DeepProSite: structure-aware protein binding site prediction using ESMFold and pretrained language model
Open accessThe DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.
Abstract
Motivation: Identifying the functional sites of a protein, such as the binding sites of proteins, peptides, or other biological components, is crucial for understanding related biological processes and drug design. However, existing sequence-based methods have limited predictive accuracy, as they only consider sequence-adjacent contextual features and lack structural information. Results: In this study, DeepProSite is presented as a new framework for identifying protein binding site that utilizes protein structure and sequence information. DeepProSite first generates protein structures from ESMFold and sequence representations from pretrained language models. It then uses Graph Transformer and formulates binding site predictions as graph node classifications. In predicting protein–protein/peptide binding sites, DeepProSite outperforms state-of-the-art sequence- and structure-based methods on most metrics. Moreover, DeepProSite maintains its performance when predicting unbound structures, in contrast to competing structure-based prediction methods. DeepProSite is also extended to the prediction of binding sites for nucleic acids and other ligands, verifying its generalization capability. Finally, an online server for predicting multiple types of residue is established as the implementation of the proposed DeepProSite. Availability and implementation: The datasets and source codes can be accessed at https://github.com/WeiLab-Biology/DeepProSite. The proposed DeepProSite can be accessed at https://inner.wei-group.net/DeepProSite/.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Highly accurate protein structure prediction with AlphaFold
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03819-2 · Cited by 43,764
Evolutionary-scale prediction of atomic level protein structure with a language model
Lin, Z., Akin, H., Rao, R., Hie, B., Zhu, Z., Lu, W., Smetanin, N., Verkuil, R., Kabeli, O., Shmueli, Y., et al. (2022) · doi:10.1101/2022.07.20.500902 · Cited by 5,990
Accurate prediction of protein structures and interactions using a 3-track network
Baek, M., DiMaio, F., Anishchenko, I., Dauparas, J., Ovchinnikov, S., Lee, G. R., Wang, J., Cong, Q., Kinch, L. N., Schaeffer, R. D., et al. (2021) · doi:10.1101/2021.06.14.448402 · Cited by 5,549
Intrinsically unstructured proteins and their functions
Dyson, H. J., & Wright, P. E. (2005) · Nature Reviews Molecular Cell Biology · doi:10.1038/nrm1589 · Cited by 3,871 · cited 1×
“In addition, several technical challenges, including the small size of peptides (Vlieghe et al. 2010), weak binding affinity (Dyson and Wright 2005), conformational flexibility (Bertolazzi et al. 2014), high tran-sience, and dynamics of protein–protein interactions, increase the difficulty in…”
BackgroundProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning
Elnaggar, A., Heinzinger, M., Dallago, C., Rehawi, G., Wang, Y., Jones, L., Gibbs, T., Feher, T., Angerer, C., Steinegger, M., et al. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3095381 · Cited by 2,329
UniRef clusters: a comprehensive and scalable alternative for improving sequence similarity searches
Suzek, B. E., Wang, Y., Huang, H., McGarvey, P. B., Wu, C. H., & the UniProt Consortium (2015) · Bioinformatics · doi:10.1093/bioinformatics/btu739 · Cited by 1,928
Therapeutic peptides: Historical perspectives, current development trends, and future directions
Lau, J. L., & Dunn, M. K. (2018) · Bioorganic & Medicinal Chemistry · doi:10.1016/j.bmc.2017.06.052 · Cited by 1,879 · cited 1×
“Moreover, therapeutic peptides have become a focus of drug development because they can bind to proteins and have therapeutic effects (Fosgerau and Hoffmann 2015, Bruzzoni-Giovanelli et al. 2018, Lau and Dunn 2018, Davenport et al. 2020).”
