Enhancing Drug-Target Affinity Prediction with Multi-scale Graph Attention Network and Attention Mechanism
DOI:
https://doi.org/10.26555/jiteki.v10i4.30425Keywords:
Drug-target affinity, Drug graph, Protein sequences, Graph attention network, Attention mechanismAbstract
Drug-target affinity (DTA) prediction is critical to drug discovery, yet traditional experimental methods are expensive and time-consuming. Existing computational approaches often struggle with limitations in representing the structural and sequential complexities of drugs and proteins, resulting in suboptimal prediction accuracy. This study proposes a novel framework integrating Graph Attention Networks (GAT) for drug molecular and motif graphs and Bidirectional Long Short-Term Memory (BiLSTM) for protein sequences. A two-sided multi-head attention mechanism is utilized to dynamically model drug-protein interactions, enhancing robustness and accuracy. This research contribution is the development of a robust computational model that improves the accuracy of DTA predictions, reducing dependency on traditional laboratory methods. The integration of structural and sequential features provides a more comprehensive representation of drug-protein interactions. The study utilizes the Davis and KIBA, a binding affinity datasets that is widely used. the proposed model achieving the lowest Mean Squared Error (MSE) of 0.3209 and 0.1864, the highest Concordance Index (CI) of 0.8646 and 0.8616, and the highest of 0.5046 and 0.6672, respectively, outperforming baseline models. In conclusion, this study showed the proposed approach as a reliable method for DTA prediction, offering a faster and more accurate alternative in the drug discovery research field. However, there are still limitations, such as high computational complexity and the GAT model still uses static attention. Future work will focus on addressing this issue, testing the model across broader datasets, and implementing additional drug and target representation for richer feature extraction.
References
[1] K. Sachdev dan M. K. Gupta, “A comprehensive review of feature based methods for drug target interaction prediction,” J. Biomed. Inform., vol. 93, pp. 103159, Mei 2019, https://doi.org/10.1016/j.jbi.2019.103159.
[2] Y. Zhang, Y. Hu, N. Han, A. Yang, X. Liu, dan H. Cai, “A survey of drug-target interaction and affinity prediction methods via graph neural networks,” Comput. Biol. Med., vol. 163, pp. 107136, Sep 2023, https://doi.org/10.1016/j.compbiomed.2023.107136.
[3] A. Suruliandi, T. Idhaya, dan S. P. Raja, “Drug Target Interaction Prediction Using Machine Learning Techniques – A Review,” Int. J. Interact. Multimed. Artif. Intell., vol. 8, no. 6, pp. 86, 2024, https://doi.org/10.9781/ijimai.2022.11.002.
[4] D. Butina, M. D. Segall, dan K. Frankcombe, “Predicting ADME properties in silico: methods and models,” Drug Discov. Today, vol. 7, no. 11, pp. S83–S88, Mei 2002, https://doi.org/10.1016/S1359-6446(02)02288-2.
[5] E. Byvatov, U. Fechner, J. Sadowski, dan G. Schneider, “Comparison of Support Vector Machine and Artificial Neural Network Systems for Drug/Nondrug Classification,” J. Chem. Inf. Comput. Sci., vol. 43, no. 6, pp. 1882–1889, Nov 2003, https://doi.org/10.1021/ci0341161.
[6] H. Bhargava, A. Sharma, dan P. Suravajhala, “Chemogenomic Approaches for Revealing Drug Target Interactions inDrug Discovery,” Curr. Genomics, vol. 22, no. 5, pp. 328–338, Des 2021, https://doi.org/10.2174/1389202922666210920125800.
[7] H. Li et al., “TarFisDock: a web server for identifying drug targets with docking approach,” Nucleic Acids Res., vol. 34, pp. W219–W224, Jul 2006, https://doi.org/10.1093/nar/gkl114.
[8] A. C. Cheng et al., “Structure-based maximal affinity model predicts small-molecule druggability,” Nat. Biotechnol., vol. 25, no. 1, pp. 71–75, Jan 2007, https://doi.org/10.1038/nbt1273.
[9] G. Pujadas et al., “Protein-ligand Docking: A Review of Recent Advances and Future Perspectives,” Curr. Pharm. Anal., vol. 4, no. 1, pp. 1–19, Feb 2008, https://doi.org/10.2174/157341208783497597.
