CoChat
Sign Up
Cited by 7Computational Drug Discovery MethodsProtein Degradation and InhibitorsMachine Learning in Bioinformatics

The DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.

DrugForm-DTA: Towards real-world drug-target binding affinity model

Open access
Khokhlov, Ivan, Tashchilova, Anna, Bugaev-Makarovskiy, Nikolai, Glushkova, Olga, Yudin, Vladimir, Keskinov, Anton, Yudin, Sergey, Svetlichnyy, Dmitry, Skvortsova, Veronika·2025·Computational and Structural Biotechnology Journal
DOI 10.1016/j.csbj.2025.09.023PMC PMC12495441

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Drug-target affinity (DTA) prediction is a fundamental problem in drug discovery. Computational methods for predicting DTA can greatly assist drug design by decreasing the search space and reducing the number of protein-ligand complexes with low affinity. Currently DTA approaches often do not require protein three-dimensional (3D) structural information, which is often not accessible. In this study we present the DrugForm-DTA model, which uses only structure-less representations of ligand and protein. It is a Transformer-based neural network with protein encoding based on ESM, and small molecule ligand encoding obtained with Chemformer. We evaluated the model on standard benchmarks Davis and KIBA, and revealed superior performance of DrugForm-DTA with best result for KIBA (MSE=0.117). Moreover, we developed a ready-to-use model using BindingDB dataset that was subjected to high-quality filtering and transformation. Overall, our method predicts drug-target affinity values with a confidence level comparable to a single in-vitro experiment. Also, we compared DrugForm-DTA against molecular modeling methods and revealed higher efficacy of the developed model for drug-target affinity predictions. Our investigation provides a high accuracy neural network model with performance comparable to experimental measurements, filtered and reassessed BindingDB dataset for further usage, and demonstrates outstanding applicability of the proposed method for DTA prediction.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1016/j.csbj.2025.09.023

Cites

6 of 6 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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

