{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T14:58:50Z","timestamp":1790607530996,"version":"4.1.0"},"reference-count":50,"publisher":"American Chemical Society (ACS)","issue":"18","license":[{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100012131","name":"Department of Science and Technology of Liaoning Province","doi-asserted-by":"publisher","award":["2025-MSLH-351"],"award-info":[{"award-number":["2025-MSLH-351"]}],"id":[{"id":"10.13039\/501100012131","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Fundamental Research Funds for the Liaoning Universities","award":["LJ212410146026"],"award-info":[{"award-number":["LJ212410146026"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,28]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Antimicrobial resistance (AMR) poses a growing threat to global public health, particularly for priority pathogens such as Staphylococcus aureus (SA) and Neisseria gonorrheae (NG), for which therapeutic options are increasingly limited. Although deep learning has accelerated antibiotic discovery, existing approaches are often constrained by single-modality molecular representations and a lack of joint modeling for antibacterial activity and host cell toxicity. Here, we present MAPViS, a task-adaptive multimodal pretrained framework for jointly predicting antibacterial activity against SA and NG together with cytotoxicity risk in three human cell lines (HepG2, HSkMC, and IMR-90). MAPViS integrates molecular graph representations, molecular fingerprints, and descriptor\/3D geometric features through a gated fusion mechanism and combines large-scale self-supervised pretraining with supervised task-specific fine-tuning. The graph encoder is pretrained on approximately 10 million molecules using hierarchical pseudolabel classification and masked graph contrastive learning to improve representation quality and transferability. On benchmark datasets derived from experimentally screened compounds, MAPViS achieves competitive performance across antibacterial activity prediction tasks and shows advantages in early recognition metrics relevant to virtual screening, including AUPRC, Precision@30, and EF@1.0%. External validation further shows strong top-ranked enrichment on the SA task, where all Top-10 candidates are experimentally active, whereas performance on NG is more modest, indicating task-dependent generalization. Overall, MAPViS provides a scalable multimodal framework for joint activity\u2013toxicity modeling and offers a practical computational tool for antibiotic virtual screening and candidate prioritization.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c02659","type":"journal-article","created":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T16:01:30Z","timestamp":1789056090000},"page":"11793-11806","source":"Crossref","is-referenced-by-count":0,"title":["A Task-Adaptive\nMultimodal Pretrained Framework for\nAntibiotic Virtual Screening with Joint Activity and Cytotoxicity\nPrediction"],"prefix":"10.1021","volume":"66","author":[{"given":"Zhijie","family":"Pan","sequence":"first","affiliation":[{"name":"University of Science and Technology Liaoning , , ,","place":["Anshan, China, 114051"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Yang","sequence":"additional","affiliation":[{"name":"Beijing Institute of Technology , , ,","place":["Beijing, China, 100081"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenchi","family":"Ge","sequence":"additional","affiliation":[{"name":"University of Science and Technology Liaoning , , ,","place":["Anshan, China, 114051"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Churong","family":"Wang","sequence":"additional","affiliation":[{"name":"University of Science and Technology Liaoning , , ,","place":["Anshan, China, 114051"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianqiang","family":"Sun","sequence":"additional","affiliation":[{"name":"Linyi University , , ,","place":["Linyi, China, 276000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9713-1864","authenticated-orcid":true,"given":"Qi","family":"Zhao","sequence":"additional","affiliation":[{"name":"University of Science and Technology Liaoning , , ,","place":["Anshan, China, 114051"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,9,10]]},"reference":[{"issue":"10459","key":"2026092810071875200_cit1","doi-asserted-by":"publisher","first-page":"1199","DOI":"10.1016\/S0140-6736(24)01867-1","article-title":"Global Burden of Bacterial\nAntimicrobial Resistance 1990\u20132021: A Systematic Analysis with\nForecasts to 2050","volume":"404","author":"Naghavi","year":"2024","journal-title":"Lancet"},{"key":"2026092810071875200_cit2","unstructured":"O\u2019Neill, J.