[10] W. Zhang, Y. Chen, dan D. Li, “Drug-Target Interaction Prediction through Label Propagation with Linear Neighborhood Information,” Molecules, vol. 22, no. 12, pp. 2056, Nov 2017, https://doi.org/10.3390/molecules22122056.
[11] X. Zhang, L. Li, M. K. Ng, dan S. Zhang, “Drug–target interaction prediction by integrating multiview network data,” Comput. Biol. Chem., vol. 69, pp. 185–193, Agu 2017, https://doi.org/10.1016/j.compbiolchem.2017.03.011.
[12] Z. Shi dan J. Li, “Drug-Target Interaction Prediction with Weighted Bayesian Ranking,” dalam Proceedings of the 2nd International Conference on Biomedical Engineering and Bioinformatics, pp. 19–24, Sep 2018, https://doi.org/10.1145/3278198.3278210.
[13] A. C. A. Nascimento, R. B. C. Prudêncio, dan I. G. Costa, “A multiple kernel learning algorithm for drug-target interaction prediction,” BMC Bioinformatics, vol. 17, no. 1, pp. 46, Jan 2016, https://doi.org/10.1186/s12859-016-0890-3.
[14] Z. Li et al., “In silico prediction of drug-target interaction networks based on drug chemical structure and protein sequences,” Sci. Rep., vol. 7, no. 1, pp. 11174, Sep 2017, https://doi.org/10.1038/s41598-017-10724-0.
[15] M. Ohue, T. Yamazaki, T. Ban, dan Y. Akiyama, “Link Mining for Kernel-Based Compound-Protein Interaction Predictions Using a Chemogenomics Approach,” Intelligent Computing Theories and Application, vol. 10362, pp. 549–558, 2017, https://doi.org/10.1007/978-3-319-63312-1_48.
[16] Y.-B. Wang, Z.-H. You, S. Yang, H.-C. Yi, Z.-H. Chen, dan K. Zheng, “A deep learning-based method for drug-target interaction prediction based on long short-term memory neural network,” BMC Med. Inform. Decis. Mak., vol. 20, no. S2, pp. 49, Mar 2020, https://doi.org/10.1186/s12911-020-1052-0.
[17] L. Xu, X. Ru, dan R. Song, “Application of Machine Learning for Drug–Target Interaction Prediction,” Front. Genet., vol. 12, pp. 680117, Jun 2021, https://doi.org/10.3389/fgene.2021.680117.
[18] N. R. C. Monteiro, B. Ribeiro, dan J. P. Arrais, “Drug-Target Interaction Prediction: End-to-End Deep Learning Approach,” IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 18, no. 6, pp. 2364–2374, Nov 2021, https://doi.org/10.1109/TCBB.2020.2977335.
[19] C. Chen, H. Shi, Y. Han, Z. Jiang, X. Cui, dan B. Yu, “DNN-DTIs: improved drug-target interactions prediction using XGBoost feature selection and deep neural network,” Computers in Biology and Medicine, vol. 136, p. 104676. 2021, https://doi.org/10.1101/2020.08.11.247437.
[20] S. M. Hasan Mahmud, W. Chen, H. Jahan, B. Dai, S. U. Din, dan A. M. Dzisoo, “DeepACTION: A deep learning-based method for predicting novel drug-target interactions,” Anal. Biochem., vol. 610, pp. 113978, Des 2020, https://doi.org/10.1016/j.ab.2020.113978.
[21] G. Liu et al., “GraphDTI: A robust deep learning predictor of drug-target interactions from multiple heterogeneous data,” J. Cheminformatics, vol. 13, no. 1, pp. 58, Agu 2021, https://doi.org/10.1186/s13321-021-00540-0.
[22] H. Öztürk, A. Özgür, dan E. Ozkirimli, “DeepDTA: deep drug–target binding affinity prediction,” Bioinformatics, vol. 34, no. 17, pp. i821–i829, Sep 2018, https://doi.org/10.1093/bioinformatics/bty593.
[23] H. Öztürk, E. Ozkirimli, dan A. Özgür, “WideDTA: prediction of drug-target binding affinity,” arXiv: arXiv:1902.04166, 2019, https://doi.org/10.48550/arXiv.1902.04166.
