{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T19:44:02Z","timestamp":1792784642044,"version":"4.5.0"},"reference-count":40,"publisher":"American Chemical Society (ACS)","issue":"41","license":[{"start":{"date-parts":[[2026,10,4]],"date-time":"2026-10-04T00:00:00Z","timestamp":1791072000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,10,4]],"date-time":"2026-10-04T00:00:00Z","timestamp":1791072000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,10,4]],"date-time":"2026-10-04T00:00:00Z","timestamp":1791072000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100004731","name":"Natural Science Foundation of Zhejiang Province","doi-asserted-by":"publisher","award":["LZ26B070001"],"award-info":[{"award-number":["LZ26B070001"]}],"id":[{"id":"10.13039\/501100004731","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Natural Science Foundation of China (NSFC)","award":["U24A20520"],"award-info":[{"award-number":["U24A20520"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,10,20]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Chemical exposure causes adverse neurological outcomes, highlighting the need for high-throughput screening of chemical neurotoxicity. However, existing computational methods rely on single-modality inputs and lack complementary molecular information to sufficiently enhance the accuracy and efficiency of neurotoxicity risk assessment. Here, we present NeuroToxPredictor, a scalable and interpretable multimodal framework built on a curated chemical neurotoxicity dataset for accurate and efficient chemical neurotoxicity prediction. It integrates graph attention networks and molecular fingerprints with ChemBERTa, a pretrained chemical language model that provides contextual encoding of SMILES. A gated fusion module adaptively weights these complementary representations according to their predictive relevance to enhance modality-specific weight allocation through a gating fusion mechanism. NeuroToxPredictor achieved competitive performance across five independent training seeds (AUC = 0.919 \u00b1 0.004; accuracy = 0.845 \u00b1 0.014; F1-score = 0.861 \u00b1 0.011; recall = 0.884 \u00b1 0.013; precision = 0.839 \u00b1 0.024; MCC = 0.688 \u00b1 0.028), outperforming both single-modality and reduced-modality baselines. It demonstrates strong generalizability with an AUC of 0.861, as externally validated on an independent set of 279 compounds. NeuroToxPredictor is deployed on a publicly platform (https:\/\/www.ai4environ.cn\/NeuroToxPredictor), enabling an end-to-end intelligent workflow from chemical retrieval, neurotoxicity prediction, applicability-domain assessment, and large language model-assisted result interpretation.<\/jats:p>","DOI":"10.1021\/acs.est.6c10122","type":"journal-article","created":{"date-parts":[[2026,10,4]],"date-time":"2026-10-04T14:01:51Z","timestamp":1791122511000},"page":"29441-29451","source":"Crossref","is-referenced-by-count":0,"title":["Scalable and Interpretable\nChemical Neurotoxicity\nPrediction via Multi-Modal Gated Fusion and ChemBERTa"],"prefix":"10.1021","volume":"60","author":[{"given":"Lilai","family":"Shen","sequence":"first","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziyu","family":"Chen","sequence":"additional","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yizhou","family":"Huang","sequence":"additional","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianing","family":"Liu","sequence":"additional","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinhui","family":"Yuan","sequence":"additional","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7774-7239","authenticated-orcid":true,"given":"Shulin","family":"Zhuang","sequence":"additional","affiliation":[{"name":"Zhejiang University , , ,","place":["Hangzhou, China, 310058"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,10,4]]},"reference":[{"issue":"3","key":"2026102007531048600_cit1","doi-asserted-by":"publisher","first-page":"330","DOI":"10.1016\/S1474-4422(13)70278-3","article-title":"Neurobehavioural effects of developmental toxicity","volume":"13","author":"Grandjean","year":"2014","journal-title":"Lancet Neurol."},{"issue":"6","key":"2026102007531048600_cit2","doi-asserted-by":"publisher","first-page":"1833","DOI":"10.1038\/s41390-024-03547-z","article-title":"Development and\nchild health in a world of synthetic\nchemicals","volume":"97","author":"Wager","year":"2025","journal-title":"Pediatr. Res."