Skip to content
2000
Volume 20, Issue 9
  • ISSN: 1574-8936
  • E-ISSN: 2212-392X

Abstract

Background

Deep learning models have gained significant traction in predicting drug-target binding affinity, primarily focusing on deciphering intricate drug-target relationships. However, these models often overlook intermediate representations, thus failing to capture the holistic characteristics of proteins crucial for discerning drug-target interactions.

Methods

This study proposes a novel deep-learning model that captures comprehensive and long-range dependencies within protein sequences. Leveraging deep feature engineering and an inverted Transformer module, it integrates multi-scale chemical information of drug molecules using graph neural networks and hierarchical attention mechanisms.

Results

The proposed model achieves state-of-the-art performance across multiple drug-target interaction datasets. It obtains MSE losses of 0.229 and 0.162 on the Davis and KIBA datasets, respectively, and AUC scores of 0.982 and 0.985 on the Human and datasets.

Conclusion

These results demonstrate the model's superior efficacy in predicting drug-target affinity and interactions, showcasing its potential to expedite drug discovery processes.

© 2025 Bentham Science Publishers
Loading

Article metrics loading...

/content/journals/cbio/10.2174/0115748936320436240826114416
2024-09-26
2025-10-25

  • From This Site

    /content/journals/cbio/10.2174/0115748936320436240826114416
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. SadybekovA.V. KatritchV. Computational approaches streamlining drug discovery.Nature2023616795867368510.1038/s41586‑023‑05905‑z 37100941
     [Google Scholar]
  2. ChenW. LiuX. ZhangS. ChenS. Artificial intelligence for drug discovery: Resources, methods, and applications.Mol. Ther. Nucleic Acids20233169170210.1016/j.omtn.2023.02.019 36923950
     [Google Scholar]
  3. AskrH. ElgeldawiE. Aboul EllaH. ElshaierY.A.M.M. GomaaM.M. HassanienA.E. Deep learning in drug discovery: An integrative review and future challenges.Artif. Intell. Rev.20235675975603710.1007/s10462‑022‑10306‑1 36415536
     [Google Scholar]
  4. NandiS. BhaduriS. DasD. GhoshP. MandalM. MitraP. Deciphering the lexicon of protein targets: a review on multifaceted drug discovery in the Era of artificial intelligence.Mol. Pharm.20242141563159010.1021/acs.molpharmaceut.3c01161 38466810
     [Google Scholar]
  5. SuJ. YangL. SunZ. ZhanX. Personalized drug therapy: Innovative concept guided with proteoformics.Mol. Cell. Proteomics202423310073710.1016/j.mcpro.2024.100737 38354979
     [Google Scholar]
  6. ZhouY. ZhangY. ZhaoD. TTD: Therapeutic target database describing target druggability information.Nucleic Acids Res.202452D1D1465D147710.1093/nar/gkad751 37713619
     [Google Scholar]
  7. LiangX. RenH. HanF. LiangR. ZhaoJ. LiuH. The new direction of drug development: Degradation of undruggable targets through targeting chimera technology.Med. Res. Rev.202444263268510.1002/med.21992 37983964
     [Google Scholar]
  8. WangH. XiongR. LaiL. Rational drug design targeting intrinsically disordered proteins.Wiley Interdiscip. Rev. Comput. Mol. Sci.2023136e168510.1002/wcms.1685
     [Google Scholar]
  9. LiuC. Application of FRET and TBET in bioimaging and biosensors.10.1117/12.3012863
     [Google Scholar]
  10. ZhangR. ZhuB. JiangT. CuiZ. WuH. Enhancing drug-target binding affinity prediction through deep learning and protein secondary structure integration.Curr. Bioinform.2024191094395210.2174/0115748936285519240110070209