[24] Y. Pu, J. Li, J. Tang, dan F. Guo, “DeepFusionDTA: Drug-Target Binding Affinity Prediction With Information Fusion and Hybrid Deep-Learning Ensemble Model,” IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 19, no. 5, pp. 2760–2769, Sep 2022, https://doi.org/10.1109/TCBB.2021.3103966.
[25] W. Yuan, G. Chen, dan C. Y.-C. Chen, “FusionDTA: attention-based feature polymerizer and knowledge distillation for drug-target binding affinity prediction,” Brief. Bioinform., vol. 23, no. 1, pp. bbab506, Jan 2022, https://doi.org/10.1093/bib/bbab506.
[26] A. Ghimire, H. Tayara, Z. Xuan, dan K. T. Chong, “CSatDTA: Prediction of Drug–Target Binding Affinity Using Convolution Model with Self-Attention,” Int. J. Mol. Sci., vol. 23, no. 15, pp. 8453, Jul 2022, https://doi.org/10.3390/ijms23158453.
[27] L. Deng, Y. Zeng, H. Liu, Z. Liu, dan X. Liu, “DeepMHADTA: Prediction of Drug-Target Binding Affinity Using Multi-Head Self-Attention and Convolutional Neural Network,” Curr. Issues Mol. Biol., vol. 44, no. 5, pp. 2287–2299, Mei 2022, https://doi.org/10.3390/cimb44050155.
[28] Q. Zhao, F. Xiao, M. Yang, Y. Li, dan J. Wang, “AttentionDTA: prediction of drug–target binding affinity using attention model,” dalam 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 64–69, 2019, https://doi.org/10.1109/BIBM47256.2019.8983125.
[29] K. Abbasi, P. Razzaghi, A. Poso, M. Amanlou, J. B. Ghasemi, dan A. Masoudi-Nejad, “DeepCDA: deep cross-domain compound–protein affinity prediction through LSTM and convolutional neural networks,” Bioinformatics, vol. 36, no. 17, pp. 4633–4642, Nov 2020, https://doi.org/10.1093/bioinformatics/btaa544.
[30] A. Mahdaddi, S. Meshoul, dan M. Belguidoum, “EA-based hyperparameter optimization of hybrid deep learning models for effective drug-target interactions prediction,” Expert Syst. Appl., vol. 185, pp. 115525, Des 2021, https://doi.org/10.1016/j.eswa.2021.115525.
[31] Q. Zhao, G. Duan, M. Yang, Z. Cheng, Y. Li, dan J. Wang, “AttentionDTA: Drug–Target Binding Affinity Prediction by Sequence-Based Deep Learning With Attention Mechanism,” IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 20, no. 2, pp. 852–863, Mar 2023, https://doi.org/10.1109/TCBB.2022.3170365.
[32] T. Nguyen, H. Le, T. P. Quinn, T. Nguyen, T. D. Le, dan S. Venkatesh, “GraphDTA: predicting drug–target binding affinity with graph neural networks,” Bioinformatics, vol. 37, no. 8, pp. 1140–1147, Mei 2021, https://doi.org/10.1093/bioinformatics/btaa921.
[33] S. Zhang, M. Jiang, S. Wang, X. Wang, Z. Wei, dan Z. Li, “SAG-DTA: Prediction of Drug–Target Affinity Using Self-Attention Graph Network,” Int. J. Mol. Sci., vol. 22, no. 16, pp. 8993, Agu 2021, https://doi.org/10.3390/ijms22168993.
[34] Y. Liang, S. Jiang, M. Gao, F. Jia, Z. Wu, dan Z. Lyu, “GLSTM-DTA: Application of Prediction Improvement Model Based on GNN and LSTM,” J. Phys. Conf. Ser., vol. 2219, no. 1, pp. 012008, Apr 2022, https://doi.org/10.1088/1742-6596/2219/1/012008.
[35] A. Banerjee, Z.-H. Zhou, E. E. Papalexakis, dan M. Riondato, Ed., Proceedings of the 2022 SIAM International Conference on Data Mining (SDM). Philadelphia, PA: Society for Industrial and Applied Mathematics, 2022. https://doi.org/10.1137/1.9781611977172.
