Enhancing Drug-Drug Interaction Prediction with Explainable AI: Integrating GANs for Improved Clinical Transparency
DOI:
https://doi.org/10.22399/ijcesen.2551Keywords:
Drug-Drug Interactions (DDIs), Artificial Intelligence (AI), Explainable AI (XAI), Generative Adversarial Networks (GANs), XGBoost, SHapley Additive exPlanations (SHAP)Abstract
Drug-drug interactions (DDIs) are critical in polypharmacy, where the concurrent use of multiple drugs can lead to synergistic effects or adverse drug events (ADEs). The latter can significantly impact patient morbidity and mortality. The rapid introduction of new drugs further complicates the prediction of DDIs, making traditional wet-lab verification methods both time-consuming and resource-intensive. While artificial intelligence (AI) models have been employed to predict DDIs, the development of highly complex "black-box" models poses challenges in terms of interpretability and trust in clinical settings. There is a pressing need for explainable AI (XAI) approaches to ensure these models are both accurate and transparent.This study utilizes a comprehensive dataset from DrugBank, encompass- ing various drug interactions. We implemented data preprocessing steps, including handling missing values, encoding categorical variables, and normalizing the data. To address data scarcity, we employed Generative Adversarial Networks (GANs) to generate synthetic data, which was combined with real data to enhance the training dataset. The augmented dataset was then used to train an XGBoost model, optimized for binary classification. To ensure interpretability, we integrated SHapley Additive exPlanations (SHAP) to analyze feature im- portance and model decision-making processes. The XGBoost model demonstrated high predictive accuracy with a validation accuracy of 99.06%, precision of 98.73%, recall of 99.03%, and an F1 score of 98.88%. SHAP analysis provided clear insights into feature importance, highlighting the most influential fea- tures in the model’s predictions and enhancing the transparency of the decision-making pro- cess. The combination of advanced machine learning techniques and explainable AI methods effectively addresses the challenges of DDI prediction. The proposed approach not only achieves high predictive performance but also ensures model interpretability, fostering trust and adoption in clinical applications. This methodology offers significant potential for improving patient safety and treatment outcomes.
References
[1] Askari, M., et al. (2013). Frequency and nature of drug-drug interactions in the intensive care unit. Pharmacoepidemiology and Drug Safety. 22(4);430-437. https://doi.org/10.1002/pds.3415 DOI: https://doi.org/10.1002/pds.3415
[2] Raschetti, R., et al. (1999). Suspected adverse drug events requiring emergency department visits or hospital admissions. European Journal of Clinical Pharmacology. 54(12);959-963. https://doi.org/10.1007/s002280050582 DOI: https://doi.org/10.1007/s002280050582
[3] Budnitz, D.S., et al. (2006). National surveillance of emergency department visits for outpatient adverse drug events. JAMA. 296(15);1858-1866. https://doi.org/10.1001/jama.296.15.1858 DOI: https://doi.org/10.1001/jama.296.15.1858
[4] Reis, A.M., & Cassiani, S.H. (2010). Evaluation of three brands of drug interaction software for use in intensive care units. Pharmacy World & Science. 32(6);822-828. https://doi.org/10.1007/s11096-010-9445-2 DOI: https://doi.org/10.1007/s11096-010-9445-2