},{"key":"2026102007531048600_cit3","doi-asserted-by":"publisher","first-page":"306","DOI":"10.14573\/altex.1712081","article-title":"Recommendation\non test readiness criteria for new approach methods in toxicology:\nExemplified for developmental neurotoxicity","author":"Bal-Price","year":"2018","journal-title":"ALTEX"},{"issue":"1","key":"2026102007531048600_cit4","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1093\/toxsci\/kfy278","article-title":"Screening\nfor developmental neurotoxicity at the National Toxicology Program:\nThe future is here","volume":"167","author":"Behl","year":"2019","journal-title":"Toxicol. Sci."},{"issue":"1","key":"2026102007531048600_cit5","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1093\/toxsci\/kfx169","article-title":"From the\ncover: Developmental neurotoxicants disrupt activity in cortical networks\non microelectrode arrays: Results of screening 86 compounds during\nneural network formation","volume":"160","author":"Frank","year":"2017","journal-title":"Toxicol. Sci."},{"issue":"8","key":"2026102007531048600_cit6","doi-asserted-by":"publisher","first-page":"4788","DOI":"10.1021\/acs.est.1c08471","article-title":"Noninvasive electrophysiology: Emerging prospects in\naquatic neurotoxicity testing","volume":"56","author":"Tomasello","year":"2022","journal-title":"Environ. Sci.\nTechnol."},{"key":"2026102007531048600_cit7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jhazmat.2024.133443","article-title":"NeuTox: A weighted ensemble model for screening potential neuronal\ncytotoxicity of chemicals based on various types of molecular representations","volume":"465","author":"He","year":"2024","journal-title":"J. Hazard. Mater."},{"issue":"51","key":"2026102007531048600_cit8","doi-asserted-by":"publisher","first-page":"22646","DOI":"10.1021\/acs.est.4c10081","article-title":"ChemNTP:\nAdvanced prediction of neurotoxicity\ntargets for environmental chemicals using a Siamese neural network","volume":"58","author":"Zhang","year":"2024","journal-title":"Environ. Sci. Technol."},{"issue":"22","key":"2026102007531048600_cit9","doi-asserted-by":"publisher","first-page":"11108","DOI":"10.1021\/acs.est.5c03311","article-title":"Development\nof an interpretable machine\nlearning model for neurotoxicity prediction of environmentally related\ncompounds","volume":"59","author":"Hao","year":"2025","journal-title":"Environ. Sci. Technol."},{"key":"2026102007531048600_cit10","doi-asserted-by":"publisher","DOI":"10.1016\/j.envint.2024.109244","article-title":"NeuTox 2.0: A hybrid deep learning architecture for screening potential\nneurotoxicity of chemicals based on multimodal feature fusion","volume":"195","author":"Pang","year":"2025","journal-title":"Environ. Int."},{"key":"2026102007531048600_cit11","doi-asserted-by":"publisher","DOI":"10.1016\/j.envint.2022.107224","article-title":"Predicting chemical\necotoxicity by learning latent space chemical representations","volume":"163","author":"Gao","year":"2022","journal-title":"Environ. Int."},{"issue":"51","key":"2026102007531048600_cit12","doi-asserted-by":"publisher","first-page":"47536","DOI":"10.1021\/acsomega.2c05693","article-title":"Machine learning toxicity\nprediction: Latest advances\nby toxicity end point","volume":"7","author":"Cavasotto","year":"2022","journal-title":"ACS Omega"},{"issue":"16","key":"2026102007531048600_cit13","doi-asserted-by":"publisher","first-page":"8202","DOI":"10.1039\/D4CS00196F","article-title":"Navigating the landscape of enzyme design: From molecular simulations\nto machine learning","volume":"53","author":"Zhou","year":"2024","journal-title":"Chem. Soc. Rev."},{"issue":"21","key":"2026102007531048600_cit14","doi-asserted-by":"publisher","first-page":"10286","DOI":"10.1021\/acs.est.4c13813","article-title":"Predicting\nbioconcentration\nfactors of per- and polyfluoroalkyl substances using a directed message\npassing neural network with multimodal feature fusion","volume":"59","author":"Yao","year":"2025","journal-title":"Environ. Sci. Technol."},{"issue":"2","key":"2026102007531048600_cit15","doi-asserted-by":"publisher","first-page":"267","DOI":"10.1002\/ieam.1885","article-title":"Recommended\napproaches to the scientific evaluation of ecotoxicological hazards\nand risks of endocrine-active substances","volume":"13","author":"Matthiessen","year":"2017","journal-title":"Integr.\nEnviron. Assess. Manage."