[36] P. Chen, H. Shen, Y. Zhang, B. Wang, dan P. Gu, “SGNet: Sequence-Based Convolution and Ligand Graph Network for Protein Binding Affinity Prediction,” IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 20, no. 5, pp. 3257–3266, Sep 2023, https://doi.org/10.1109/TCBB.2023.3262821.
[37] W. Qiu, Q. Liang, L. Yu, X. Xiao, W. Qiu, dan W. Lin, “LSTM-SAGDTA: Predicting Drug-target Binding Affinity with an Attention GraphNeural Network and LSTM Approach,” Curr. Pharm. Des., vol. 30, no. 6, pp. 468–476, Feb 2024, https://doi.org/10.2174/0113816128282837240130102817.
[38] H. N. T. Tran, J. J. Thomas, dan N. H. Ahamed Hassain Malim, “DeepNC: a framework for drug-target interaction prediction with graph neural networks,” PeerJ, vol. 10, pp. e13163, Mei 2022, https://doi.org/10.7717/peerj.13163.
[39] S. Wang et al., “MSGNN-DTA: Multi-Scale Topological Feature Fusion Based on Graph Neural Networks for Drug–Target Binding Affinity Prediction,” Int. J. Mol. Sci., vol. 24, no. 9, pp. 8326, Mei 2023, https://doi.org/10.3390/ijms24098326.
[40] M. I. Davis et al., “Comprehensive analysis of kinase inhibitor selectivity,” Nat. Biotechnol., vol. 29, no. 11, pp. 1046–1051, Nov 2011, https://doi.org/10.1038/nbt.1990.
[41] J. Tang et al., “Making Sense of Large-Scale Kinase Inhibitor Bioactivity Data Sets: A Comparative and Integrative Analysis,” J. Chem. Inf. Model., vol. 54, no. 3, pp. 735–743, Mar 2014, https://doi.org/10.1021/ci400709d.
[42] B. Ramsundar, P. Eastman, P. Walters, dan V. Pande, Deep learning for the life sciences: applying deep learning to genomics, microscopy, drug discovery and more, First edition. Sebastopol, CA: O’Reilly Media, 2019. https://books.google.co.id/books?hl=id&lr=&id=6OiRDwAAQBAJ.
[43] P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Liò, dan Y. Bengio, “Graph Attention Networks,” arXiv: arXiv:1710.10903, 2024, http://arxiv.org/abs/1710.10903.
[44] S. Hochreiter dan J. Schmidhuber, “Long Short-Term Memory,” Neural Comput., vol. 9, no. 8, pp. 1735–1780, Nov 1997, https://doi.org/10.1162/neco.1997.9.8.1735.
[45] M. Gönen dan G. Heller, “Concordance probability and discriminatory power in proportional hazards regression,” Biometrika, vol. 92, no. 4, pp. 965–970, Des 2005, https://doi.org/10.1093/biomet/92.4.965.
[46] K. Roy, P. Chakraborty, I. Mitra, P. K. Ojha, S. Kar, dan R. N. Das, “Some case studies on application of ‘ r m 2 ’ metrics for judging quality of quantitative structure–activity relationship predictions: Emphasis on scaling of response data,” J. Comput. Chem., vol. 34, no. 12, pp. 1071–1082, Mei 2013, https://doi.org/10.1002/jcc.23231.
[47] T. Pahikkala et al., “Toward more realistic drug-target interaction predictions,” Brief. Bioinform., vol. 16, no. 2, pp. 325–337, Mar 2015, https://doi.org/10.1093/bib/bbu010.
[48] T. N. Kipf dan M. Welling, “Semi-Supervised Classification with Graph Convolutional Networks,” arXiv: arXiv:1609.02907, 2017, http://arxiv.org/abs/1609.02907.
[49] D. E. Rumelhart, G. E. Hinton, dan R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, no. 6088, pp. 533–536, Okt 1986, https://doi.org/10.1038/323533a0.
[50] K. Cho et al., “Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation,” arXiv: arXiv:1406.1078, 2024. [Daring], http://arxiv.org/abs/1406.1078.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Muhammad Rizky Yusfian Yusuf, Isman Kurniawan
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with JITEKI agree to the following terms:
- Authors retain copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY-SA 4.0) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
This work is licensed under a Creative Commons Attribution 4.0 International License
