{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T08:52:14Z","timestamp":1764751934554,"version":"3.46.0"},"reference-count":38,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T00:00:00Z","timestamp":1764720000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Drug-induced liver injury (DILI) is a leading cause of drug development failures and post-market withdrawals, presenting significant challenges to both pharmaceutical safety and clinical care. Conventional screening methods, such as laboratory-based cell assays and animal models, are limited in their ability to accurately predict human hepatotoxicity. To address this, we propose a robust graph-based deep learning model that combines graph attention networks (GAT) and graph convolutional networks (GCN) for the prediction of DILI-associated compounds. Molecular structures are represented as graphs, where atoms and bonds are encoded with rich chemical and topological features derived from molecular formulas. The model effectively learns complex interaction patterns within each compound and demonstrates superior performance over existing machine learning and deep learning baselines. It achieves an area under the receiver operating characteristic curve (ROC-AUC) of 0.8481, an area under the precision\u2013recall curve (PR-AUC) of 0.9008, and an F1-score of 0.8017. Furthermore, evaluation across twenty independent trials confirms strong reproducibility. These findings highlight the model\u2019s reliability and potential as a practical tool for early-stage hepatotoxicity screening in drug discovery.<\/jats:p>","DOI":"10.7717\/peerj-cs.3416","type":"journal-article","created":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T08:45:29Z","timestamp":1764751529000},"page":"e3416","source":"Crossref","is-referenced-by-count":0,"title":["A robust graph-based computational model for predicting drug-induced liver injury compounds"],"prefix":"10.7717","volume":"11","author":[{"given":"Qi","family":"Yao","sequence":"first","affiliation":[]},{"given":"Xingming","family":"Ji","sequence":"additional","affiliation":[]},{"given":"Congwei","family":"Liu","sequence":"additional","affiliation":[]},{"given":"Ye","family":"Wang","sequence":"additional","affiliation":[]}],"member":"4443","published-online":{"date-parts":[[2025,12,3]]},"reference":[{"issue":"2","key":"10.7717\/peerj-cs.3416\/ref-1","doi-asserted-by":"publisher","first-page":"e13340","DOI":"10.7759\/cureus.13340","article-title":"Acute cholestatic liver injury due to ciprofloxacin in a young healthy adult","volume":"13","author":"Ahmad","year":"2021","journal-title":"Cureus"},{"issue":"1","key":"10.7717\/peerj-cs.3416\/ref-2","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1093\/toxsci\/kfy121","article-title":"Predicting drug-induced liver injury using ensemble learning methods and molecular fingerprints","volume":"165","author":"Ai","year":"2018","journal-title":"Toxicological Sciences"},{"issue":"3","key":"10.7717\/peerj-cs.3416\/ref-3","doi-asserted-by":"publisher","first-page":"135","DOI":"10.5582\/ddt.2020.03029","article-title":"Malnutrition as an important risk factor for drug-induced liver injury in patients on anti-tubercular therapy: an experience from a tertiary care center in south India","volume":"14","author":"Ali","year":"2020","journal-title":"Drug Discoveries & Therapeutics"},{"issue":"1","key":"10.7717\/peerj-cs.3416\/ref-4","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1038\/s41572-019-0105-0","article-title":"Drug-induced liver injury","volume":"5","author":"Andrade","year":"2019","journal-title":"Nature Reviews Disease Primers"},{"issue":"7","key":"10.7717\/peerj-cs.3416\/ref-5","doi-asserted-by":"publisher","first-page":"940","DOI":"10.1111\/apt.15168","article-title":"A prospective study of the incidence of drug-induced liver injury by the modern volatile anaesthetics sevoflurane and desflurane","volume":"49","author":"Bishop","year":"2019","journal-title":"Alimentary