},{"issue":"10","key":"2026102007531048600_cit16","doi-asserted-by":"publisher","first-page":"7952","DOI":"10.1021\/acs.est.5c17401","article-title":"Screening of respiratory toxicity of environmental compounds based\non multimodal feature fusion model","volume":"60","author":"Xu","year":"2026","journal-title":"Environ.\nSci. Technol."},{"issue":"1","key":"2026102007531048600_cit17","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1021\/acs.jcim.3c01250","article-title":"Chemprop:\nA machine learning package for chemical property prediction","volume":"64","author":"Heid","year":"2024","journal-title":"J. Chem. Inf. Model."},{"issue":"12","key":"2026102007531048600_cit18","doi-asserted-by":"publisher","first-page":"1353","DOI":"10.1038\/s41551-022-00942-x","article-title":"Graph representation\nlearning in\nbiomedicine and healthcare","volume":"6","author":"Li","year":"2022","journal-title":"Nat. Biomed. Eng."},{"issue":"1","key":"2026102007531048600_cit19","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-017-0247-6","article-title":"The CompTox Chemistry Dashboard: A community data resource\nfor environmental chemistry","volume":"9","author":"Williams","year":"2017","journal-title":"J. Cheminf."},{"issue":"2","key":"2026102007531048600_cit20","doi-asserted-by":"publisher","first-page":"189","DOI":"10.1021\/acs.chemrestox.0c00264","article-title":"The\nTox21 10K compound library: Collaborative chemistry advancing toxicology","volume":"34","author":"Richard","year":"2021","journal-title":"Chem. Res. Toxicol."},{"issue":"D1","key":"2026102007531048600_cit21","doi-asserted-by":"publisher","first-page":"D1074","DOI":"10.1093\/nar\/gkx1037","article-title":"DrugBank\n5.0: A major update to the DrugBank database for 2018","volume":"46","author":"Wishart","year":"2018","journal-title":"Nucleic Acids Res."},{"issue":"D1","key":"2026102007531048600_cit22","doi-asserted-by":"publisher","first-page":"D1388","DOI":"10.1093\/nar\/gkaa971","article-title":"PubChem in 2021: New data content and improved\nweb interfaces","volume":"49","author":"Kim","year":"2021","journal-title":"Nucleic Acids Res."},{"issue":"D1","key":"2026102007531048600_cit23","doi-asserted-by":"publisher","first-page":"D1257","DOI":"10.1093\/nar\/gkac833","article-title":"Comparative Toxicogenomics Database (CTD): Update 2023","volume":"51","author":"Davis","year":"2023","journal-title":"Nucleic Acids Res."},{"issue":"8","key":"2026102007531048600_cit24","doi-asserted-by":"publisher","first-page":"1225","DOI":"10.1021\/acs.chemrestox.6b00135","article-title":"ToxCast\nchemical landscape: Paving the road to 21st century toxicology","volume":"29","author":"Richard","year":"2016","journal-title":"Chem. Res. Toxicol."},{"issue":"3","key":"2026102007531048600_cit25","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1021\/acs.chemrestox.2c00344","article-title":"Evaluation of per- and polyfluoroalkyl substances (PFAS) in vitro\ntoxicity testing for developmental neurotoxicity","volume":"36","author":"Carstens","year":"2023","journal-title":"Chem. Res. Toxicol."},{"issue":"D1","key":"2026102007531048600_cit26","doi-asserted-by":"publisher","first-page":"D1075","DOI":"10.1093\/nar\/gkv1075","article-title":"The SIDER database\nof drugs and side effects","volume":"44","author":"Kuhn","year":"2016","journal-title":"Nucleic Acids Res."},{"issue":"1","key":"2026102007531048600_cit27","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-45566-8","article-title":"Transfer learning\nwith graph neural networks for improved molecular property prediction\nin the multi-fidelity setting","volume":"15","author":"Buterez","year":"2024","journal-title":"Nat. Commun."},{"issue":"8","key":"2026102007531048600_cit28","doi-asserted-by":"publisher","first-page":"3370","DOI":"10.1021\/acs.jcim.9b00237","article-title":"Analyzing learned molecular\nrepresentations for property prediction","volume":"59","author":"Yang","year":"2019","journal-title":"J.\nChem. Inf. Model."},{"issue":"2","key":"2026102007531048600_cit29","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/C7SC02664A","article-title":"MoleculeNet: A benchmark\nfor molecular machine learning","volume":"9","author":"Wu","year":"2018","journal-title":"Chem. Sci."},{"issue":"15","key":"2026102007531048600_cit30","doi-asserted-by":"publisher","first-page":"2887","DOI":"10.1021\/jm9602928","article-title":"The properties of\nknown drugs. 1. Molecular frameworks","volume":"39","author":"Bemis","year":"1996","journal-title":"J. Med.