BackgroundLGA: a method for finding 3D similarities in protein structures
Zemla, A. (2003) · Nucleic Acids Research · doi:10.1093/nar/gkg571 · Cited by 1,069 · cited 4×
“According to predictions made by DeepProSite, the top 30% of proteins with the highest GDT (average GDT ¼ 0.964) have an average AUPRC of 0.627, which is significantly higher than that of the bottom 30% of proteins with the lowest GDT (average GDT ¼ 0.526), which showed an average AUPRC of 0.375 according to Mann–Whit…”
MethodInfluentialStructure-based prediction of protein–protein interactions on a genome-wide scale
Zhang, Q. C., Petrey, D., Deng, L., Qiang, L., Shi, Y., Thu, C. A., Bisikirska, B., Lefebvre, C., Accili, D., Hunter, T., et al. (2012) · Nature · doi:10.1038/nature11503 · Cited by 723 · cited 1×
“…molecules form the basis of protein function in most biological processes, such as gene expression regulation, signal transduction, and metabolic pathway regulation (Pawson and Nash 2003, Kortemme et al. 2004, Hermann et al. 2007, Rubinstein and Niv 2009, Babu et al. 2011, Zhang et al. 2012).”
BackgroundSmall-molecule MDM2 antagonists reveal aberrant p53 signaling in cancer: implications for therapy
Tovar, C., Rosinski, J., Filipovic, Z., Higgins, B., Kolinsky, K., Hilton, H., Zhao, X., Vu, B. T., Qing, W., Packman, K., et al. (2006) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0507493103 · Cited by 708
Research works printed but not linked: 64
- Abdin, 2020, Sequence and structure based deep learning models for the identification of peptide binding sites, Adv Neural Inf Process Syst, 33
- Babu, 2011, Intrinsically disordered proteins: regulation and disease, Curr Opin Struct Biol, 21, 432, 10.1016/j.sbi.2011.03.011 · DOI · Paper page
- Batra, 2018, Protein interaction mapping identifies RBBP6 as a negative regulator of Ebola virus replication, Cell, 175, 1917, 10.1016/j.cell.2018.08.044 · DOI · Paper page
- Berman, 2000, The protein data bank, Nucleic Acids Res, 28, 235, 10.1093/nar/28.1.235 · DOI · Paper page
- Bertolazzi, 2014, Predicting protein-ligand and protein-peptide interfaces, Eur Phys J Plus, 129, 1, 10.1140/epjp/i2014-14132-1 · DOI
- Brettner, 2012, Protein stickiness, rather than number of functional protein-protein interactions, predicts expression noise and plasticity in yeast, BMC Syst Biol, 6, 1, 10.1186/1752-0509-6-128 · DOI
- Bruzzoni-Giovanelli, 2018, Interfering peptides targeting protein-protein interactions: the next generation of drugs?, Drug Discov Today, 23, 272, 10.1016/j.drudis.2017.10.016 · DOI
- Chen, 2017, Understand protein functions by comparing the similarity of local structural environments, Biochim Biophys Acta Proteins Proteom, 1865, 142, 10.1016/j.bbapap.2016.11.008 · DOI
- Chen, 2021, Structure-aware protein solubility prediction from sequence through graph convolutional network and predicted contact map, J Cheminform, 13, 7, 10.1186/s13321-021-00488-1 · DOI · Paper page
- Clare, 2004, Computational studies of protein–peptide interactions with systematic mutation of residues, Mol Phys, 102, 939, 10.1080/00268970410001704407 · doi:10.1080/00268970410001704407
- Davenport, 2020, Advances in therapeutic peptides targeting G protein-coupled receptors, Nat Rev Drug Discov, 19, 389, 10.1038/s41573-020-0062-z · DOI · Paper page
- De Las Rivas, 2012, Protein–protein interaction networks: unraveling the wiring of molecular machines within the cell, Brief Funct Genomics, 11, 489, 10.1093/bfgp/els036 · DOI