Pharmacology & Therapeutics"},{"issue":"1","key":"10.7717\/peerj-cs.3416\/ref-6","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1111\/j.1365-2036.2010.04320.x","article-title":"Review article: drug-induced liver injury in clinical practice","volume":"32","author":"Bj\u00f6rnsson","year":"2010","journal-title":"Alimentary Pharmacology & Therapeutics"},{"issue":"2","key":"10.7717\/peerj-cs.3416\/ref-7","doi-asserted-by":"publisher","first-page":"51","DOI":"10.1383\/anes.2006.7.2.51","article-title":"Liver: metabolic functions","volume":"7","author":"Campbell","year":"2006","journal-title":"Anaesthesia & Intensive Care Medicine"},{"issue":"14","key":"10.7717\/peerj-cs.3416\/ref-8","doi-asserted-by":"publisher","first-page":"7693","DOI":"10.1021\/acs.analchem.7b01671","article-title":"Investigation of drug-induced hepatotoxicity and its remediation pathway with reaction-based fluorescent probes","volume":"89","author":"Cheng","year":"2017","journal-title":"Analytical Chemistry"},{"issue":"12","key":"10.7717\/peerj-cs.3416\/ref-9","doi-asserted-by":"publisher","first-page":"2302","DOI":"10.1124\/dmd.110.035113","article-title":"A predictive ligand-based Bayesian model for human drug-induced liver injury","volume":"38","author":"Ekins","year":"2010","journal-title":"Drug Metabolism and Disposition"},{"issue":"10","key":"10.7717\/peerj-cs.3416\/ref-10","doi-asserted-by":"publisher","first-page":"1450","DOI":"10.1038\/ajg.2015.283","article-title":"Persistent liver biochemistry abnormalities are more common in older patients and those with cholestatic drug induced liver injury","volume":"110","author":"Fontana","year":"2015","journal-title":"American Journal of Gastroenterology"},{"issue":"7","key":"10.7717\/peerj-cs.3416\/ref-11","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1021\/ci100176x","article-title":"Trust, but verify: on the importance of chemical structure curation in cheminformatics and QSAR modeling research","volume":"50","author":"Fourches","year":"2010","journal-title":"Journal of Chemical Information and Modeling"},{"issue":"7","key":"10.7717\/peerj-cs.3416\/ref-12","doi-asserted-by":"publisher","first-page":"610","DOI":"10.2174\/1574893618666230504143647","article-title":"Predicting herb-disease associations through graph convolutional network","volume":"18","author":"Hu","year":"2023","journal-title":"Current Bioinformatics"},{"issue":"11","key":"10.7717\/peerj-cs.3416\/ref-13","doi-asserted-by":"publisher","first-page":"1925","DOI":"10.1021\/acs.chemrestox.2c00269","article-title":"Developing role for artificial intelligence in drug discovery in drug design, development, and safety assessment","volume":"35","author":"Hurben","year":"2022","journal-title":"Chemical Research in Toxicology"},{"issue":"Suppl. 2","key":"10.7717\/peerj-cs.3416\/ref-14","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1016\/j.tox.2017.06.003","article-title":"Predicting drug-induced liver injury: the importance of data curation","volume":"389","author":"Kotsampasakou","year":"2017","journal-title":"Toxicology"},{"key":"10.7717\/peerj-cs.3416\/ref-15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-023-00796-8","article-title":"Interdili: interpretable prediction of drug-induced liver injury through permutation feature importance and attention mechanism","volume":"16","author":"Lee","year":"2024","journal-title":"Journal of Cheminformatics"},{"issue":"1","key":"10.7717\/peerj-cs.3416\/ref-16","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1016\/j.mayocp.2013.09.016","article-title":"Drug-induced liver injury","volume":"89","author":"Leise","year":"2014","journal-title":"Mayo Clinic Proceedings"},{"issue":"2","key":"10.7717\/peerj-cs.3416\/ref-17","doi-asserted-by":"publisher","first-page":"550","DOI":"10.1021\/acs.chemrestox.0c00374","article-title":"DeepDILI: deep learning-powered drug-induced liver injury prediction using model-level