Flagged first, then by citations

Research works printed but not linked: 60

  1. Zhang S. Jiang M. Wang S. Wang X. Wei Z. and Li Z.: SAG-DTA: prediction of Drug–Target affinity using Self-Attention graph network. Int J Mol Sci 2021; 2210.3390/ijms22168993. 〈https://www.mdpi.com/1422-0067/22/16/8993〉. (URL). · DOI
  2. 10.1038/nbt.1990 · DOI · Paper page
  3. Tang J. Szwajda A. Shakyawar S. Xu T. Hintsanen P. Wennerberg K. and Aittokallio T.: Making sense of Large-Scale kinase inhibitor bioactivity data sets: a comparative and integrative analysis. J Chem Inf Model 2014; 54: 73510.1021/ci400709d. 〈https://doi.org/10.1021/ci400709d〉. URL https://doi.org/10.1021/ci400709d. · DOI
  4. 10.1093/nar/gkv1072 · DOI
  5. Benson M.L. Smith R.D. Khazanov N.A. Dimcheff B. Beaver J. Dresslar P. Nerothin J. and Carlson H.A.: Binding MOAD a highquality protein–ligand database. Nucleic Acids Res 2007; 36: D67410.1093/nar/gkm911. 〈https://academic.oup.com/nar/articlepdf/36/suppl_1/D674/7633925/gkm911.pdf〉. URL https://doi.org/10.1093/nar/gkm911. · DOI
  6. 10.1021/ci00057a005 · DOI · Paper page
  7. 10.1021/ci100050t · DOI
  8. 10.1021/c160017a018 · DOI
  9. Matsuzaka Y. and Uesawa Y.: Optimization of a Deep-Learning method based on the classification of images generated by parameterized deep snap a novel Molecular-Image-Input technique for quantitative Structure–Activity relationship (QSAR) analysis. Front Bioeng Biotechnol 2019; 710.3389/fbioe.2019.00065. 〈https://www.frontiersin.org/journals/bioengineering-and-biotechnology/arti〉. (URL). · DOI
  10. Khokhlov I. Krasnov L. Fedorov M.V. and Sosnin S.: Image2SMILES: Transformer-Based molecular optical recognition engine. ChemMethods 2022; 2: e20210006910.1002/cmtd.202100069. 〈https://chemistryeurope.onlinelibrary.wiley.com/doi/pdf/10.1002/cmtd.202100069〉. URL https://chemistry-europe.onlinelibrary.wiley.com/doi/abs/10.1002/cmtd.2021. · DOI
  11. 10.26599/BDMA.2022.9020005 · DOI
  12. Nguyen T. Le H. Quinn T.P. Nguyen T. Le T.D. and Venkatesh S.: GraphDTA: predicting drug–target binding affinity with graph neural networks. Bioinformatics 2020; 37: 114010.1093/bioinformatics/btaa921. 〈https://academic.oup.com/bioinformatics/articlepdf/37/8/1140/50340643/btaa921.pdf〉. URL https://doi.org/10.1093/bioinformatics/btaa921. · DOI
  13. 10.1039/D1SC05180F · DOI
  14. Qian Y. Ni W. Xianyu X. Tao L. and Wang Q.: DoubleSG-DTA: deep learning for drug discovery: case study on the Non-Small cell lung cancer with EGFRT790M mutation. Pharmaceutics 2023; 1510.3390/pharmaceutics15020675. 〈https://www.mdpi.com/1999-4923/15/2/675〉. (URL). · DOI
  15. H. Ye Y. Song B. Wang L. Wu S. He X.-C. Bo Z. Zhang HSGCL-DTA: Hybridscale Graph Contrastive Learning based Drug-Target Binding Affinity Prediction (2023) 947–954 doi:10.1109/ICTAI59109.2023.00142. · DOI
  16. 10.1371/journal.pcbi.1012100 · DOI
  17. öztüürk H. özgür A. and Ozkirimli E.: DeepDTA: deep drug–target binding¨ affinity prediction. Bioinformatics 2018; 34: i821. 〈https://academic.oup.com/bioinformatics/articlepdf/34/17/i821/50582374/bioinformatics 34 17 i821.pdf. doi:10.1093/bioinformatics/bty593〉. URL https://doi.org/10.1093/bioinformatics/bty593. · DOI
  18. Zhang L. Wang C.-C. and Chen X.: Predicting drug–target binding affinity through molecule representation block based on multi-head attention and skip connection. Brief Bioinforma 2022; 23: bbac46810.1093/bib/bbac468. 〈https://academic.oup.com/bib/articlepdf/23/6/bbac468/47144652/bbac468.pdf〉. URL https://doi.org/10.1093/bib/bbac468. · DOI
  19. Hua Y. Song X. Feng Z. and Wu X.: MFR-DTA: a multi-functional and robust model for predicting drug–target binding affinity and region. Bioinformatics 2023; 39: btad05610.1093/bioinformatics/btad056. 〈https://academic.oup.com/bioinformatics/articlepdf/39/2/btad056/49096141/btad056.pdf〉. URL https://doi.org/10.1093/bioinformatics/btad056. · DOI
  20. Chen H. Li D. Liao J. Wei L. and Wei L.: MultiscaleDTA: a multiscale-based method with a self-attention mechanism for drug-target binding affinity prediction. Methods 2022; 207: 10310.1016/j.ymeth.2022.09.006. 〈https://www.sciencedirect.com/science/article/pii/S1046202322002146〉. (URL). · DOI