     [Google Scholar]
  11. WangC. ZouQ. Prediction of protein solubility based on sequence physicochemical patterns and distributed representation information with DeepSoluE.BMC Biol.20232111210.1186/s12915‑023‑01510‑8 36694239
     [Google Scholar]
  12. TianC. WangL. CuiZ. WuH. GTAMP-DTA: Graph transformer combined with attention mechanism for drug-target binding affinity prediction.Comput. Biol. Chem.202410810798210.1016/j.compbiolchem.2023.107982 38039800
     [Google Scholar]
  13. ShahinM.H. BarthA. PodichettyJ.T. Artificial intelligence: from buzzword to useful tool in clinical pharmacology.Clin. Pharmacol. Ther.2024115469870910.1002/cpt.3083 37881133
     [Google Scholar]
  14. RyanD.K. MacleanR.H. BalstonA. ScourfieldA. ShahA.D. RossJ. Artificial intelligence and machine learning for clinical pharmacology.Br. J. Clin. Pharmacol.202490362963910.1111/bcp.15930 37845024
     [Google Scholar]
  15. SharmaV. SinghA. ChauhanS. Role of artificial intelligence in drug discovery and target identification in cancer.Curr. Drug Deliv.202421687088610.2174/1567201821666230905090621 37670704
     [Google Scholar]
  16. AlamS. IsrarJ. KumarA. Artificial intelligence and machine learning in bioinformatics. advances in bioinformatics.SingaporeSpringer Nature202432134510.1007/978‑981‑99‑8401‑5_16
     [Google Scholar]
  17. D’SouzaS. PremaK.V. BalajiS. Machine learning models for drug-target interactions: Current knowledge and future directions.Drug Discov. Today202025474875610.1016/j.drudis.2020.03.003 32171918
     [Google Scholar]
  18. AhmedA. MamB. SowdhaminiR. DEELIG: A deep learning approach to predict protein-ligand binding affinity.Bioinform. Biol. Insights20211510.1177/11779322211030364 34290496
     [Google Scholar]
  19. GorantlaR. KubincovaA. WeißeA.Y. MeyA.S. From proteins to ligands: Decoding deep learning methods for binding affinity prediction.J. Chem. Inf. Model.2023 37983381
     [Google Scholar]
  20. QiH. YuT. YuW. LiuC. Drug-target affinity prediction with extended graph learning-convolutional networks.BMC Bioinformatics20242517510.1186/s12859‑024‑05698‑6 38365583
     [Google Scholar]
  21. PengX OuyangC LiuY YuY LiuJ ChenM. Multimodal drug target binding affinity prediction using graph local substructure.IEEE J Biomed Health Inform202411110.1109/JBHI.2024.3386815 38598378
     [Google Scholar]
  22. LuL. GouX. TanS.K. De novo design of drug-binding proteins with predictable binding energy and specificity.Science2024384669110611210.1126/science.adl5364 38574125
     [Google Scholar]
  23. KalematiM. Zamani EmaniM. KoohiS. DCGAN-DTA: Predicting drug-target binding affinity with deep convolutional generative adversarial networks.BMC Genomics202425141110.1186/s12864‑024‑10326‑x 38724911
     [Google Scholar]
  24. ZhangY. HuY. HanN. YangA. LiuX. CaiH. A survey of drug-target interaction and affinity prediction methods via graph neural networks.Comput. Biol. Med.202316310713610.1016/j.compbiomed.2023.107136 37329615
     [Google Scholar]
  25. WhiteJ.J. New methods for predicting drug molecule activity using deep learning.Biosci Methods2024151410.5376/bm.2024.15.0004
     [Google Scholar]
  26. ZhangH. LiuX. ChengW. WangT. ChenY. Prediction of drug-target binding affinity based on deep learning models.Comput. Biol. Med.202417410843510.1016/j.compbiomed.2024.108435 38608327
     [Google Scholar]