- Dhole, 2014, Sequence-based prediction of protein–protein interaction sites with L1-logreg classifier, J Theor Biol, 348, 47, 10.1016/j.jtbi.2014.01.028 · DOI
- Fosgerau, 2015, Peptide therapeutics: current status and future directions, Drug Discov Today, 20, 122, 10.1016/j.drudis.2014.10.003 · DOI · Paper page
- Gainza, 2020, Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning, Nat Methods, 17, 184, 10.1038/s41592-019-0666-6 · DOI · Paper page
- Guilliam, 2017, Molecular basis for PrimPol recruitment to replication forks by RPA, Nat Commun, 8, 15222, 10.1038/ncomms15222 · DOI
- Hermann, 2007, Structure-based activity prediction for an enzyme of unknown function, Nature, 448, 775, 10.1038/nature05981 · DOI · Paper page
- Hu, 2018, ATPbind: accurate protein–ATP binding site prediction by combining sequence-profiling and structure-based comparisons, J Chem Inf Model, 58, 501, 10.1021/acs.jcim.7b00397 · DOI · Paper page
- Hu, 2016, Recognizing metal and acid radical ion-binding sites by integrating ab initio modeling with template-based transferals, Bioinformatics, 32, 3260, 10.1093/bioinformatics/btw396 · DOI · Paper page
- Huynh, 2009, Metrics for 3D rotations: comparison and analysis, J Math Imaging Vis, 35, 155, 10.1007/s10851-009-0161-2 · DOI · Paper page
- Hwang, 2008, Protein–protein docking benchmark version 3.0, Prot Structure Funct Bioinform, 73, 705, 10.1002/prot.22106 · DOI · Paper page
- Ingraham, 2019, Generative models for graph-based protein design, Adv Neural Inf Process Syst, 32
- Kabsch, 1983, Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features, Biopolymers, 22, 2577, 10.1002/bip.360221211 · DOI · Paper page
- Kortemme, 2004, Computational redesign of protein-protein interaction specificity, Nat Struct Mol Biol, 11, 371, 10.1038/nsmb749 · DOI · Paper page
- Krapp, 2023, PeSTo: parameter-free geometric deep learning for accurate prediction of protein binding interfaces, Nat Commun, 14, 2175, 10.1038/s41467-023-37701-8 · DOI · Paper page
- Lavi, 2013, Detection of peptide-binding sites on protein surfaces: the first step toward the modeling and targeting of peptide-mediated interactions, Proteins, 81, 2096, 10.1002/prot.24422 · DOI · Paper page
- Lee, 2015, GalaxyPepDock: a protein–peptide docking tool based on interaction similarity and energy optimization, Nucleic Acids Res, 43, W431, 10.1093/nar/gkv495 · DOI · Paper page
- Li, 2021, DELPHI: accurate deep ensemble model for protein interaction sites prediction, Bioinformatics, 37, 896, 10.1093/bioinformatics/btaa750 · DOI · Paper page
- London, 2012, Modeling peptide-protein interactions, Methods Mol Biol, 857, 375, 10.1007/978-1-61779-588-6_17 · DOI · Paper page
- Murakami, 2010, Applying the naïve Bayes classifier with kernel density estimation to the prediction of protein–protein interaction sites, Bioinformatics, 26, 1841, 10.1093/bioinformatics/btq302 · DOI · Paper page
- Orengo, 1997, CATH—a hierarchic classification of protein domain structures, Structure, 5, 1093, 10.1016/S0969-2126(97)00260-8 · DOI · Paper page
- Pawson, 2003, Assembly of cell regulatory systems through protein interaction domains, Science, 300, 445, 10.1126/science.1083653 · DOI · Paper page
- Penna, 2007, Spontaneous and prostatic steroid binding protein peptide-induced autoimmune prostatitis in the nonobese diabetic mouse, J Immunol, 179, 1559, 10.4049/jimmunol.179.3.1559 · DOI
- Petsalaki, 2009, Accurate prediction of peptide binding sites on protein surfaces, PLoS Comput Biol, 5, e1000335, 10.1371/journal.pcbi.1000335 · DOI · Paper page