\n          \n          Tackling\nDrug-Resistant Infections Globally: Final Report and Recommendations. Final Report, Review on Antimicrobial Resistance, 2016."},{"issue":"10325","key":"2026092810071875200_cit3","doi-asserted-by":"publisher","first-page":"629","DOI":"10.1016\/S0140-6736(21)02724-0","article-title":"Global Burden of Bacterial\nAntimicrobial Resistance in 2019: A Systematic Analysis","volume":"399","author":"Murray","year":"2022","journal-title":"Lancet"},{"key":"2026092810071875200_cit4","doi-asserted-by":"crossref","DOI":"10.15620\/cdc:82532","volume-title":"Antibiotic Resistance Threats in the United\nStates, 2019","author":"Centers for Disease Control and Prevention.","year":"2019"},{"key":"2026092810071875200_cit5","first-page":"12","volume-title":"WHO Bacterial Priority\nPathogens List","author":"World Health Organization.","year":"2024"},{"issue":"1","key":"2026092810071875200_cit6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1086\/595011","article-title":"Bad Bugs, No Drugs:\nNo ESKAPE! An Update from the Infectious Diseases Society of America","volume":"48","author":"Boucher","year":"2009","journal-title":"Clin. Infect. Dis."},{"issue":"6","key":"2026092810071875200_cit7","doi-asserted-by":"publisher","first-page":"3091","DOI":"10.1021\/acs.jcim.6c00033","article-title":"Dual-Modal Fusion Based\non Peptide Sequence and Molecular Graph with Bidirectional Cross-Attention\nfor Precise Peptide Toxicity Prediction","volume":"66","author":"Li","year":"2026","journal-title":"J.\nChem. Inf. Model."},{"issue":"5","key":"2026092810071875200_cit8","doi-asserted-by":"publisher","first-page":"371","DOI":"10.1038\/nrd3975","article-title":"Platforms\nfor Antibiotic Discovery","volume":"12","author":"Lewis","year":"2013","journal-title":"Nat. Rev. Drug Discovery"},{"issue":"7885","key":"2026092810071875200_cit9","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1038\/s41586-021-04045-6","article-title":"A Synthetic Antibiotic\nClass Overcoming Bacterial Multidrug Resistance","volume":"599","author":"Mitcheltree","year":"2021","journal-title":"Nature"},{"issue":"3","key":"2026092810071875200_cit10","doi-asserted-by":"publisher","first-page":"946","DOI":"10.1021\/acsinfecdis.5c00988","article-title":"Recent Advances in Pleuromutilin\nDerivatives\nas Antimicrobial Agents (2015\u20132025)","volume":"12","author":"Liu","year":"2026","journal-title":"ACS Infect. Dis."},{"key":"2026092810071875200_cit11","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2020.603615","article-title":"Indicator\nRegularized Non-Negative Matrix Factorization\nMethod-Based Drug Repurposing for COVID-19","volume":"11","author":"Tang","year":"2021","journal-title":"Front. Immunol."},{"issue":"13","key":"2026092810071875200_cit12","doi-asserted-by":"publisher","DOI":"10.1002\/advs.202412987","article-title":"A Multi-Task\nSelf-Supervised\nStrategy for Predicting Molecular Properties and FGFR1 Inhibitors","volume":"12","author":"Yang","year":"2025","journal-title":"Adv. Sci."},{"issue":"4","key":"2026092810071875200_cit13","doi-asserted-by":"publisher","first-page":"688","DOI":"10.1016\/j.cell.2020.01.021","article-title":"A Deep Learning Approach\nto Antibiotic Discovery","volume":"180","author":"Stokes","year":"2020","journal-title":"Cell"},{"issue":"7997","key":"2026092810071875200_cit14","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1038\/s41586-023-06887-8","article-title":"Discovery of a Structural\nClass of Antibiotics with Explainable Deep Learning","volume":"626","author":"Wong","year":"2024","journal-title":"Nature"},{"issue":"21","key":"2026092810071875200_cit15","doi-asserted-by":"publisher","first-page":"5962","DOI":"10.1016\/j.cell.2025.07.033","article-title":"A Generative Deep Learning\nApproach to de Novo Antibiotic Design","volume":"188","author":"Krishnan","year":"2025","journal-title":"Cell"},{"key":"2026092810071875200_cit16","doi-asserted-by":"publisher","first-page":"1154","DOI":"10.1038\/s41587-025-02814-6","article-title":"Deep-Learning-Based\nVirtual Screening of Antibacterial Compounds","volume":"44","author":"Scalia","year":"2026","journal-title":"Nat. Biotechnol."