\nChem."},{"issue":"4","key":"2026102007531048600_cit31","doi-asserted-by":"publisher","first-page":"747","DOI":"10.1021\/ci9803381","article-title":"Unsupervised\ndata base clustering based on Daylight\u2019s fingerprint and Tanimoto\nsimilarity: A fast and automated way to cluster small and large data\nsets","volume":"39","author":"Butina","year":"1999","journal-title":"J. Chem. Inf. Comput. Sci."},{"issue":"3","key":"2026102007531048600_cit32","doi-asserted-by":"publisher","first-page":"697","DOI":"10.1021\/acs.jcim.3c01774","article-title":"Real-world\nmolecular out-of-distribution: Specification and investigation","volume":"64","author":"Tossou","year":"2024","journal-title":"J. Chem. Inf. Model."},{"issue":"13","key":"2026102007531048600_cit33","doi-asserted-by":"publisher","first-page":"4012","DOI":"10.1021\/acs.jcim.3c00373","article-title":"Characterizing uncertainty\nin machine learning for chemistry","volume":"63","author":"Heid","year":"2023","journal-title":"J. Chem.\nInf. Model."},{"issue":"8","key":"2026102007531048600_cit34","doi-asserted-by":"publisher","first-page":"3770","DOI":"10.1021\/acs.jcim.0c00502","article-title":"Uncertainty\nquantification using neural networks for molecular property prediction","volume":"60","author":"Hirschfeld","year":"2020","journal-title":"J. Chem. Inf. Model."},{"issue":"5","key":"2026102007531048600_cit35","doi-asserted-by":"publisher","first-page":"2268","DOI":"10.1021\/acs.jcim.4c01371","article-title":"Large language models\nas tools for molecular toxicity prediction: AI insights into cardiotoxicity","volume":"65","author":"Yang","year":"2025","journal-title":"J. Chem. Inf. Model."},{"issue":"7992","key":"2026102007531048600_cit36","doi-asserted-by":"publisher","first-page":"570","DOI":"10.1038\/s41586-023-06792-0","article-title":"Autonomous chemical\nresearch with large language models","volume":"624","author":"Boiko","year":"2023","journal-title":"Nature"},{"issue":"1","key":"2026102007531048600_cit37","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-39016-5","article-title":"High-throughput\nscreening and classification of chemicals and their effects on neuronal\ngene expression using RASL-seq","volume":"9","author":"Simon","year":"2019","journal-title":"Sci. Rep."},{"issue":"14","key":"2026102007531048600_cit38","doi-asserted-by":"publisher","first-page":"7187","DOI":"10.1021\/acs.est.4c14787","article-title":"Artificial\nintelligence for the discovery of safe and effective flame retardants","volume":"59","author":"Chen","year":"2025","journal-title":"Environ. Sci. Technol."},{"issue":"19","key":"2026102007531048600_cit39","doi-asserted-by":"publisher","first-page":"13911","DOI":"10.1021\/acs.est.6c00505","article-title":"Integrating adverse outcome pathway-Bayesian\nnetwork and physiologically based pharmacokinetic modeling for quantitative\nprediction of developmental neurotoxicity of F-53B under human-relevant\nexposure scenarios","volume":"60","author":"Feng","year":"2026","journal-title":"Environ. Sci. Technol."},{"issue":"1","key":"2026102007531048600_cit40","doi-asserted-by":"publisher","DOI":"10.1038\/ncomms10425","article-title":"Modelling the Tox21\n10K chemical profiles for in vivo toxicity prediction and mechanism\ncharacterization","volume":"7","author":"Huang","year":"2016","journal-title":"Nat. Commun."}],"container-title":["Environmental Science &amp;\nTechnology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.est.6c10122","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/esthag\/article-pdf\/doi\/10.1021\/acs.est.6c10122\/68740243\/acs.est.6c10122.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/esthag\/article-pdf\/doi\/10.1021\/acs.est.6c10122\/68740243\/acs.est.6c10122.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T12:20:46Z","timestamp":1792498846000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/esthag\/article\/60\/41\/29441\/5444254\/Scalable-and-Interpretable-Chemical-Neurotoxicity"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,4]]},"references-count":40,"journal-issue":{"issue":"41","published-online":{"date-parts":[[2026,10,4]]},"published-print":{"date-parts":[[2026,10,20]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.est.6c10122","relation":{},"ISSN":["0013-936X","1520-5851"],"issn-type":[{"value":"0013-936X","type":"print"},{"value":"1520-5851","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,4]]}}}