[5] Vonbach, P., et al. (2008). Evaluation of frequently used drug interaction screening programs. Pharmacy World & Science. 30(4);367-374. https://doi.org/10.1007/s11096-008-9191-x DOI: https://doi.org/10.1007/s11096-008-9191-x
[6] Cheng, F., & Zhao, Z. (2014). Machine learning-based prediction of drug–drug interactions by integrating drug phenotypic, therapeutic, chemical, and genomic properties. Journal of the American Medical Informatics Association. 21(e2);e278-e286. https://doi.org/10.1136/amiajnl-2013-002512 DOI: https://doi.org/10.1136/amiajnl-2013-002512
[7] Ryu, J.Y., Kim, H.U., & Lee, S.Y. (2018). Deep learning improves prediction of drug–drug and drug–food interactions. Proceedings of the National Academy of Sciences. 115(18);E4304. https://doi.org/10.1073/pnas.1803294115 DOI: https://doi.org/10.1073/pnas.1803294115
[8] Vilar, S., et al. (2014). Similarity-based modeling in large-scale prediction of drug-drug interactions. Nature Protocols. 9(9);2147-2163. https://doi.org/10.1038/nprot.2014.151 DOI: https://doi.org/10.1038/nprot.2014.151
[9] Vilar, S., Uriarte, E., Santana, L., Tatonetti, N.P., & Friedman, C. (2013). Detection of drug-drug interactions by modeling interaction profile fingerprints. PLoS ONE. 8(3);e58321. https://doi.org/10.1371/journal.pone.0058321 DOI: https://doi.org/10.1371/journal.pone.0058321
[10] Gunning, D., et al. (2019). XAI—explainable artificial intelligence. Science Robotics. 4(37);eaay7120. https://doi.org/10.1126/scirobotics.aay7120 DOI: https://doi.org/10.1126/scirobotics.aay7120
[11] Hailu, N., Hunter, L., & Cohen, K.B. (2013). Ucolorado_som: extraction of drug-drug interactions from biomedical text using knowledge-rich and knowledge-poor features. Second Joint Conference on Lexical and Computational Semantics (SEM), Volume 2: Proceedings of the Seventh International Workshop on Semantic Evaluation (SemEval 2013). https://aclanthology.org/S13-2112/
[12] Hunta, S., Aunsri, N., & Yooyativong, T. (2017). Integrated action crossing method for drug-drug interactions prediction in noncommunicable diseases based on neural networks. 2017 International Conference on Digital Arts, Media and Technology (ICDAMT). https://doi.org/10.1109/icdamt.2017.7904973 DOI: https://doi.org/10.1109/ICDAMT.2017.7904973
[13] Song, D., et al. (2019). Similarity-based machine learning support vector machine predictor of drug-drug interactions with improved accuracies. Journal of Clinical Pharmacy and Therapeutics. 44(2);268-275. https://doi.org/10.1111/jcpt.12786 DOI: https://doi.org/10.1111/jcpt.12786
[14] Wang, H., et al. (2021). Gognn: graph of graphs neural network for predicting structured entity interactions. Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence. Article 183. https://doi.org/10.24963/ijcai.2020/183 DOI: https://doi.org/10.24963/ijcai.2020/183
[15] Minard, A.-L., et al. (2011). Feature selection for drug-drug interaction detection using machine-learning based approaches. Challenge Task on Drug-Drug Interaction Extraction (DDI) SEPLN. 43–50. https://hal.science/hal-02289969v1
[16] Boyce, R., Gardner, G., & Harkema, H. (2012). Using natural language processing to identify pharmacokinetic drug-drug interactions described in drug package inserts. Proceedings of the 2012 Workshop on Biomedical Natural Language Processing. 206-213.
