Graph Neural Network Model for Predicting Mitochondrial Toxicity via Complementary Molecular Features
Online First: 22/08/2026
Corressponding author's email:
datnp@hcmute.edu.vnDOI:
https://doi.org/10.54644/jte.2026.1954Keywords:
Toxicity prediction, Machine learning, Deep learning, Graph neural network, Mitochondrial toxicityAbstract
In this study, we introduce a hybrid Graph Neural Network model for predicting mitochondrial toxicity, a significant contributor to drug-induced liver injury. This model effectively captures both the structural and physicochemical properties of compounds by integrating graph embeddings and molecular fingerprints. After systematically optimizing the model architecture, we found that the combination of the GATv2Conv convolutional layer, 64 hidden neurons, and a single predictor layer was the most effective configuration for our purposes. When benchmarked against well-established 3D GNN models like DimeNet++ and ViSNet, our hybrid model consistently outperformed them, exhibiting improvements in AUC-ROC by up to 18.3% and 6.2%, respectively. These results emphasize the importance of achieving an optimal balance when integrating features and demonstrate the potential of our hybrid GNN model for computational toxicity prediction. This model could notably expedite drug discovery, reduce dependence on animal testing, and provide valuable insights into molecular toxicity.
Downloads: 0
References
J. A. Dykens and Y. Will, "The significance of mitochondrial toxicity testing in drug development," Drug Discovery Today, vol. 12, no. 17, pp. 777–785, 2007.
K. Chan, T. Don, S. Nandita, and P. J. and O’Brien, "Drug-induced mitochondrial toxicity," Expert Opinion on Drug Metabolism & Toxicology, vol. 1, no. 4, pp. 655–669, 2005.
Y. Will and J. and Dykens, "Mitochondrial toxicity assessment in industry – a decade of technology development and insight," Expert Opinion on Drug Metabolism & Toxicology, vol. 10, no. 8, pp. 1061–1067, 2014.
S. Nadanaciva and Y. Will, "New insights in drug-induced mitochondrial toxicity," Current Pharmaceutical Design, vol. 17, no. 20, pp. 2100–2112, 2011.
M. Porceddu, N. Buron, P. Rustin, B. Fromenty, and A. Borgne-Sanchez, "In vitro assessment of mitochondrial toxicity to predict drug-induced liver injury," Drug-induced liver toxicity, pp. 283–300, 2018.
C. Bai, L. Wu, R. Li, Y. Cao, S. He, and X. Bo, "Machine learning-enabled drug-induced toxicity prediction," Advanced Science, vol. 12, no. 16, p. 2413405, 2025.
H. Zhang, Q. Y. Chen, M. L. Xiang, C. Y. Ma, Q. Huang, and S. Y. Yang, "In silico prediction of mitochondrial toxicity by using GA-CG-SVM approach," Toxicology in Vitro, vol. 23, no. 1, pp. 134–140, 2009.
M. D. Nelms, C. L. Mellor, M. T. D. Cronin, J. C. Madden, and S. J. Enoch, "Development of an in silico profiler for mitochondrial toxicity," Chemical Research in Toxicology, vol. 28, no. 10, pp. 1891–1902, 2015.
H. Zhang et al., "Development of novel prediction model for drug-induced mitochondrial toxicity by using naïve Bayes classifier method," Food and Chemical Toxicology, vol. 110, pp. 122–129, 2017.
D. A. Dreier, N. D. Denslow, and C. J. Martyniuk, "Computational in vitro toxicology uncovers chemical structures impairing mitochondrial membrane potential," Journal of Chemical Information and Modeling, vol. 59, no. 2, pp. 702–712, 2019.
J. Hemmerich, F. Troger, B. Füzi, and G. F.Ecker, "Using machine learning methods and structural alerts for prediction of mitochondrial toxicity," Molecular Informatics, vol. 39, no. 5, p. 2000005, 2020.
W. Tang, J. Chen, and H. Hong, "Discriminant models on mitochondrial toxicity improved by consensus modeling and resolving imbalance in training," Chemosphere, vol. 253, p. 126768, 2020.
F. Bringezu, J. Carlos Gómez-Tamayo, and M. Pastor, "Ensemble prediction of mitochondrial toxicity using machine learning technology," Computational Toxicology, vol. 20, p. 100189, 2021.
P. Zhao et al., "In silico prediction of mitochondrial toxicity of chemicals using machine learning methods," Journal of Applied Toxicology, vol. 41, no. 10, pp. 1518–1526, 2021.
K. Jaganathan, M. U. Rehman, H. Tayara, and K. T. Chong, "XML-CIMT: Explainable machine learning (XML) model for predicting chemical-induced mitochondrial toxicity," International Journal of Molecular Sciences, vol. 23, no. 24, p. 15655, 2022.
R. Zhang, Z. Chen, B. Wang, Y. Li, Y. Mu, and X. Li, "Modeling and insights into the structural characteristics of chemical mitochondrial toxicity," ACS Omega, vol. 8, no. 35, pp. 31675–31682, 2023.
M. G. De Lomana, P. M. Zapata, and F. Montanari, "Predicting the mitochondrial toxicity of small molecules: insights from mechanistic assays and cell painting data," Chemical Research in Toxicology, vol. 36, pp. 1107–1120, 2023-07-06 2023.
Y. Igarashi, R. Kojima, S. Matsumoto, H. Iwata, Y. Okuno, and H. Yamada, "Developing a GNN-based AI model to predict mitochondrial toxicity using the bagging method," The Journal of Toxicological Sciences, vol. 49, no. 3, pp. 117–126, 2024.
Y. Zhou et al., "ToxMPNN: A deep learning model for small molecule toxicity prediction," Journal of Applied Toxicology, vol. 44, no. 7, pp. 953–964, 2024.
Y. T. Lin, K. H. Lin, C. J. Huang, and A. C. Wei, "MitoTox: a comprehensive mitochondrial toxicity database," BMC Bioinformatics, vol. 22, no. 10, p. 369, 2021.
B. Ramsundar, P. Eastman, P. Walters, and V. Pande, Deep learning for the life sciences: applying deep learning to genomics, microscopy, drug discovery, and more. " O'Reilly Media, Inc.", 2019.
J. Gasteiger, S. Giri, J. T. Margraf, and S. Günnemann, "Fast and uncertainty-aware directional message passing for non-equilibrium molecules," arXiv preprint arXiv:2011.14115, 2020.
Y. Wang et al., "Enhancing geometric representations for molecules with equivariant vector-scalar interactive message passing," Nature Communications, vol. 15, no. 1, p. 313, 2024.
Downloads
Published
How to Cite
License
Copyright (c) 2026 Journal of Technical Education Science
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Copyright © JTE.
