GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation
Chemical Research in Toxicology · 20.11.2023 · 36 (11) · 1717-1730
DOI: 10.1021/acs.chemrestox.3c00199 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chemical Research in Toxicology, online erschienen am 15.10.2023. Verfasst von Wenxuan Wu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai 201203, China. Hauptthemen laut MEDLINE: Chemical and Drug Induced Liver Injury. Die Arbeit wurde bisher 31-mal zitiert (Crossref) und nennt 103 Literaturangaben. Gefördert von: Natural Science Foundation of Hunan Province, National Natural Science Foundation of China.
Abstract (Originalsprache)
Drug-induced liver injury (DILI) is a significant cause of drug failure and withdrawal due to liver damage. Accurate prediction of hepatotoxic compounds is crucial for safe drug development. Several DILI prediction models have been published, but they are built on different data sets, making it difficult to compare model performance. Moreover, most existing models are based on molecular fingerprints or descriptors, neglecting molecular geometric properties and lacking interpretability. To address these limitations, we developed GeoDILI, an interpretable graph neural network that uses a molecular geometric representation. First, we utilized a geometry-based pretrained molecular representation and optimized it on the DILI data set to improve predictive performance. Second, we leveraged gradient information to obtain high-precision atomic-level weights and deduce the dominant substructure. We benchmarked GeoDILI against recently published DILI prediction models, as well as popular GNN models and fingerprint-based machine learning models using the same data set, showing superior predictive performance of our proposed model. We applied the interpretable method in the DILI data set and derived seven precise and mechanistically elucidated structural alerts. Overall, GeoDILI provides a promising approach for accurate and interpretable DILI prediction with potential applications in drug discovery and safety assessment. The data and source code are available at GitHub repository (https://github.com/CSU-QJY/GeoDILI).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai 201203, China.
- School of Mathematics and Statistics, Central South University, Changsha, Hunan 410083, China.
ORCID-Profile: Wenxuan Wu, Jiayu Qian, Guangbo Ge, Qingping Zhou, Xiaoqing Guan
Förderung
- Natural Science Foundation of Hunan Province
- National Natural Science Foundation of China
Zitieren
APA
Wu, W., Qian, J., Liang, C., Yang, J., Ge, G., Zhou, Q., & Guan, X. (2023). GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation. Chemical Research in Toxicology, 36(11), 1717-1730. https://doi.org/10.1021/acs.chemrestox.3c00199
Vancouver
Wu W, Qian J, Liang C, Yang J, Ge G, Zhou Q, et al. GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation. Chemical Research in Toxicology. 2023;36(11):1717-1730. doi:10.1021/acs.chemrestox.3c00199
BibTeX
@article{Wu2023,
author = {Wu, Wenxuan and Qian, Jiayu and Liang, Changjie and Yang, Jingya and Ge, Guangbo and Zhou, Qingping and Guan, Xiaoqing},
title = {GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation},
journal = {Chemical Research in Toxicology},
year = {2023},
volume = {36},
number = {11},
pages = {1717--1730},
doi = {10.1021/acs.chemrestox.3c00199}
}
EUVIMED