[17] Thomas, P., et al. (2011). Relation extraction for drug-drug interactions using ensemble learning. 1st Challenge task on Drug-Drug Interaction Extraction (DDIExtraction 2011). 11-18. https://ceur-ws.org/Vol-761/paper1.pdf
[18] Bokharaeian, B., Diaz, A., & Chitsaz, H. (2016). Enhancing extraction of drug-drug interaction from literature using neutral candidates, negation, and clause dependency. PLoS ONE. 11(10);e0163480. https://doi.org/10.1371/journal.pone.0163480 DOI: https://doi.org/10.1371/journal.pone.0163480
[19] Dhami, D.S., et al. (2018). Drug-drug interaction discovery: kernel learning from heterogeneous similarities. Smart Health. 9;88-100. https://doi.org/10.1016/j.smhl.2018.07.007 DOI: https://doi.org/10.1016/j.smhl.2018.07.007
[20] Zhang, Y., et al. (2012). A single kernel-based approach to extract drug-drug interactions from biomedical literature. PLoS ONE. 7(11);e48901. https://doi.org/10.1371/journal.pone.0048901 DOI: https://doi.org/10.1371/journal.pone.0048901
[21] Zhang, P., Wang, F., Hu, J., & Sorrentino, R. (2015). Label propagation prediction of drug-drug interactions based on clinical side effects. Scientific Reports. 5;12339. https://doi.org/10.1038/srep12339 DOI: https://doi.org/10.1038/srep12339
[22] Xie, W., et al. (2021). Integrated random negative sampling and uncertainty sampling in active learning improve clinical drug safety drug-drug interaction information retrieval. Frontiers in Pharmacology. 11;2225. https://doi.org/10.3389/fphar.2020.582470 DOI: https://doi.org/10.3389/fphar.2020.582470
[23] Yan, S., Jiang, X., & Chen, Y. (2013). Text mining driven drug-drug interaction detection. Proceedings. IEEE International Conference on Bioinformatics and Biomedicine. 349-355. https://doi.org/10.1109/bibm.2013.6732517 DOI: https://doi.org/10.1109/BIBM.2013.6732517
[24] Zhan, C., et al. (2020). Detecting high-quality signals of adverse drug-drug interactions from spontaneous reporting data. Journal of Biomedical Informatics. 112;103603. https://doi.org/10.1016/j.jbi.2020.103603 DOI: https://doi.org/10.1016/j.jbi.2020.103603
[25] Qian, S., Liang, S., & Yu, H. (2019). Leveraging genetic interactions for adverse drug-drug interaction prediction. PLoS Computational Biology. 15(5);e1007068. https://doi.org/10.1371/journal.pcbi.1007068 DOI: https://doi.org/10.1371/journal.pcbi.1007068
[26] Dang, L.H., et al. (2021). Machine learning-based prediction of drug-drug interactions for histamine antagonist using hybrid chemical features. Cells. 10(11);3092. https://doi.org/10.3390/cells10113092 DOI: https://doi.org/10.3390/cells10113092
[27] Park, C., Park, J., & Park, S. (2020). Agcn: Attention-based graph convolutional networks for drug-drug interaction extraction. Expert Systems with Applications. 159;113538. https://doi.org/10.1016/j.eswa.2020.113538 DOI: https://doi.org/10.1016/j.eswa.2020.113538
[28] Zhang, Y., et al. (2017). Drug–drug interaction extraction via hierarchical rnns on sequence and shortest dependency paths. Bioinformatics. 34(5);828-835. https://doi.org/10.1093/bioinformatics/btx659 DOI: https://doi.org/10.1093/bioinformatics/btx659
[29] Mahadevan, A.A., et al. (2019). A predictive model for drug-drug interaction using a similarity measure. 2019 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB). https://doi.org/10.1109/cibcb.2019.8791458 DOI: https://doi.org/10.1109/CIBCB.2019.8791458
[30] Patrick, M.T., et al. (2021). Advancement in predicting interactions between drugs used to treat psoriasis and its comorbidities by integrating molecular and clinical resources. Journal of the American Medical Informatics Association. 28(6);1159-1167. https://doi.org/10.1093/jamia/ocaa335 DOI: https://doi.org/10.1093/jamia/ocaa335