representation","volume":"34","author":"Li","year":"2020","journal-title":"Chemical Research in Toxicology"},{"issue":"5","key":"10.7717\/peerj-cs.3416\/ref-18","doi-asserted-by":"publisher","first-page":"332","DOI":"10.1111\/1751-2980.13205","article-title":"Mapping the incidence of drug-induced liver injury: a systematic review and meta-analysis","volume":"24","author":"Li","year":"2023","journal-title":"Journal of Digestive Diseases"},{"issue":"1","key":"10.7717\/peerj-cs.3416\/ref-19","doi-asserted-by":"publisher","first-page":"bbad431","DOI":"10.1093\/bib\/bbad431","article-title":"Drug repositioning based on weighted local information augmented graph neural network","volume":"25","author":"Meng","year":"2023","journal-title":"Briefings in Bioinformatics"},{"issue":"7","key":"10.7717\/peerj-cs.3416\/ref-20","doi-asserted-by":"publisher","first-page":"2628","DOI":"10.1021\/acs.molpharmaceut.0c00326","article-title":"Comparing machine learning algorithms for predicting drug-induced liver injury (DILI)","volume":"17","author":"Minerali","year":"2020","journal-title":"Molecular Pharmaceutics"},{"issue":"6","key":"10.7717\/peerj-cs.3416\/ref-21","doi-asserted-by":"publisher","first-page":"385","DOI":"10.3390\/toxics12060385","article-title":"Machine learning to predict drug-induced liver injury and its validation on failed drug candidates in development","volume":"12","author":"Mostafa","year":"2024","journal-title":"Toxics"},{"issue":"2","key":"10.7717\/peerj-cs.3416\/ref-22","doi-asserted-by":"publisher","first-page":"356","DOI":"10.3390\/ijms18020356","article-title":"Cell imaging counting as a novel ex vivo approach for investigating drug-induced hepatotoxicity in zebrafish larvae","volume":"18","author":"Nguyen","year":"2017","journal-title":"International Journal of Molecular Sciences"},{"issue":"24","key":"10.7717\/peerj-cs.3416\/ref-23","doi-asserted-by":"publisher","first-page":"9173","DOI":"10.1021\/acs.jcim.4c01240","article-title":"From SMILES to enhanced molecular property prediction: a unified multimodal framework with predicted 3D conformers and contrastive learning techniques","volume":"64","author":"Nguyen","year":"2024","journal-title":"Journal of Chemical Information and Modeling"},{"issue":"39","key":"10.7717\/peerj-cs.3416\/ref-24","doi-asserted-by":"publisher","first-page":"25432","DOI":"10.1021\/acsomega.0c03866","article-title":"Predicting drug-induced liver injury using convolutional neural network and molecular fingerprint-embedded features","volume":"5","author":"Nguyen-Vo","year":"2020","journal-title":"ACS Omega"},{"issue":"3","key":"10.7717\/peerj-cs.3416\/ref-25","doi-asserted-by":"publisher","first-page":"519","DOI":"10.1007\/s12293-024-00414-6","article-title":"Molecular representations in bio-cheminformatics","volume":"16","author":"Nguyen-Vo","year":"2024","journal-title":"Memetic Computing"},{"key":"10.7717\/peerj-cs.3416\/ref-26","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1186\/s40360-018-0282-6","article-title":"eToxPred: a machine learning-based approach to estimate the toxicity of drug candidates","volume":"20","author":"Pu","year":"2019","journal-title":"BMC Pharmacology and Toxicology"},{"key":"10.7717\/peerj-cs.3416\/ref-27","doi-asserted-by":"publisher","first-page":"116350","DOI":"10.1016\/j.scriptamat.2024.116350","article-title":"Harmonizing physical and deep learning modeling: a computationally efficient and interpretable approach for property prediction","volume":"255","author":"Ren","year":"2025","journal-title":"Scripta Materialia"},{"issue":"6","key":"10.7717\/peerj-cs.3416\/ref-28","doi-asserted-by":"publisher","first-page":"2065","DOI":"10.1002\/hep.23937","article-title":"Drug-induced acute liver failure: results of a U.S. multicenter, prospective study","volume":"52","author":"Reuben","year":"2010","journal-title":"Hepatology"},{"issue":"10","key":"10.7717\/peerj-cs.3416\/ref-29","doi-asserted-by":"publisher","first-page":"2067","DOI":"10.1021\/tx300075j","article-title":"Preclinical