  27. ShiW. YangH. XieL. YinX.X. ZhangY. A review of machine learning-based methods for predicting drug-target interactions.Health Inf. Sci. Syst.20241213010.1007/s13755‑024‑00287‑6 38617016
     [Google Scholar]
  28. WangC. JuY. ZouQ. LinC. DeepAc4C: A convolutional neural network model with hybrid features composed of physicochemical patterns and distributed representation information for identification of N4-acetylcytidine in mRNA.Bioinformatics2021381525710.1093/bioinformatics/btab611 34427581
     [Google Scholar]
  29. MostofianB. MartinH.J. RazaviA. Targeted protein degradation: Advances, challenges, and prospects for computational methods.J. Chem. Inf. Model.202363175408543210.1021/acs.jcim.3c00603 37602861
     [Google Scholar]
  30. WuH. LiuJ. ZhangR. A review of deep learning methods for ligand based drug virtual screening.Fundamental Res2024471573710.1016/j.fmre.2024.02.011
     [Google Scholar]
  31. YingngamB. NavabhatraA. SillapapiboolP. AI-driven decision-making applications in pharmaceutical sciences. Using Traditional Design Methods to Enhance AI-Driven Decision Making.Pennsylvania, United StatesIGI Global202416310.4018/979‑8‑3693‑0639‑0.ch001
     [Google Scholar]
  32. SiddiquiB. YadavC.S. AkilM. Artificial Intelligence in Computer-Aided Drug Design (CADD) tools for the finding of potent biologically active small molecules.SSRN2024
     [Google Scholar]
  33. RahemR.A.A. Al-AkaylehF. Integrative Analysis of Genomic Data Types and AI Methodologies in Healthcare Applications. 2024 2nd International Conference on Cyber Resilience (ICCR), Dubai, United Arab Emirates,20241510.1109/ICCR61006.2024.10532802
     [Google Scholar]
  34. XuK. HuW. LeskovecJ. JegelkaS. How powerful are graph neural networks?arXiv:1810008262018
     [Google Scholar]
  35. VeličkovićP. CucurullG. CasanovaA. Graph attention networks.arXiv:1710109032017
     [Google Scholar]
  36. JuW. LuoX. QuM. TGNN: A joint semi-supervised framework for graph-level classification.arXiv:2304116882023
     [Google Scholar]
  37. JuW. MaoZ. YiS. Hypergraph-enhanced Dual Semi-supervised Graph Classification.arXiv:240504772024
     [Google Scholar]
  38. WuH CaoC XiaX LuQ Unified deep learning architecture for modeling biology sequence.IEEE/ACM Transact Computat Biol Bioinform201815514455210.1109/TCBB.2017.2760832
     [Google Scholar]
  39. LiuJ. LuY. GuanS. Drug-target interaction prediction by combining transformer and graph neural networks.CBIO2023184142610.2174/1574893618666230912141426
     [Google Scholar]
  40. SiddiqueF. AnwaarA. BashirM. Revisiting methotrexate and phototrexate Zinc15 library-based derivatives using deep learning in-silico drug design approach.Front Chem.202412138026610.3389/fchem.2024.1380266 38576849
     [Google Scholar]
  41. WangD.D. WuW. WangR. Structure-based, deep-learning models for protein-ligand binding affinity prediction.J. Cheminform.2024161210.1186/s13321‑023‑00795‑9 38173000
     [Google Scholar]
  42. DhudumR. GaneshpurkarA. PawarA. Revolutionizing drug discovery: A comprehensive review of AI applications.Drugs and Drug Candidates20243114817110.3390/ddc3010009
     [Google Scholar]
  43. Perez-LopezC. MolinaA. LozoyaE. SegarraV. MunicoyM. GuallarV. Combining machine-learning and molecular-modeling methods for drug-target affinity predictions.Wiley Interdiscip. Rev. Comput. Mol. Sci.2023134e165310.1002/wcms.1653
     [Google Scholar]
