{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T14:09:13Z","timestamp":1789135753068,"version":"build-2803163510"},"reference-count":32,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T00:00:00Z","timestamp":1785369600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T00:00:00Z","timestamp":1789084800000},"content-version":"vor","delay-in-days":43,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Traditional QSAR toxicity models are, in general, assessed with random train-test splits where structural overlaps between train and test compounds are allowed. This usually results in an inflated predictive performance. Therefore, this paper looks at the problem of toxicity prediction under the structural distribution shift and checks if representation-level invariance can contribute to better cross-family generalization. A set of 1792 structurally diverse organic molecules for which toxicity data (e.g., Tetrahymena pyriformis pIGC\u2085\u2080) were determined experimentally was modeled with physicochemical descriptors. In order to depict the realistic scenarios of model use, the leave-one-cluster-out (LOCO) protocol was applied to enforce strict structural separation of training and test domains. Baseline neural models lost a lot of their prediction accuracy under LOCO versus random splits, thus exposing a very large generalization gap. On the other hand, invariant learning methods such as invariant risk minimization, contrastive alignment, and domain-adversarial training managed not only to reduce the cross-domain error but also to make residual distributions more stable. The embedding of latent space further demonstrated that invariance helps to get rid of cluster-specific signals while keeping toxicity-relevant gradients intact. From a practical perspective, these results suggest that robust in silico predictive toxicology, further under structural distribution shift, can be achieved through domain-aware validation and invariant representation learning.<\/jats:p>","DOI":"10.1038\/s41598-026-64174-8","type":"journal-article","created":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T09:48:05Z","timestamp":1785404885000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Domain-generalized representation learning for cross-chemical-family toxicity prediction"],"prefix":"10.1038","volume":"16","author":[{"given":"Wael A.","family":"Mahdi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adel","family":"Alhowyan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmad J.","family":"Obaidullah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,30]]},"reference":[{"key":"64174_CR1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tox.2025.154230","author":"D Kim","year":"2025","unstructured":"Kim, D. & Choi, J. AI-based toxicity prediction models using ToxCast data: Current status and future directions for explainable models. Toxicology https:\/\/doi.org\/10.1016\/j.tox.2025.154230 (2025).","journal-title":"Toxicology"},{"issue":"6","key":"64174_CR2","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1002\/jat.4586","volume":"44","author":"Y Zhou","year":"2024","unstructured":"Zhou, Y. et al. In silico prediction of ocular toxicity of compounds using explainable machine learning and deep learning approaches. J. Appl. Toxicol. 44(6), 892\u2013907 (2024).","journal-title":"J. Appl. Toxicol."},{"issue":"7","key":"64174_CR3","doi-asserted-by":"publisher","first-page":"953","DOI":"10.1002\/jat.4591","volume":"44","author":"Y Zhou","year":"2024","unstructured":"Zhou, Y. et al. ToxMPNN: A deep learning model for small molecule toxicity prediction. J. Appl. Toxicol. 44(7), 953\u2013964 (2024).","journal-title":"J. Appl. Toxicol."},{"issue":"Suppl 13","key":"64174_CR4","doi-asserted-by":"publisher","first-page":"387","DOI":"10.1186\/s12859-020-03682-4","volume":"21","author":"R Zhou","year":"2020","unstructured":"Zhou, R. et al. NEDD: A network embedding based method for predicting drug-disease associations. BMC Bioinform. 21(Suppl 13), 387 (2020).","journal-title":"BMC Bioinform."},{"issue":"9","key":"64174_CR5","doi-asserted-by":"publisher","first-page":"1714","DOI":"10.3390\/molecules24091714","volume":"24","author":"L Zhou","year":"2019","unstructured":"Zhou, L. et al. Revealing drug-target interactions with computational models and algorithms. Molecules 24(9), 1714 (2019).","journal-title":"Molecules"},{"issue":"6","key":"64174_CR6","doi-asserted-by":"publisher","first-page":"4201","DOI":"10.1039\/D3RA07322J","volume":"14","author":"A Setiya","year":"2024","unstructured":"Setiya, A., Jani, V., Sonavane, U. & Joshi, R. MolToxPred: Small molecule toxicity prediction using machine learning approach. RSC Adv. 14(6), 4201\u20134220 (2024).","journal-title":"RSC Adv."