},{"issue":"7","key":"2026092810071875200_cit17","doi-asserted-by":"publisher","first-page":"854","DOI":"10.1038\/s41551-024-01201-x","article-title":"Deep-Learning-Enabled\nAntibiotic Discovery through Molecular de-Extinction","volume":"8","author":"Wan","year":"2024","journal-title":"Nat. Biomed. Eng."},{"key":"2026092810071875200_cit18","unstructured":"Zhou, G.; Janarthanan, S.; Lu, Y.; et al.\n          CL-MFAP: A Contrastive\nLearning-Based Multimodal Foundation Model for Molecular Property\nPrediction and Antibiotic Screening\n          arXiv, preprint arXiv:2502.11001, 2025."},{"issue":"12","key":"2026092810071875200_cit19","doi-asserted-by":"publisher","first-page":"6855","DOI":"10.1021\/acs.jcim.5c03071","article-title":"MPMFMol: Multitask Self-Supervised\nPretraining with Multimodal Fine-Tuning for Molecular Property Prediction","volume":"66","author":"Xia","year":"2026","journal-title":"J. Chem. Inf. Model."},{"issue":"3","key":"2026092810071875200_cit20","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1038\/s42256-022-00447-x","article-title":"Molecular\nContrastive Learning of Representations via Graph Neural Networks","volume":"4","author":"Wang","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"2026092810071875200_cit21","first-page":"4849","article-title":"Multi-Objective\nMolecule Generation Using Interpretable Substructures","volume-title":"International Conference on Machine Learning","author":"Jin","year":"2020"},{"issue":"2","key":"2026092810071875200_cit22","doi-asserted-by":"publisher","first-page":"1766","DOI":"10.1021\/acs.jmedchem.5c03551","article-title":"Multimodal Multi-Granularity\nFusion\nModel with Mamba Architecture for Ames Mutagenicity Prediction","volume":"69","author":"Han","year":"2026","journal-title":"J. Med. Chem."},{"key":"2026092810071875200_cit23","doi-asserted-by":"publisher","DOI":"10.34133\/research.0869","article-title":"E-SegNet: E-Shaped Structure\nNetworks for Accurate 2D and 3D Medical Image Segmentation","volume":"8","author":"Wu","year":"2025","journal-title":"Research"},{"key":"2026092810071875200_cit24","doi-asserted-by":"publisher","first-page":"471","DOI":"10.1007\/s12539-025-00742-2","article-title":"AMPred-MFG: Investigating\nthe Mutagenicity of Compounds Using Motif-Based Graph Combined with\nMolecular Fingerprints and Graph Attention Mechanism","volume":"18","author":"Hu","year":"2026","journal-title":"Interdiscip. Sci.:Comput. Life Sci."},{"key":"2026092810071875200_cit25","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2026.113431","article-title":"MFDL-DDI: An Effective\nDeep Learning-Based\nFramework for Predicting Drug-Drug Interactions through Multimodal\nInformation Fusion","volume":"178","author":"Xia","year":"2026","journal-title":"Pattern Recognit."},{"issue":"1","key":"2026092810071875200_cit26","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-91190-x","article-title":"A Machine\nLearning Method for Predicting Molecular\nAntimicrobial Activity","volume":"15","author":"Lin","year":"2025","journal-title":"Sci. Rep."},{"key":"2026092810071875200_cit27","doi-asserted-by":"publisher","DOI":"10.1093\/jac\/dkag006","article-title":"A Disproportionality Analysis of Adverse\nEvents Associated with Omadacycline Based on the FDA Adverse Event\nReporting System Database","volume":"81","author":"Zhao","year":"2026","journal-title":"J. Antimicrob. Chemother."},{"key":"2026092810071875200_cit28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cbpc.2026.110608","article-title":"Reproductive Toxicity\nof Benzo[\u03b1]Pyrene: Effects, Mechanisms, and Perspectives","volume":"308","author":"Wang","year":"2026","journal-title":"Comp. Biochem. Physiol., Part C:Toxicol. Pharmacol."