[31] Zhang, Y., & Lu, Z. (2019). Exploring semi-supervised variational autoencoders for biomedical relation extraction. Methods. 166. https://doi.org/10.1016/j.ymeth.2019.02.021 DOI: https://doi.org/10.1016/j.ymeth.2019.02.021
[32] Hung, T.N.K., et al. (2022). An ai-based prediction model for drug-drug interactions in osteoporosis and paget's diseases from smiles. Molecular Informatics. 2100264. https://doi.org/10.1002/minf.202100264 DOI: https://doi.org/10.1002/minf.202100264
[33] Bobic, T., Fluck, J., & Hofmann, M. (2013). Scai: Extracting drug-drug interactions using a rich feature vector. Second Joint Conference on Lexical and Computational Semantics (SEM), Volume 2: Proceedings of the Seventh International Workshop on Semantic Evaluation (SemEval 2013). https://aclanthology.org/S13-2111/
[34] Zhang, Y., et al. (2020). Predicting drug-drug interactions using multi-modal deep auto-encoders based network embedding and positive-unlabeled learning. Methods. 179;37-46. https://doi.org/10.1016/j.ymeth.2020.05.007 DOI: https://doi.org/10.1016/j.ymeth.2020.05.007
[35] Zhang, W., et al. (2019). Sflln: A sparse feature learning ensemble method with linear neighborhood regularization for predicting drug–drug interactions. Information Sciences. 497;189-201. https://doi.org/10.1016/j.ins.2019.05.017 DOI: https://doi.org/10.1016/j.ins.2019.05.017
[36] Li, D., et al. (2016). A topic-modeling based framework for drug-drug interaction classification from biomedical text. AMIA Annual Symposium Proceedings. 789-798. https://pmc.ncbi.nlm.nih.gov/articles/PMC5333320/
[37] Shukla, P.K., et al. (2020). Efficient prediction of drug-drug interaction using deep learning models. IET Systems Biology. 14(4);211-216. https://doi.org/10.1049/iet-syb.2019.0116 DOI: https://doi.org/10.1049/iet-syb.2019.0116
[38] Sejnowski, T.J. (2020). The unreasonable effectiveness of deep learning in artificial intelligence. Proceedings of the National Academy of Sciences. 117(48);30033-30038. https://doi.org/10.1073/pnas.1907373117 DOI: https://doi.org/10.1073/pnas.1907373117
[39] Hou, W.J., & Ceesay, B. (2018). Extraction of drug-drug interaction using neural embedding. Journal of Bioinformatics and Computational Biology. 16(6);1840027. https://doi.org/10.1142/s0219720018400279 DOI: https://doi.org/10.1142/S0219720018400279
[40] Shtar, G., Rokach, L., & Shapira, B. (2019). Detecting drug-drug interactions using artificial neural networks and classic graph similarity measures. PLoS ONE. 14(8);e0219796. https://doi.org/10.1371/journal.pone.0219796 DOI: https://doi.org/10.1371/journal.pone.0219796
[41] Rohani, N., & Eslahchi, C. (2019). Drug-drug interaction predicting by neural network using integrated similarity. Scientific Reports. 9(1);13645. https://doi.org/10.1038/s41598-019-50121-3 DOI: https://doi.org/10.1038/s41598-019-50121-3
[42] Masumshah, R., Aghdam, R., & Eslahchi, C. (2021). A neural network-based method for polypharmacy side effects prediction. BMC Bioinformatics. 22(1);385. https://doi.org/10.1186/s12859-021-04298-y DOI: https://doi.org/10.1186/s12859-021-04298-y
[43] Collobert, R., et al. (2011). Natural language processing (almost) from scratch. Journal of Machine Learning Research. 12;2493-2537. https://doi.org/10.48550/arXiv.1103.0398
[44] Sutskever, I., Vinyals, O., & Le, Q.V. (2014). Sequence to sequence learning with neural networks. Proceedings of the 27th International Conference on Neural Information Processing Systems - Volume 2. 3104-3112. https://doi.org/10.48550/arXiv.1409.3215
[45] Zhang, S., et al. (2015). Bidirectional long short-term memory networks for relation classification. PACLIC. https://aclanthology.org/Y15-1009/