  44. SreeramanS. KannanM.P. Singh KushwahR.B. Drug design and disease diagnosis: The potential of deep learning models in biology.Curr. Bioinform.202318320822010.2174/1574893618666230227105703
     [Google Scholar]
  45. ArafatM.Y. HossainM.J. AlamM.M. Machine learning scopes on microgrid predictive maintenance: Potential frameworks, challenges, and prospects.Renew. Sustain. Energy Rev.202419011408810.1016/j.rser.2023.114088
     [Google Scholar]
  46. GauriauO. GalárragaL. BrunF. TermierA. DavadanL. JoudelatF. Comparing machine-learning models of different levels of complexity for crop protection: A look into the complexity-accuracy tradeoff.Smart Agricultural Technology2024710038010.1016/j.atech.2023.100380
     [Google Scholar]
  47. AliY. HussainF. HaqueM.M. Advances, challenges, and future research needs in machine learning-based crash prediction models: A systematic review.Accid. Anal. Prev.202419410737810.1016/j.aap.2023.107378 37976634
     [Google Scholar]
  48. ZengT. JinB. GladeT. Assessing the imperative of conditioning factor grading in machine learning-based landslide susceptibility modeling: A critical inquiry.Catena202423610773210.1016/j.catena.2023.107732
     [Google Scholar]
  49. WangJ. ChenY. ZouQ. Inferring gene regulatory network from single-cell transcriptomes with graph autoencoder model.PLoS Genet.2023199e101094210.1371/journal.pgen.1010942 37703293
     [Google Scholar]
  50. SinghR. SledzieskiS. BrysonB. CowenL. BergerB. Contrastive learning in protein language space predicts interactions between drugs and protein targets.Proc. Natl. Acad. Sci. USA202312024e222077812010.1073/pnas.2220778120 37289807
     [Google Scholar]
  51. HuX. LuY. TianG. BingP. WangB. HeB. Predicting herb-disease associations through graph convolutional network.Curr. Bioinform.202318761061910.2174/1574893618666230504143647
     [Google Scholar]
  52. de Sena Murteira PinheiroP. FrancoL.S. MontagnoliT.L. FragaC.A.M. Molecular hybridization: A powerful tool for multitarget drug discovery.Expert Opin. Drug Discov.202419445147010.1080/17460441.2024.2322990 38456452
     [Google Scholar]
  53. MonteiroN.R.C. OliveiraJ.L. ArraisJ.P. TAG-DTA: Binding-region-guided strategy to predict drug-target affinity using transformers.Expert Syst. Appl.202423812233410.1016/j.eswa.2023.122334
     [Google Scholar]
  54. GuanJ. ZhouX. YangY. DecompDiff: Diffusion models with decomposed priors for structure-based drug design.ar-Xiv:2403079022024
     [Google Scholar]
  55. Datta-MannanA. ChoiH. JinZ. Reducing target binding affinity improves the therapeutic index of anti-MET antibody-drug conjugate in tumor bearing animals.PLoS One2024194e029370310.1371/journal.pone.0293703 38630694
     [Google Scholar]
  56. Martorell-MarugánJ. ChiericiM. Bandres-CigaS. JurmanG. Carmona-SáezP. Machine Learning Applications in the Study of Parkinson’s Disease: A Systematic Review.Curr. Bioinform.202318757658610.2174/1574893618666230406085947
     [Google Scholar]
  57. TanH. WangZ. HuG. GAABind: A geometry-aware attention-based network for accurate protein-ligand binding pose and binding affinity prediction.Brief. Bioinform.2023251bbad46210.1093/bib/bbad462 38102069
     [Google Scholar]
  58. ZhangS. TianX. ChenC. AIGO-DTI: Predicting Drug-Target Interactions Based on Improved Drug Properties Combined with Adaptive Iterative Algorithms.J. Chem. Inf. Model.202464104373438410.1021/acs.jcim.4c00584 38743013
     [Google Scholar]
  59. LiuY. CuiM. GaoX. Structural connectome combining DTI features predicts postoperative language decline and its recovery in glioma patients.Eur. Radiol.20233442759277110.1007/s00330‑023‑10212‑2 37736802