strategy to reduce clinical hepatotoxicity using in vitro bioactivation data for 200 compounds","volume":"25","author":"Sakatis","year":"2012","journal-title":"Chemical Research in Toxicology"},{"issue":"5","key":"10.7717\/peerj-cs.3416\/ref-30","doi-asserted-by":"publisher","first-page":"26","DOI":"10.3892\/etm.2020.9154","article-title":"Abnormal metabolic processes involved in the pathogenesis of non-alcoholic fatty liver disease (review)","volume":"20","author":"Shao","year":"2020","journal-title":"Experimental and Therapeutic Medicine"},{"issue":"8","key":"10.7717\/peerj-cs.3416\/ref-31","doi-asserted-by":"publisher","first-page":"2230\u20132241.e11","DOI":"10.1053\/j.gastro.2019.02.002","article-title":"Incidence and etiology of drug-induced liver injury in mainland China","volume":"156","author":"Shen","year":"2019","journal-title":"Gastroenterology"},{"issue":"9","key":"10.7717\/peerj-cs.3416\/ref-32","doi-asserted-by":"publisher","first-page":"2628","DOI":"10.1021\/acs.jcim.3c00200","article-title":"Artificial intelligence in drug toxicity prediction: recent advances, challenges, and future perspectives","volume":"63","author":"Tran","year":"2023","journal-title":"Journal of Chemical Information and Modeling"},{"issue":"2","key":"10.7717\/peerj-cs.3416\/ref-33","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1093\/toxsci\/kfy058","article-title":"Prediction of drug-induced hepatotoxicity using long-term stable primary hepatic 3d spheroid cultures in chemically defined conditions","volume":"163","author":"Vorrink","year":"2018","journal-title":"Toxicological Sciences"},{"issue":"6","key":"10.7717\/peerj-cs.3416\/ref-34","doi-asserted-by":"publisher","first-page":"1373","DOI":"10.1002\/hep.32283","article-title":"Prediction of biochemical nonresolution in patients with chronic drug-induced liver injury: a large multicenter study","volume":"75","author":"Wang","year":"2022","journal-title":"Hepatology"},{"issue":"10","key":"10.7717\/peerj-cs.3416\/ref-35","doi-asserted-by":"publisher","first-page":"2085","DOI":"10.1021\/acs.jcim.5b00238","article-title":"Deep learning for drug-induced liver injury","volume":"55","author":"Xu","year":"2015","journal-title":"Journal of Chemical Information and Modeling"},{"issue":"3","key":"10.7717\/peerj-cs.3416\/ref-36","doi-asserted-by":"publisher","first-page":"208","DOI":"10.1038\/tpj.2013.39","article-title":"Data mining reveals a network of early-response genes as a consensus signature of drug-induced in vitro and in vivo toxicity","volume":"14","author":"Zhang","year":"2013","journal-title":"The Pharmacogenomics Journal"},{"issue":"3\u20134","key":"10.7717\/peerj-cs.3416\/ref-37","doi-asserted-by":"publisher","first-page":"136","DOI":"10.1002\/minf.201500055","article-title":"In silico prediction of drug induced liver toxicity using substructure pattern recognition method","volume":"35","author":"Zhang","year":"2016","journal-title":"Molecular Informatics"},{"issue":"9","key":"10.7717\/peerj-cs.3416\/ref-38","doi-asserted-by":"publisher","first-page":"1714","DOI":"10.3390\/molecules24091714","article-title":"Revealing drug-target interactions with computational models and algorithms","volume":"24","author":"Zhou","year":"2019","journal-title":"Molecules"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3416.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3416.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3416.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3416.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T08:45:31Z","timestamp":1764751531000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3416"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,3]]},"references-count":38,"alternative-id":["10.7717\/peerj-cs.3416"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3416","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,3]]},"article-number":"e3416"}}