  21. Chu Z. Huang F. Fu H. Quan Y. Zhou X. Liu S. and Zhang W.: Hierarchical graph representation learning for the prediction of drug-target binding affinity. Inf Sci 2022; 613: 50710.1016/j.ins.2022.09.043. 〈https://www.sciencedirect.com/science/article/pii/S0020025522010908〉. (URL). · DOI
  22. 10.1039/D3RA00281K · DOI
  23. 10.1371/journal.pcbi.1011036 · DOI
  24. 10.1109/JBHI.2023.3240305 · DOI
  25. X. Xiao W. Wang J. Xie L. Zhu G. Chen Z. Li T. Wang M. Xu HGTDP-DTA: Hybrid Graph-Transformer with Dynamic Prompt for Drug-Target Binding Affinity Prediction (Jun. 2024). doi:10.48550/arXiv.2406.17697. · DOI
  26. T. Mikolov K. Chen G. Corrado J. Dean Efficient Estimation of Word Representations in Vector Space eprint: 1301.3781 (2013). URL 〈https://arxiv.org/abs/1301.3781〉. · arXiv
  27. Ahmad W. Simon E. Chithrananda S. Grand G. and Ramsundar B.: ChemBERTa-2. Towards Chem Found Models 2022; . 〈https://arxiv.org/abs/2209.01712〉. (URL). · arXiv
  28. 10.1021/ed083p1633 · DOI
  29. 10.1038/s41598-021-94082-y · DOI
  30. Heller S.R. McNaught A. Pletnev I.V. Stein S. and Tchekhovskoi D.V.: InChI the IUPAC international chemical identifier. J Chemin 2015; 7. 〈https://api.semanticscholar.org/CorpusID:9709647〉. (URL). · DOI
  31. Vaswani A. Shazeer N. Parmar N. Uszkoreit J. Jones L. Gomez A.N. Kaiser L. and Polosukhin I.: Atten Is All You Need 2017; . 〈http://arxiv.org/abs/1706.03762〉. (URL). · arXiv
  32. 10.1088/2632-2153/ac3ffb · DOI
  33. 10.18653/v1/2020.acl-main.703 · DOI · Paper page
  34. Gorriz J.M. Clemente R.M. Segovia F. Ramirez J. Ortiz A. and Suckling J.: Is KFold Cross Valid Best Model Sel Method Mach Learn 2024; . 〈https://arxiv.org/abs/2401.16407〉. (URL). · arXiv
  35. Ott M. Edunov S. Baevski A. Fan A. Gross S. Ng N. Grangier D. and Auli M.: fairseq A Fast Extensible Toolkit Seq Model 2019; . 〈http://arxiv.org/abs/1904.01038〉. (URL). · DOI
  36. Heagerty P.J. and Zheng Y.: Survival model predictive accuracy and ROC curves. Biometrics 2005; 61: 9210.1111/j.0006341X.2005.030814.x. 〈https://academic.oup.com/biometrics/articlepdf/61/1/92/52029971/biometrics 61 1 92.pdf〉. URL https://doi.org/10.1111/j.0006-341X.2005.030814.x. · DOI
  37. 10.1038/s41598-022-08787-9 · DOI
  38. Coudert E. Gehant S. de Castro E. Pozzato M. Baratin D. Neto T. Sigrist C.J.A. Redaschi N. Bridge A. and Consortium T.U.: Annotation of biologically relevant ligands in UniProtKB using ChEBI. Bioinformatics 2022; 39: btac79310.1093/bioinformatics/btac793. 〈https://academic.oup.com/bioinformatics/articlepdf/39/1/btac793/48520867/btac793.pdf〉. URL https://doi.org/10.1093/bioinformatics/btac793. · DOI · Paper page
  39. Sharma P. Vaiwala R. Gopinath A.K. Chockalingam R. and Ayappa K.G.: Structure of the bacterial cell envelope and interactions with antimicrobials: insights from molecular dynamics simulations. Langmuir 2024; 40: 779110.1021/acs.langmuir.3c03474. 〈https://doi.org/10.1021/acs.langmuir.3c03474〉. URL https://doi.org/10.1021/acs.langmuir.3c03474. · DOI
  40. Strahl H. and Errington J.: Bacterial membranes: structure domains and function. Annu Rev Microbiol 2017; 71: 51910.1146/annurev-micro-102215-095630. 〈https://www.annualreviews.org/content/journals/10.1146/annurev-micro-10221〉. (URL). · DOI
  41. Hoare S.R.: The problems of applying classical pharmacology analysis to modern in vitro drug discovery assays: slow binding kinetics and high target concentration. SLAS Discov 2021; 26: 83510.1177/24725552211019653. 〈https://linkinghub.elsevier.com/retrieve/pii/S2472555222067351〉. (URL). · DOI
  42. Bastos M. Abian O. Johnson C.M. Ferreira-da Silva F. Vega S. Jimenez-Alesanco A. Ortega-Alarcon D. and Velazquez-Campoy A.: Isothermal titration calorimetry. Nat Rev Methods Prim 2023; 3: 1710.1038/s43586-023-00199-x. 〈https://www.nature.com/articles/s43586-023-00199-x〉. (URL). · DOI
  43. Berman H.M. Westbrook J. Feng Z. Gilliland G. Bhat T.N. Weissig H. Shindyalov I.N. and Bourne P.E.: The protein data bank. Nucleic Acids Res 2000; 48: 23510.1093/nar/28.1.235. 〈http://www.rcsb.org/pdb/〉. (URL). · DOI · Paper page