     [Google Scholar]
  60. ZhouG. XiaoN.G. SunY. LiH. LiuJ. Face recognition ability can be predicted by microstructural properties of white matter: A study of diffusion tensor imaging (DTI).Cereb. Cortex202333138442845510.1093/cercor/bhad130 37170639
     [Google Scholar]
  61. DehghanA. AbbasiK. RazzaghiP. BanadkukiH. GharaghaniS. CCL-DTI: Contributing the contrastive loss in drug-target interaction prediction.BMC Bioinformatics20242514810.1186/s12859‑024‑05671‑3 38291364
     [Google Scholar]
  62. SofosF RoukaE TriantafylliaV Development and validation of a symbolic regression-based machine learning method to predict COVID-19 in-hospital mortality among vaccinated patients.Health Technol (Berl)202410.1007/s12553‑024‑00886‑z
     [Google Scholar]
  63. NguyenT. LeH. QuinnT.P. GraphDTA: Predicting drug-target binding affinity with graph neural networks.Bioinformatics202137811401147
     [Google Scholar]
  64. TsubakiM. TomiiK. SeseJ. Compound-protein interaction prediction with end-to-end learning of neural networks for graphs and sequences.Bioinformatics201935230931810.1093/bioinformatics/bty535 29982330
     [Google Scholar]
  65. ChenL. TanX. WangD. TransformerCPI: Improving compound-protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments.Bioinformatics202036164406441410.1093/bioinformatics/btaa524 32428219
     [Google Scholar]
  66. LiP. LiY. HsiehC.Y. TrimNet: Learning molecular representation from triplet messages for biomedicine.Brief. Bioinform.2021224bbaa26610.1093/bib/bbaa266 33147620
     [Google Scholar]
  67. MitchellD.C. KuljaninM. LiJ. A proteome-wide atlas of drug mechanism of action.Nat. Biotechnol.202341684585710.1038/s41587‑022‑01539‑0 36593396
     [Google Scholar]
  68. SunarjoM.S. GanH.S. SetiadiD.R.I.M. High-performance convolutional neural network model to identify COVID-19 in medical images.Journal of Computing Theories and Applications202311193010.33633/jcta.v1i1.8936
     [Google Scholar]
  69. RastogiD. JohriP. TiwariV. ElngarA.A. Multi-class classification of brain tumour magnetic resonance images using multi-branch network with inception block and five-fold cross validation deep learning framework.Biomed. Signal Process. Control20248810560210.1016/j.bspc.2023.105602
     [Google Scholar]
  70. LiuY. HuT. ZhangH. iTransformer: Inverted transformers are effective for time series forecasting.arXiv:2310066252023
     [Google Scholar]
  71. Quispe-RicaldeM.A. Castelán-SánchezH.G. Meza-RodríguezP.M. Evidence of natural selection and dominance of SARS-CoV-2 variant Lambda (C.37) over variants of concern in Cusco, Peru.Arch. Virol.202316838810.1007/s00705‑022‑05645‑x 36786950
     [Google Scholar]
  72. YangR. ZhangC. GaoR. ZhangL. A novel feature extraction method with feature selection to identify golgi-resident protein types from imbalanced data.Int. J. Mol. Sci.201617221810.3390/ijms17020218 26861308
     [Google Scholar]
  73. NietoJ.J. TorresA. GeorgiouD.N. KarakasidisT.E. Fuzzy polynucleotide spaces and metrics.Bull. Math. Biol.200668370372510.1007/s11538‑005‑9020‑5 16794951
     [Google Scholar]
  74. BommidiB.S. TeeparthiK. KosanaV. Hybrid wind speed forecasting using ICEEMDAN and transformer model with novel loss function.Energy202326512638310.1016/j.energy.2022.126383
     [Google Scholar]
