, H. M., Cragnolini, T., & Thalassinos, K. (2021) · Chemical Reviews · doi:10.1021/acs.chemrev.1c00356 · Cited by 89
Surfing the wave of oxyfunctionalization chemistry by engineering fungal unspecific peroxygenases
Beltrán-Nogal, A., Sánchez‐Moreno, I., Méndez‐Sánchez, D., de Santos, P. G., Hollmann, F., & Alcalde, M. (2022) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2022.102342 · Cited by 86
Wang, Z., Combs, S. A., Brand, R., Calvo, M. R., Xu, P., Price, G., Golovach, N., Salawu, E. O., Wise, C. J., Ponnapalli, S. P., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-10775-y · Cited by 85
“Recent breakthroughs in deep learning approaches including AlphaFold 2,3 , trRosetta 4 and a three-track deep neural net approach by Baek et al. 5 , have exceeded performance based on traditional approaches for modeling the folding processes.”
MethodGenome interpretation using in silico predictors of variant impact
Katsonis, P., Wilhelm, K. O., Williams, A. M., & Lichtarge, O. (2022) · Human Genetics · doi:10.1007/s00439-022-02457-6 · Cited by 100
“Alternatively, stability prediction methods may take advantage of the recent development of high-quality protein structure predictors, such as AlphaFold (Jumper et al. 2021) and RosettaFold (Baek et al. 2021), which broaden the available number of protein structures.”
BackgroundLindorff-Larsen, K., & Kragelund, B. B. (2021) · Journal of Molecular Biology · doi:10.1016/j.jmb.2021.167196 · Cited by 85
Bittrich, S., Bhikadiya, C., Bi, C., Chao, H., Duarte, J. M., Dutta, S., Fayazi, M., Henry, J., Khokhriakov, I., Lowe, R., et al. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.167994 · Cited by 87
“17 RoseTTAFold, 18 developed subsequently by David Baker and co-workers, supports prediction of both individual protein structures and those of binary complexes, exemplified with 1,106 heterodimers of Saccharomyces cerevisiae proteins.”
BackgroundDeep language models for interpretative and predictive materials science
Hu, Y., & Buehler, M. J. (2023) · APL Machine Learning · doi:10.1063/5.0134317 · Cited by 78
Shor, B., & Schneidman-Duhovny, D. (2024) · Nature Methods · doi:10.1038/s41592-024-02174-0 · Cited by 85
“7) and RosettaFold 8 .”
InfluentialRoom-temperature crystallography reveals altered binding of small-molecule fragments to PTP1B
Skaist Mehlman, T., Biel, J. T., Azeem, S. M., Nelson, E. R., Hossain, S., Dunnett, L., Paterson, N. G., Douangamath, A., Talon, R., Axford, D., et al. (2023) · eLife · doi:10.7554/elife.84632 · Cited by 84
Rui, H., Ashton, K. S., Min, J., Wang, C. W., & Potts, P. R. (2023) · RSC Chemical Biology · doi:10.1039/d2cb00207h · Cited by 94
Diffusion models in protein structure and docking
Yim, J., Stärk, H., Corso, G., Jing, B., Barzilay, R., & Jaakkola, T. (2024) · Wiley Interdisciplinary Reviews Computational Molecular Science · doi:10.1002/wcms.1711 · Cited by 108
“A fundamental limitation of AF2 and related methods 5,6 is their inability to generate distributions of structures conditioned on biophysical properties.”
BackgroundRoom-temperature crystallography reveals altered binding of small-molecule fragments to PTP1B
Mehlman, T. (., Biel, J. T., Azeem, S. M., Nelson, E. R., Hossain, S., Dunnett, L. E., Paterson, N. G., Douangamath, A., Talon, R., Axford, D., et al. (2022) · doi:10.1101/2022.11.02.514751 · Cited by 84
Machine learning solutions for predicting protein–protein interactions
Casadio, R., Martelli, P. L., & Savojardo, C. (2022) · WIREs Computational Molecular Science · doi:10.1002/wcms.1618 · Cited by 80
“115 Furthermore, other papers still in bioRxiv, support the notion that the problem of predicting the interfaces of protein complexes, can take advantage of the deep learning-based methods included in AlphaFold2.”
BackgroundImpact of AlphaFold on structure prediction of protein complexes: The CASP15‐CAPRI experiment
Lensink, M. F., Brysbaert, G., Raouraoua, N., Bates, P. A., Giulini, M., Honorato, R. V., van Noort, C., Teixeira, J. M. C., Bonvin, A. M. J. J., Kong, R., et al. (2023) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.26609 · Cited by 77
“More specifically, the extent to which the power of AlphaFold and other DL-based prediction methods, such as RoseTTAFold, 35 will be harnessed by the community to produce a major leap in performance over more classical modeling approaches in the context of blind predictions.”