[46] Sahu, S.K., & Anand, A. (2018). Drug-drug interaction extraction from biomedical texts using long short-term memory network. Journal of Biomedical Informatics. 86;15-24. https://doi.org/10.1016/j.jbi.2018.08.005 DOI: https://doi.org/10.1016/j.jbi.2018.08.005
[47] Wang, W., et al. (2017). Dependency-based long short term memory network for drug-drug interaction extraction. BMC Bioinformatics. 18(16);578. https://doi.org/10.1186/s12859-017-1962-8 DOI: https://doi.org/10.1186/s12859-017-1962-8
[48] Kim, Y., et al. (2016). Character-aware neural language models. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. 2741-2749. https://doi.org/10.1609/aaai.v30i1.10362 DOI: https://doi.org/10.1609/aaai.v30i1.10362
[49] Zhang, X., Zhao, J., & LeCun, Y. (2015). Character-level convolutional networks for text classification. Proceedings of the 28th International Conference on Neural Information Processing Systems - Volume 1. 649-657. https://doi.org/10.48550/arXiv.1509.01626
[50] Kavuluru, R., Rios, A., & Tran, T. (2017). Extracting drug-drug interactions with word and character-level recurrent neural networks. IEEE International Conference on Healthcare Informatics. https://doi.org/10.1109/ichi.2017.15 DOI: https://doi.org/10.1109/ICHI.2017.15
[51] Luo, Q., et al. (2021). Novel deep learning-based transcriptome data analysis for drug-drug interaction prediction with an application in diabetes. BMC Bioinformatics. 22(1);318. https://doi.org/10.1186/s12859-021-04241-1 DOI: https://doi.org/10.1186/s12859-021-04241-1
[52] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation. 9(8);1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735 DOI: https://doi.org/10.1162/neco.1997.9.8.1735
[53] Cho, K., et al. (2014). On the properties of neural machine translation: Encoder–decoder approaches. Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation. https://doi.org/10.3115/v1/w14-4012 DOI: https://doi.org/10.3115/v1/W14-4012
[54] Zhou, D., Miao, L., & He, Y. (2018). Position-aware deep multi-task learning for drug-drug interaction extraction. Artificial Intelligence in Medicine. 87;1-8. https://doi.org/10.1016/j.artmed.2018.03.001 DOI: https://doi.org/10.1016/j.artmed.2018.03.001
[55] Jiang, Z., Gu, L., & Jiang, Q. (2017). Drug drug interaction extraction from literature using a skeleton long short term memory neural network. 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm.2017.8217708 DOI: https://doi.org/10.1109/BIBM.2017.8217708
[56] Zaikis, D., & Vlahavas, I. (2020). Drug-drug interaction classification using attention based neural networks. 11th Hellenic Conference on Artificial Intelligence. 34-40. https://doi.org/10.1145/3411408.3411461 DOI: https://doi.org/10.1145/3411408.3411461
[57] Yi, Z., et al. (2017). Drug-drug interaction extraction via recurrent neural network with multiple attention layers. International Conference on Advanced Data Mining and Applications. 448-462. https://doi.org/10.1007/978-3-319-69179-4_39 DOI: https://doi.org/10.1007/978-3-319-69179-4_39
[58] Zheng, W., et al. (2017). An attention-based effective neural model for drug-drug interactions extraction. BMC Bioinformatics. 18(1);445. https://doi.org/10.1186/s12859-017-1855-x DOI: https://doi.org/10.1186/s12859-017-1855-x
[59] Adadi, A., & Berrada, M. (2018). Peeking inside the black-box: A survey on explainable artificial intelligence (xai). IEEE Access. 2169-3536. https://doi.org/10.1109/access.2018.2870052 DOI: https://doi.org/10.1109/ACCESS.2018.2870052
[60] Guidotti, R., et al. (2018). A survey of methods for explaining black box models. ACM Computing Surveys. 51(5), 93. https://doi.org/10.1145/3236009 DOI: https://doi.org/10.1145/3236009
[61] Simonyan, K., Vedaldi, A., & Zisserman, A. (2014). Deep inside convolutional networks: Visualising image classification models and saliency maps. CoRR. abs/1312.6034.