  75. ZhangW. A novel loss function considering the distance between forecasting and historical values in financial time series forecasting models.Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference8-10 December, 2023Guangzhou, China2023
     [Google Scholar]
  76. SoutoH.G. MoradiA. A novel loss function for neural network models exploring stock realized volatility using Wasserstein Distance.Decision Analytics Journal20241010036910.1016/j.dajour.2023.100369
     [Google Scholar]
  77. WulczynE. SteinerD.F. XuZ. Deep learning-based survival prediction for multiple cancer types using histopathology images.PLoS One2020156e023367810.1371/journal.pone.0233678 32555646
     [Google Scholar]
  78. BlankemeierL. YaoL. LongJ. Skeletal muscle area on CT: Determination of an optimal height scaling power and testing for mortality risk prediction.AJR Am. J. Roentgenol.20242221e2329889 37877596
     [Google Scholar]
  79. NahmF.S. Receiver operating characteristic curve: Overview and practical use for clinicians.Korean J. Anesthesiol.2022751253610.4097/kja.21209 35124947
     [Google Scholar]
  80. KannanA.S. Asymptotic confidence interval approach to estimate the portion of area under multi-class ROC curve.Stat. Appl.2024221263277
     [Google Scholar]
  81. IbrarF. AliS. ShahI. A comparison of single- and double-threshold ROC plots for mixture distributions.J. Appl. Stat.202451225627810.1080/02664763.2022.2122027 38283053
     [Google Scholar]
  82. WangL CarvalhoL. Multiclass ROC2024Available From: https://paperswithcode.com/paper/multiclass-roc
  83. ZhouQ.M. ZheL. BrookeR.J. HudsonM.M. YuanY. A relationship between the incremental values of area under the ROC curve and of area under the precision-recall curve.Diagn. Progn. Res.2021511310.1186/s41512‑021‑00102‑w 34261544
     [Google Scholar]
  84. BeyeneK.M. ChenD.G. KifleY.G. A novel nonparametric time-dependent precision-recall curve estimator for right-censored survival data.Biom. J.2024663230013510.1002/bimj.202300135 38637327
     [Google Scholar]
  85. SzabóS. HolbI.J. Abriha-MolnárV.É. SzatmáriG. SinghS.K. AbrihaD. Classification assessment tool: A program to measure the uncertainty of classification models in terms of class-level metrics.Appl. Soft Comput.202415511146810.1016/j.asoc.2024.111468
     [Google Scholar]
  86. RichardsonE. The receiver operating characteristic curve accurately assesses imbalanced datasets.Patterns20245610099411468
     [Google Scholar]
  87. KarimiM. WuD. WangZ. ShenY. KronRLS: A sequence-based deep-learning model for predicting protein-small molecule affinity.Bioinformatics201935183329333810.1093/bioinformatics/btz111
     [Google Scholar]
  88. YangZ. ZhongW. ZhaoL. Yu-Chian ChenC. MGraphDTA: Deep multiscale graph neural network for explainable drug-target binding affinity prediction.Chem. Sci. (Camb.)202213381683310.1039/D1SC05180F 35173947
     [Google Scholar]
/content/journals/cbio/10.2174/0115748936320436240826114416
Loading
Predicting Drug-target Binding Affinity Based on Graph Isomorphism Network and iTransformer
Curr Bioinform 20, 777 (2025); https://doi.org/10.2174/0115748936320436240826114416
/content/journals/cbio/10.2174/0115748936320436240826114416
/content/journals/cbio/10.2174/0115748936320436240826114416
Loading

Data & Media loading...


  • Article Type:     Research Article
Keyword(s): binding affinity; drug molecules; Drug-target interactions; graph isomorphism network; hierarchical attention mechanisms; iTransformer; protein sequence

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test