},{"key":"64174_CR7","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1016\/B978-0-443-15339-6.00013-8","volume-title":"QSAR in Safety Evaluation and Risk Assessment","author":"J Liu","year":"2024","unstructured":"Liu, J. et al. QSAR models for predicting in vivo reproductive toxicity. In QSAR in Safety Evaluation and Risk Assessment 315\u2013327 (Elsevier, 2024)."},{"key":"64174_CR8","doi-asserted-by":"publisher","DOI":"10.1016\/j.yrtph.2024.105716","volume":"154","author":"J Achar","year":"2024","unstructured":"Achar, J. et al. Analysis of implicit and explicit uncertainties in QSAR prediction of chemical toxicity: A case study of neurotoxicity. Regul. Toxicol. Pharmacol. 154, 105716 (2024).","journal-title":"Regul. Toxicol. Pharmacol."},{"key":"64174_CR9","doi-asserted-by":"crossref","unstructured":"Voigt, M. & Jaeger, M. In silico and in vivo ecotoxicity\u2014QSAR-based predictions and experimental assays for the aquatic environment. QSAR Saf. Eval. risk Assess. 495\u2013509 (2024).","DOI":"10.1016\/B978-0-443-15339-6.00018-7"},{"issue":"4","key":"64174_CR10","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1080\/1062936X.2024.2345618","volume":"35","author":"S Danishuddin","year":"2024","unstructured":"Danishuddin, S., Malik, M. Z., Kashif, M., Haque, S. & Kim, J. J. Exploring chemical space, scaffold diversity, and activity landscape of spleen tyrosine kinase active inhibitors. SAR QSAR Environ. Res. 35(4), 325\u2013342 (2024).","journal-title":"SAR QSAR Environ. Res."},{"key":"64174_CR11","doi-asserted-by":"publisher","DOI":"10.1016\/j.envint.2026.110096","author":"D Lopez-Rodriguez","year":"2026","unstructured":"Lopez-Rodriguez, D., Guerrero-Lim\u00f3n, G. & Ch\u00e8vre, N. A chemical space model for the exploration of eco-toxicological data. Environ. Int. https:\/\/doi.org\/10.1016\/j.envint.2026.110096 (2026).","journal-title":"Environ. Int."},{"issue":"2","key":"64174_CR12","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1016\/j.therap.2023.10.008","volume":"79","author":"A Destere","year":"2024","unstructured":"Destere, A. et al. Drug-induced cardiac toxicity and adverse drug reactions, a narrative review. Therapies 79(2), 161\u2013172 (2024).","journal-title":"Therapies"},{"issue":"02","key":"64174_CR13","first-page":"13","volume":"2","author":"WT Almagharbeh","year":"2024","unstructured":"Almagharbeh, W. T., Abdullah, S. & Ullah, R. Mechanisms of drug-induced liver injury: Exploring pathways and risk factors for liver toxicity. Trends Life Sci. Biotechnol. 2(02), 13\u201324 (2024).","journal-title":"Trends Life Sci. Biotechnol."},{"issue":"8","key":"64174_CR14","doi-asserted-by":"publisher","first-page":"960","DOI":"10.1038\/s41589-024-01679-1","volume":"20","author":"DB Catacutan","year":"2024","unstructured":"Catacutan, D. B., Alexander, J., Arnold, A. & Stokes, J. M. Machine learning in preclinical drug discovery. Nat. Chem. Biol. 20(8), 960\u2013973 (2024).","journal-title":"Nat. Chem. Biol."},{"key":"64174_CR15","doi-asserted-by":"publisher","first-page":"1355044","DOI":"10.3389\/fddsv.2024.1355044","volume":"4","author":"T Hartung","year":"2024","unstructured":"Hartung, T. The (misleading) role of animal models in drug development. Front. Drug Discov. 4, 1355044 (2024).","journal-title":"Front. Drug Discov."},{"issue":"1","key":"64174_CR16","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1186\/s40364-025-00758-2","volume":"13","author":"A Ocana","year":"2025","unstructured":"Ocana, A. et al. Integrating artificial intelligence in drug discovery and early drug development: A transformative approach. Biomark. Res. 13(1), 45 (2025).","journal-title":"Biomark. Res."},{"issue":"6","key":"64174_CR17","doi-asserted-by":"publisher","first-page":"827","DOI":"10.1021\/acs.chemrestox.3c00352","volume":"37","author":"AMB Amorim","year":"2024","unstructured":"Amorim, A. M. B. et al. Advancing drug safety in drug development: Bridging computational predictions for enhanced toxicity prediction. Chem. Res. Toxicol. 37(6), 827\u2013849 (2024).","journal-title":"Chem. Res. Toxicol."},{"issue":"5","key":"64174_CR18","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1021\/acs.chemrestox.5c00033","volume":"38","author":"S Seal","year":"2025","unstructured":"Seal, S. et al. Machine learning for toxicity prediction using chemical structures: Pillars for success in the real world. Chem. Res. Toxicol. 38(5), 759\u2013807 (2025).","journal-title":"Chem. Res. Toxicol."},{"issue":"16","key":"64174_CR19","doi-asserted-by":"publisher","first-page":"2413405","DOI":"10.1002\/advs.202413405","volume":"12","author":"C Bai","year":"2025","unstructured":"Bai, C. et al. Machine learning-enabled drug-induced toxicity prediction. Adv. Sci. 12(16), 2413405 (2025).","journal-title":"Adv. Sci."},{"key":"64174_CR20","doi-asserted-by":"publisher","first-page":"1632046","DOI":"10.3389\/fchem.2025.1632046","volume":"13","author":"H Lee","year":"2025","unstructured":"Lee, H., Kim, J., Kim, J.-W. & Lee, Y. Recent advances in AI-based toxicity prediction for drug discovery. Front. Chem. 13, 1632046 (2025).","journal-title":"Front. Chem."