- Porollo, 2007, Prediction-based fingerprints of protein–protein interactions, Prot Struct Fun Bioinform, 66, 630, 10.1002/prot.21248 · DOI · Paper page
- Qiu, 2020, ProNA2020 predicts protein–DNA, protein–RNA, and protein–protein binding proteins and residues from sequence, J Mol Biol, 432, 2428, 10.1016/j.jmb.2020.02.026 · DOI · Paper page
- Rigden, 2009, From Protein Structure to Function with Bioinformatics, 10.1007/978-1-4020-9058-5 · doi:10.1007/978-1-4020-9058-5
- Rubinstein, 2009, Peptidic modulators of protein-protein interactions: progress and challenges in computational design, Biopolymers, 91, 505, 10.1002/bip.21164 · DOI · Paper page
- Shoemaker, 2007, Deciphering protein–protein interactions. Part I. Experimental techniques and databases, PLoS Comput Biol, 3, e42, 10.1371/journal.pcbi.0030042 · DOI · Paper page
- Taherzadeh, 2016, Sequence-based prediction of protein-peptide binding sites using support vector machine, J Comput Chem, 37, 1223, 10.1002/jcc.24314 · DOI · Paper page
- Taherzadeh, 2018, Structure-based prediction of protein-peptide binding regions using random Forest, Bioinformatics, 34, 477, 10.1093/bioinformatics/btx614 · DOI
- Terentiev, 2009, Dynamic proteomics in modeling of the living cell. Protein-protein interactions, Biochemistry (Mosc), 74, 1586, 10.1134/S0006297909130112 · DOI
- Valkov, 2016, Structure of the Dcp2-Dcp1 mRNA-decapping complex in the activated conformation, Nat Struct Mol Biol, 23, 574, 10.1038/nsmb.3232 · DOI
- Vaswani, 2017, Attention is all you need, Adv Neural Inf Process Syst, 30, 5998
- Vlieghe, 2010, Synthetic therapeutic peptides: science and market, Drug Discov Today, 15, 40, 10.1016/j.drudis.2009.10.009 · DOI · Paper page
- Wang, 2022, Predicting protein-peptide binding residues via interpretable deep learning, Bioinformatics, 38, 3351, 10.1093/bioinformatics/btac352 · DOI · Paper page
- Wang, 2022, RGN: residue-Based graph attention and convolutional network for protein–protein interaction site prediction, J Chem Inform Model, 62, 5961, 10.1021/acs.jcim.2c01092 · DOI · Paper page
- Wardah, 2020, Predicting protein-peptide binding sites with a deep convolutional neural network, J Theor Biol, 496, 110278, 10.1016/j.jtbi.2020.110278 · DOI
- Wells, 2007, Reaching for high-hanging fruit in drug discovery at protein–protein interfaces, Nature, 450, 1001, 10.1038/nature06526 · DOI · Paper page
- Wodak, 2013, Protein–protein interaction networks: the puzzling riches, Curr Opin Struct Biol, 23, 941, 10.1016/j.sbi.2013.08.002 · DOI
- Xia, 2020, Protein–ligand binding residue prediction enhancement through hybrid deep heterogeneous learning of sequence and structure data, Bioinformatics, 36, 3018, 10.1093/bioinformatics/btaa110 · DOI · Paper page
- Xia, 2021, GraphBind: protein structural context embedded rules learned by hierarchical graph neural networks for recognizing nucleic-acid-binding residues, Nucleic Acids Res, 49, e51, 10.1093/nar/gkab044 · DOI · Paper page
- Yang, 2013, BioLiP: a semi-manually curated database for biologically relevant ligand-protein interactions, Nucleic Acids Res, 41, D1096, 10.1093/nar/gks966 · DOI · Paper page
- Yang, 2013, Protein–ligand binding site recognition using complementary binding-specific substructure comparison and sequence profile alignment, Bioinformatics, 29, 2588, 10.1093/bioinformatics/btt447 · DOI · Paper page