  44. 10.1007/s10822-0139644-8 · DOI
  45. 10.1002/jcc.10051 · DOI
  46. Friesner R.A. Banks J.L. Murphy R.B. Halgren T.A. Klicic J.J. Mainz D.T. Repasky M.P. Knoll E.H. Shelley M. Perry J.K. Shaw D.E. Francis P. and Shenkin P.S.: Glide: a new approach for rapid accurate docking and scoring. 1. method and assessment of docking accuracy. J Med Chem 2004; 47: 173910.1021/jm0306430. 〈https://doi.org/10.1021/jm0306430〉. URL https://doi.org/10.1021/jm0306430. · DOI · Paper page
  47. 10.1517/17460441.2015.1032936 · DOI
  48. Jacobson M.P. Pincus D.L. Rapp C.S. Day T.J. Honig B. Shaw D.E. and Friesner R.A.: A hierarchical approach to all-atom protein loop prediction. Protein Struct Funct Bioinforma 2004; 55: 35110.1002/prot.10613. 〈https://onlinelibrary.wiley.com/doi/pdf/10.1002/prot.10613〉. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/prot.10613. · DOI
  49. Hostaˇs J. Reza´ˇc J. and Hobza P.: On the performance of the semiem-ˇ pirical quantum mechanical PM6 and PM7 methods for noncovalent interactions. Chem Phys Lett 2013; 568-569: 16110.1016/j.cplett.2013.02.069. 〈https://www.sciencedirect.com/science/article/pii/S0009261413002935〉. (URL). · DOI
  50. 10.1039/P29930000799 · DOI
  51. Y. Kim Y. Jeong J. Kim E.K. Lee W.J. Kim I.S. Choi MolNet: A Chemically Intuitive Graph Neural Network for Prediction of Molecular Properties eprint: 2203.09456 (2022). URL 〈https://arxiv.org/abs/2203.09456〉. · DOI
  52. Bishop C.: Neural networks for pattern recognition. USA: Oxford University Press 1995. · DOI · Paper page
  53. K. Huang T. Fu W. Gao Y. Zhao Y. Roohani J. Leskovec C.W. Coley C. Xiao J. Sun M. Zitnik Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development eprint: 2102.09548 (2021). URL 〈https://arxiv.org/abs/2102.09548〉. · arXiv
  54. 10.1186/s13321-022-00591-x · DOI
  55. Hu Z. Liu W. Zhang C. Huang J. Zhang S. Yu H. Xiong Y. Liu H. Ke S. and Hong L.: SAM-DTA: a sequence-agnostic model for drug–target binding affinity prediction. Brief Bioinforma 2022; 24: bbac53310.1093/bib/bbac533. 〈https://academic.oup.com/bib/articlepdf/24/1/bbac533/48783303/bbac533.pdf〉. URL https://doi.org/10.1093/bib/bbac533. · DOI
  56. 10.1186/s13321-017-0232-0 · DOI · Paper page
  57. Q. Feng E. Dueva A. Cherkasov M. Ester PADME: A Deep Learningbased Framework for Drug-Target Interaction Prediction eprint: 1807.09741 (2019). URL 〈https://arxiv.org/abs/1807.09741〉. · arXiv
  58. Zhang Y.Z.: CrossDocked 2020; 10.21227/45C9-VG74. 〈https://ieee-dataport.org/documents/crossdocked2020〉. (URL). · DOI
  59. 10.1111/1541-4337.13280 · DOI
  60. Georgakis N. Ioannou E. Varotsou C. Premetis G. Chronopoulou E.G. and Labrou N.E.: Determination of Half-Maximal inhibitory concentration of an enzyme inhibitor. In: . Edited by N.E. Labrou. New York NY: Springer US 2020: 4110.1007/978-1-0716-0163-1 3. URL http://link.springer.com/10.1007/978-1-0716-0163-1 3. · DOI

Metadata

Authors
Khokhlov, I., Tashchilova, A., Bugaev-Makarovskiy, N., Glushkova, O., Yudin, V., Keskinov, A., Yudin, S., Svetlichnyy, D., & Skvortsova, V.
Affiliations
  • Federal State Budgetary Institution "Centre for Strategic Planning and Management of Biomedical Health Risks" of the Federal Medical and Biological Agency (RU)
  • Federal Medical-Biological Agency (RU)
Corresponding author
Svetlichnyy, Dmitry
ORCID
Venue
Computational and Structural Biotechnology Journal
Publisher
American Association for the Advancement of Science (AAAS)
Published
2025-01-01
Volume
27
Pages
4106-4120
ISSN
2001-0370 (ISSN-L)
DOI
10.1016/j.csbj.2025.09.023
PMC
PMC12495441
OpenAlex
W4414443788
Cited by
7
References
6
Citations by year
2026: 7
FWCI
3.62
Type
journal article
Language
en
Open access
Licence
Locations
  • Computational and Structural Biotechnology Journal (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Degradation and Inhibitors
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
Keywords
drug-target affinity prediction, neural network model

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