Revolutionizing CRISPR technology with artificial intelligence
Kim, M., Go, M.-J., Kang, S.-H., Jeong, S.-H., & Lim, K. (2025) · Experimental & Molecular Medicine · doi:10.1038/s12276-025-01462-9 · Cited by 101
“Around the same time, the Baker laboratory introduced RoseTTA-Fold 30 , another DL-based protein structure prediction model.”
Current structure predictors are not learning the physics of protein folding
Outeiral, C., Nissley, D. A., & Deane, C. M. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btab881 · Cited by 82
Nanobodies: Robust miniprotein binders in biomedicine
Kim, J. Y. J., Sang, Z., Xiang, Y., Shen, Z., & Shi, Y. (2023) · Advanced Drug Delivery Reviews · doi:10.1016/j.addr.2023.114726 · Cited by 87
Recent advances in predicting and modeling protein–protein interactions
Durham, J., Zhang, J., Humphreys, I. R., Pei, J., & Cong, Q. (2023) · Trends in Biochemical Sciences · doi:10.1016/j.tibs.2023.03.003 · Cited by 89
Self-citationIntroduction to latent variable energy-based models: a path toward autonomous machine intelligence
Dawid, A., & LeCun, Y. (2024) · Journal of Statistical Mechanics Theory and Experiment · doi:10.1088/1742-5468/ad292b · Cited by 128
“In biology, the recent models such as AlphaFold [ 3 ] and RoseTTA Fold [ 25 ] can accurately predict the three-dimensional structure of proteins from their amino acid sequence, an extremely important and challenging problem whose solution opens new pathways to drug design.”
MethodchiLife: An open-source Python package for in silico spin labeling and integrative protein modeling
Tessmer, M. H., & Stoll, S. (2023) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1010834 · Cited by 74
PDBe-KB: collaboratively defining the biological context of structural data
PDBe-KB consortium, Varadi, M., Anyango, S., Armstrong, D., Berrisford, J., Choudhary, P., Deshpande, M., Nadzirin, N., Nair, S. S., Pravda, L., et al. (2022) · Nucleic Acids Research · doi:10.1093/nar/gkab988 · Cited by 82
Quantifying the use and potential benefits of artificial intelligence in scientific research
Gao, J., & Wang, D. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-02020-5 · Cited by 81
“…AI’s recent remarkable success in advancing research frontiers across several fields 31–44 , from predicting the structure of proteins in biology 45–47 to designing new drug candidates in medicine 48–51 , from discovering natural laws in physics 52–54 to solving complicated equations and…”
Protein–Ligand Blind Docking Using CB-Dock2
Liu, Y., & Cao, Y. (2023) · Methods in molecular biology · doi:10.1007/978-1-0716-3441-7_6 · Cited by 95
Strengthening nucleic acid biosecurity screening against generative protein design tools
Wittmann, B. J., Alexanian, T., Bartling, C. M., Beal, J., Clore, A. J., Diggans, J., Flyangolts, K., Gemler, B. T., Mitchell, T., Murphy, S. T., et al. (2025) · Science · doi:10.1126/science.adu8578 · Cited by 100
Language Modeling Materializes a World Model of Protein BiologyPreprint
Candido, S., Hayes, T., Derry, A., Rao, R., Lin, Z., Verkuil, R., Wu, B. Z., Lee, J. S., Bruguera, E. S., Keval, J. A., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.06.03.729735 · Cited by 151
Sala, D., Hildebrand, P. W., & Meiler, J. (2023) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2023.1121962 · Cited by 80
“Among them, RoseTTAFold was the first approach that was able to predict both active and inactive GPCR conformations by using templates in a uniform functional state, outperforming comparative homology modeling methods (Baek et al., 2021).”
BackgroundEvolution and Functions of Plant U-Box Proteins: From Protein Quality Control to Signaling
Trenner, J., Monaghan, J., Saeed, B., Quint, M., Shabek, N., & Trujillo, M. (2022) · Annual Review of Plant Biology · doi:10.1146/annurev-arplant-102720-012310 · Cited by 88
Toxicology and pharmacology of botulinum and tetanus neurotoxins: an update
Pirazzini, M., Montecucco, C., & Rossetto, O. (2022) · Archives of Toxicology · doi:10.1007/s00204-022-03271-9 · Cited by 90
The evolution of post-translational modifications
Bradley, D. T. C. (2022) · Current Opinion in Genetics & Development · doi:10.1016/j.gde.2022.101956 · Cited by 88
Updated protein domain annotation of the PARP protein family sheds new light on biological function
Suskiewicz, M. J., Munnur, D., Strømland, Ø., Yang, J.-C., Easton, L. E., Chatrin, C., Zhu, K., Baretić, D., Goffinont, S., Schuller, M., et al. (2023) · Nucleic Acids Research · doi:10.1093/nar/gkad514 · Cited by 88
“Recentl y, experimental a ppr oaches to studying pr otein structure have been supplemented with highly efficient artificial intelligence (AI) prediction systems including AlphaFold –– especially its AF2 release ( 40 ) –– and its analogues or deri vati v es, including RoseTTAFold ( 41 ).”
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.