- Yang, 2012, A new size-independent score for pairwise protein structure alignment and its application to structure classification and nucleic-acid binding prediction, Prot Struc Func Bioinform, 80, 2080, 10.1002/prot.24100 · DOI · Paper page
- Ying, 2021, Do transformers really perform badly for graph representation?, Adv Neural Inf Process Syst, 34, 28877
- Yu, 2013, Designing template-free predictor for targeting protein-ligand binding sites with classifier ensemble and spatial clustering, IEEE ACM Trans Comput Biol Bioinform, 10, 994, 10.1109/TCBB.2013.104 · DOI · Paper page
- Yuan, 2021, Structure-aware protein-protein interaction site prediction using deep graph convolutional network, Bioinformatics, 38, 125, 10.1093/bioinformatics/btab643 · DOI · Paper page
- Zeng, 2020, Protein–protein interaction site prediction through combining local and global features with deep neural networks, Bioinformatics, 36, 1114, 10.1093/bioinformatics/btz699 · DOI · Paper page
- Zhang, 2019, Sequence-based prediction of protein-protein interaction sites by simplified long short-term memory network, Neurocomputing, 357, 86, 10.1016/j.neucom.2019.05.013 · DOI · Paper page
- Zhang, 2019, SCRIBER: accurate and partner type-specific prediction of protein-binding residues from proteins sequences, Bioinformatics, 35, i343, 10.1093/bioinformatics/btz324 · DOI · Paper page
- Zhao, 2018, Improving sequence-based prediction of protein-peptide binding residues by introducing intrinsic disorder and a consensus method, J Chem Inf Model, 58, 1459, 10.1021/acs.jcim.8b00019 · DOI · Paper page
- Zheng, 2020, Predicting retrosynthetic reactions using self-corrected transformer neural networks, J Chem Inf Model, 60, 47, 10.1021/acs.jcim.9b00949 · DOI · Paper page
- Zhu, 2019, DNAPred: accurate identification of DNA-binding sites from protein sequence by ensembled hyperplane-distance-based support vector machines, J Chem Inf Model, 59, 3057, 10.1021/acs.jcim.8b00749 · DOI · Paper page
Cited by
Large language models transform biological research: from architecture to utilization
Wang, T., & Luo, Z. (2025) · Science China Information Sciences · doi:10.1007/s11432-024-4466-3 · Cited by 14
(2024) · doi:10.1093/bib/bbae146 · Cited by 6
Zhang, S., Xie, L., Tiourine, D., & Xie, L. (2026) · Journal of Cheminformatics · doi:10.1186/s13321-026-01227-0 · Cited by 2
Artificially intelligent modeling of macromolecular assemblies
Bini, M., & Tozzini, V. (2026) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2026.1856649
Metadata
- Authors
- Fang, Y., Jiang, Y., Wei, L., Ma, Q., Ren, Z., Yuan, Q., & Wei, D.-Q.
- Venue
- Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2023-11-28
- Volume
- 39
- Issue
- 12
- ISSN
- 1367-4803 (ISSN-L)
- 1367-4811
- OpenAlex
- W4389076414
- Cited by
- 100
- References
- 16
- FWCI
- 7.47
- Type
- journal article
- Language
- en
- Open access
- gold
- Funders
- National Science Foundation of China (32070662, 61832019, 32030063)
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.
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: DeepProSite: structure-aware protein binding site prediction using ESMFold and pretrained language model. https://app.cochat.ai/papers/10.1093/bioinformatics/btad718
- The paper
- Fang, Y., Jiang, Y., Wei, L., Ma, Q., Ren, Z., Yuan, Q., & Wei, D.-Q. (2023). DeepProSite: structure-aware protein binding site prediction using ESMFold and pretrained language model. Bioinformatics. https://doi.org/10.1093/bioinformatics/btad718
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.