[62] Liu, S., et al. (2016). Dependency-based convolutional neural network for drug-drug interaction extraction. 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm.2016.7822671 DOI: https://doi.org/10.1109/BIBM.2016.7822671
[63] Sun, X., et al. (2018). Deep convolution neural networks for drug-drug interaction extraction. 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm.2018.8621405 DOI: https://doi.org/10.1109/BIBM.2018.8621405
[64] Shrikumar, A., Greenside, P., & Kundaje, A. (2017). Learning important features through propagating activation differences. Proceedings of the 34th International Conference on Machine Learning - Volume 70. 3145-3153. https://doi.org/10.1109/tnnls.2022.3228102 DOI: https://doi.org/10.1109/TNNLS.2022.3228102
[65] Zeng, T., et al. (2015). Deep convolutional neural networks for annotating gene expression patterns in the mouse brain. BMC Bioinformatics. 16(1);147. https://doi.org/10.1186/s12859-015-0553-9 DOI: https://doi.org/10.1186/s12859-015-0553-9
[66] Springenberg, J.T., et al. (2015). Striving for simplicity: The all convolutional net. CoRR. abs/1412.6806.
[67] Lei, T., Barzilay, R., & Jaakkola, T. (2016). Rationalizing neural predictions. 2016 Conference on Empirical Methods in Natural Language Processing. 2143-2152. https://doi.org/10.18653/v1/d16-1011 DOI: https://doi.org/10.18653/v1/D16-1011
[68] Quan, C., et al. (2016). Multichannel convolutional neural network for biological relation extraction. BioMed Research International. 2016;1850404. https://doi.org/10.1155/2016/1850404 DOI: https://doi.org/10.1155/2016/1850404
[69] Xiong, W., et al. (2019). Extracting drug-drug interactions with a dependency-based graph convolution neural network. 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm47256.2019.8983150 DOI: https://doi.org/10.1109/BIBM47256.2019.8983150
[70] Bolukbasi, T., et al. (2016). Man is to computer programmer as woman is to homemaker? debiasing word embeddings. Advances in Neural Information Processing Systems. 29. https://doi.org/10.48550/arXiv.1607.06520
[71] Zügner, D., Akbarnejad, A., & Günnemann, S. (2018). Adversarial attacks on neural networks for graph data. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery Data Mining. 2847-2856. https://doi.org/10.1145/3219819.3220078 DOI: https://doi.org/10.1145/3219819.3220078
[72] Sun, M., Wang, F., Elemento, O., & Zhou, J. (2020). Structure-based drug-drug interaction detection via expressive graph convolutional networks and deep sets. Proceedings of the AAAI Conference on Artificial Intelligence. https://doi.org/10.1609/aaai.v34i10.7236 DOI: https://doi.org/10.1609/aaai.v34i10.7236
[73] Lin, Z.Q.X., et al. (2020). Knowledge graph neural network for drug-drug interaction prediction. Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence. https://doi.org/10.24963/ijcai.2020/380 DOI: https://doi.org/10.24963/ijcai.2020/380
[74] Miyato, T., Dai, A.M., & Goodfellow, I. (2016). Adversarial training methods for semi-supervised text classification. arXiv preprint arXiv:1605.07725.
[75] Feng, S., et al. (2018). Pathologies of neural models make interpretations difficult. arXiv preprint. https://doi.org/10.48550/arXiv.1605.07725 DOI: https://doi.org/10.18653/v1/D18-1407
[76] Schwarz, K., et al. (2021). Attentionddi: Siamese attention-based deep learning method for drug–drug interaction predictions. BMC Bioinformatics. 22(1);412. https://doi.org/10.1186/s12859-021-04325-y DOI: https://doi.org/10.1186/s12859-021-04325-y
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering

This work is licensed under a Creative Commons Attribution 4.0 International License.