},{"key":"64174_CR21","doi-asserted-by":"publisher","DOI":"10.1016\/j.tox.2022.153325","volume":"480","author":"Z Fang","year":"2022","unstructured":"Fang, Z., Yu, X. & Zeng, Q. Random forest algorithm-based accurate prediction of chemical toxicity to Tetrahymena pyriformis. Toxicology 480, 153325 (2022).","journal-title":"Toxicology"},{"issue":"7","key":"64174_CR22","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1021\/ci100176x","volume":"50","author":"D Fourches","year":"2010","unstructured":"Fourches, D., Muratov, E. & Tropsha, A. Trust, but verify: On the importance of chemical structure curation in cheminformatics and QSAR modeling research. J. Chem. Inf. Model. 50(7), 1189 (2010).","journal-title":"J. Chem. Inf. Model."},{"key":"64174_CR23","doi-asserted-by":"publisher","DOI":"10.1016\/j.psep.2025.108314","author":"M Sun","year":"2025","unstructured":"Sun, M. et al. Predicting the toxicity of organic chemicals to Tetrahymena pyriformis based on machine learning and interpretable models. Process Saf. Environ. Prot. https:\/\/doi.org\/10.1016\/j.psep.2025.108314 (2025).","journal-title":"Process Saf. Environ. Prot."},{"issue":"1","key":"64174_CR24","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1007\/s13755-025-00344-8","volume":"13","author":"W Zheng","year":"2025","unstructured":"Zheng, W. et al. GEP-DNN4Mol: automatic chemical molecular design based on deep neural networks and gene expression programming. Heal. Inf. Sci. Syst. 13(1), 31 (2025).","journal-title":"Heal. Inf. Sci. Syst."},{"issue":"29","key":"64174_CR25","doi-asserted-by":"publisher","first-page":"25990","DOI":"10.1021\/jacs.5c08771","volume":"147","author":"W-Z Yang","year":"2025","unstructured":"Yang, W.-Z. et al. Giant five-shell polyoxometalate cages and the single-cluster-based nanowire superstructures. J. Am. Chem. Soc. 147(29), 25990\u201325997 (2025).","journal-title":"J. Am. Chem. Soc."},{"issue":"2","key":"64174_CR26","doi-asserted-by":"publisher","first-page":"2217076","DOI":"10.1080\/21645515.2023.2217076","volume":"19","author":"Y Jia","year":"2023","unstructured":"Jia, Y., Fu, B., Dong, L. & Zhao, M. Sweet syndrome induced by SARS-CoV-2 vaccines: A systematic review of patient-report studies. Hum. Vaccin. Immunother. 19(2), 2217076 (2023).","journal-title":"Hum. Vaccin. Immunother."},{"key":"64174_CR27","unstructured":"Yoshida, K. & Naganuma, H. Towards understanding variants of invariant risk minimization through the lens of calibration. arXiv:2401.17541 (2024)."},{"key":"64174_CR28","unstructured":"Lai, Z.-R. & Wang, W. Invariant risk minimization is a total variation model. arXiv:2405.01389 (2024)."},{"key":"64174_CR29","doi-asserted-by":"crossref","unstructured":"Liu, J. & Chen, S. Timesurl: Self-supervised contrastive learning for universal time series representation learning. In Proceedings of the AAAI Conference on Artificial Intelligence 13918\u201313926 (2024).","DOI":"10.1609\/aaai.v38i12.29299"},{"key":"64174_CR30","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2024.110874","volume":"157","author":"ZZ Darban","year":"2025","unstructured":"Darban, Z. Z., Webb, G. I., Pan, S., Aggarwal, C. C. & Salehi, M. CARLA: Self-supervised contrastive representation learning for time series anomaly detection. Pattern Recognit. 157, 110874 (2025).","journal-title":"Pattern Recognit."},{"key":"64174_CR31","doi-asserted-by":"publisher","DOI":"10.1016\/j.measurement.2025.117113","volume":"250","author":"B Wang","year":"2025","unstructured":"Wang, B. et al. Enhancing robustness of cross-machine fault diagnosis via an improved domain adversarial neural network and self-adversarial training. Measurement 250, 117113 (2025).","journal-title":"Measurement"},{"key":"64174_CR32","doi-asserted-by":"publisher","DOI":"10.1016\/j.intimp.2024.112178","volume":"134","author":"W Fang","year":"2024","unstructured":"Fang, W., Sun, W., Fang, W., Zhao, S. & Wang, C. Clinical features, treatment, and outcomes of patients with carfilzomib induced thrombotic microangiopathy. Int. Immunopharmacol. 134, 112178 (2024).","journal-title":"Int. Immunopharmacol."}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-64174-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-64174-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-64174-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T13:31:32Z","timestamp":1789133492000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-64174-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,30]]},"references-count":32,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["64174"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-64174-8","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,30]]},"assertion":[{"value":"1 May 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors confirm that they have no conflict of interest.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"28484"}}