A Comparative Study of Graph Neural Network Models for Drug-Target Interaction Prediction

• Jaqueline Bitencourt UFRGS
• Anderson Tavares UFRGS

DOI: https://doi.org/10.5753/sbcas.2025.7729

Resumo

Accurately predicting drug-target interactions (DTI) is crucial for computational drug discovery, yet there’s a research gap in evaluating existing graph neural network (GNN) models rather than developing novel architectures. This study provides a comparative analysis of three state-of-the-art GNN architectures – GraphSAGE, Graph Attention Network (GAT), and Graph Isomorphism Network (GIN) – for predicting interactions between chemical compounds and five protein targets. Using a dataset of 73,938 samples representing interactions between compounds and five protein targets derived from PubChem, we implement a robust evaluation framework with hyperparameter optimization and cross-validation. Our results show GraphSAGE achieves the highest accuracy (93%) and precision (79%), while GIN exhibits superior recall (72%). This work contributes to the field by: (1) providing a comprehensive evaluation framework for GNN models in DTI prediction; (2) offering empirical evidence of architecture-specific advantages for different application contexts; and (3) introducing a new benchmark dataset that facilitates reproducibility and further research in computational drug discovery.

Referências

Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A nextgeneration hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.

Cheng, Z., Yan, C., Wu, F.-X., and Wang, J. (2021). Drug-target interaction prediction using multi-head self-attention and graph attention network. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 19(4):2208–2218.

Dandibhotla, S., Samudrala, M., Kaneriya, A., and Dakshanamurthy, S. (2025). Gnnseq: A sequence-based graph neural network for predicting protein–ligand binding affinity. Pharmaceuticals, 18(3):329.

Elbasani, E., Njimbouom, S. N., Oh, T.-J., Kim, E.-H., Lee, H., and Kim, J.-D. (2021). Gcrnn: graph convolutional recurrent neural network for compound–protein interaction prediction. BMC bioinformatics, 22(Suppl 5):616.

Hamilton, W., Ying, Z., and Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in neural information processing systems, 30.

Hasebe, T. (2021). Knowledge-embedded message-passing neural networks: improving molecular property prediction with human knowledge. ACS omega, 6(42):27955–27967.

Krenn, M., Häse, F., Nigam, A., Friederich, P., and Aspuru-Guzik, A. (2020). Self-referencing embedded strings (selfies): A 100% robust molecular string representation. Machine Learning: Science and Technology, 1(4):045024.

Li, Y., Qiao, G., Wang, K., and Wang, G. (2022). Drug–target interaction predication via multi-channel graph neural networks. Briefings in Bioinformatics, 23(1):bbab346.

Liu, S., Wang, Y., Deng, Y., He, L., Shao, B., Yin, J., Zheng, N., Liu, T.-Y., and Wang, T. (2022). Improved drug–target interaction prediction with intermolecular graph transformer. Briefings in Bioinformatics, 23(5):bbac162.

Nguyen, T., Le, H., Quinn, T. P., Nguyen, T., Le, T. D., and Venkatesh, S. (2021). Graphdta: predicting drug–target binding affinity with graph neural networks. Bioinformatics, 37(8):1140–1147.

Öztürk, H., Özgür, A., and Ozkirimli, E. (2018). Deepdta: deep drug–target binding affinity prediction. Bioinformatics, 34(17):i821–i829.

Reiser, P., Neubert, M., Eberhard, A., Torresi, L., Zhou, C., Shao, C., Metni, H., van Hoesel, C., Schopmans, H., Sommer, T., et al. (2022). Graph neural networks for materials science and chemistry. Communications Materials, 3(1):93.

Sachdev, K. and Gupta, M. K. (2019). A comprehensive review of feature based methods for drug target interaction prediction. Journal of Biomedical Informatics, 93:103159.

Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., and Monfardini, G. (2008). The graph neural network model. IEEE transactions on neural networks, 20(1):61–80.

Su, X., Hu, P., You, Z.-H., Philip, S. Y., and Hu, L. (2024). Dual-channel learning framework for drug-drug interaction prediction via relation-aware heterogeneous graph transformer. In Proceedings of AAAI, volume 38, pages 249–256.

Tang, X., Lei, X., and Zhang, Y. (2024). Prediction of drug-target affinity using attention neural network. International Journal of Molecular Sciences, 25(10):5126.

Tran, H. N. T., Thomas, J. J., and Malim, N. H. A. H. (2022). Deepnc: a framework for drug-target interaction prediction with graph neural networks. PeerJ, 10:e13163.

Tsubaki, M., Tomii, K., and Sese, J. (2019). Compound–protein interaction prediction with end-to-end learning of neural networks for graphs and sequences. Bioinformatics, 35(2):309–318.

Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903.

Wan, W., Silva, R., Odenweller, D. J., and Leeuwon, S. (2024). 6.03 - proprietary strategies in precision medicine. In Ramos, K. S., editor, Comprehensive Precision Medicine (First Edition), pages 197–220. Elsevier, Oxford, first edition edition.

Wang, H., Zhou, G., Liu, S., Jiang, J.-Y., and Wang, W. (2021). Drug-target interaction prediction with graph attention networks. arXiv preprint arXiv:2107.06099.

Wang, X., Wen, Y., Zhang, Y., Dai, C., Yang, Y., Bo, X., He, S., and Peng, S. (2024). A hierarchical attention network integrating multi-scale relationship for drug response prediction. Information Fusion, 110:102485.

Weisfeiler, B. and Leman, A. (1968). A reduction of a graph to a canonical form and an algebra arising during this reduction. Nauchno-Technicheskaya Informatsiya, 2(9):12–16.

Xu, K., Hu, W., Leskovec, J., and Jegelka, S. (2018). How powerful are graph neural networks? arXiv preprint arXiv:1810.00826.

Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., and Sun, M. (2020). Graph neural networks: A review of methods and applications. AI Open, 1:57–81.