},{"issue":"34","key":"2026092810071875200_cit29","doi-asserted-by":"publisher","DOI":"10.1002\/advs.202523984","article-title":"INB3P: A Multi-Modal and Interpretable Co-Attention\nFramework Integrating Property-Aware Explanations and Memory-Bank\nContrastive Fusion for Blood\u2013Brain Barrier Penetrating Peptide\nDiscovery","volume":"13","author":"Lv","year":"2026","journal-title":"Adv. Sci."},{"issue":"23","key":"2026092810071875200_cit30","doi-asserted-by":"publisher","first-page":"12799","DOI":"10.1021\/acs.jcim.5c02280","article-title":"Multimodal Feature Fusion\nfor Bone Toxicity Prediction and Local Platform","volume":"65","author":"Pan","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026092810071875200_cit31","doi-asserted-by":"publisher","first-page":"1080","DOI":"10.1007\/s12539-025-00768-6","article-title":"hERG-MFFGNN:\nAn Explainable Deep Learning Model for Predicting Cardiotoxicity Using\nMulti-Feature Fusion and Graph Neural Networks","volume":"18","author":"Jin","year":"2026","journal-title":"Interdiscip. Sci.: Comput. Life Sci."},{"issue":"1","key":"2026092810071875200_cit32","doi-asserted-by":"publisher","first-page":"796","DOI":"10.1021\/acs.jcim.5c02664","article-title":"HetGAT-LMI: Interpretable\nHeterogeneous Graph Attention Method for Predicting lncRNA\u2013miRNA\nInteractions","volume":"66","author":"Liu","year":"2026","journal-title":"J. Chem. Inf. Model."},{"key":"2026092810071875200_cit33","doi-asserted-by":"publisher","first-page":"1276","DOI":"10.1609\/aaai.v40i2.37100","article-title":"Generalizable Drug\u2013Target\nInteraction Prediction via ESM-2 Representations and Progressive Contrastive\nCurriculum Learning","volume":"40","author":"Wu","year":"2026","journal-title":"Proc. AAAI Conf. Artif.\nIntell."},{"key":"2026092810071875200_cit34","unstructured":"Gu, A.; Dao, T.\n          Mamba: Linear-Time Sequence\nModeling with Selective State Spaces\n          arXiv, preprint arXiv:2312.00752, 2023."},{"key":"2026092810071875200_cit35","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1609\/aaai.v39i1.32009","article-title":"Mol-Mamba: Enhancing Molecular Representation\nwith Structural & Electronic Insights","volume":"39","author":"Hu","year":"2025","journal-title":"Proc.\nAAAI Conf. Artif. Intell."},{"key":"2026092810071875200_cit36","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1214\/12-EJS668","article-title":"Regularized K-Means Clustering of\nHigh-Dimensional Data and Its Asymptotic Consistency","volume":"6","author":"Sun","year":"2012","journal-title":"Electron. J. Stat."},{"key":"2026092810071875200_cit37","unstructured":"Hu, W.; Liu, B.; Gomes, J.; et al.\n          Strategies for Pre-Training\nGraph Neural Networks\n          arXiv, preprintarXiv:1905.12265, 2019."},{"key":"2026092810071875200_cit38","unstructured":"Landrum, G.\n          \n          Rdkit\nDocumentation.\nRelease. 2013; Vol. 1(1\u201379), , p 4."},{"issue":"5","key":"2026092810071875200_cit39","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-Connectivity Fingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J. Chem.\nInf. Model."},{"issue":"2","key":"2026092810071875200_cit40","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1021\/ci00046a002","article-title":"Atom Pairs as Molecular Features\nin Structure-Activity Studies: Definition and Applications","volume":"25","author":"Carhart","year":"1985","journal-title":"J. Chem. Inf. Comput. Sci."},{"issue":"2","key":"2026092810071875200_cit41","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1021\/ci00054a008","article-title":"Topological\nTorsion: A New Molecular Descriptor for SAR Applications. Comparison\nwith Other Descriptors","volume":"27","author":"Nilakantan","year":"1987","journal-title":"J. Chem. Inf. Comput.\nSci."},{"issue":"46","key":"2026092810071875200_cit42","doi-asserted-by":"publisher","first-page":"56225","DOI":"10.1021\/acsomega.5c07964","article-title":"Multimodal Cross-Attention\nMolecular Property Prediction for Text, Sequence, Graph, and Geometry","volume":"10","author":"Sun","year":"2025","journal-title":"ACS Omega"},{"key":"2026092810071875200_cit43","unstructured":"Kingma, D. P.; Ba, J.\n          Adam: A Method for Stochastic\nOptimization\n          arXiv, preprint Adam:\narXiv:1412.6980, 2014."},{"key":"2026092810071875200_cit44","unstructured":"Loshchilov, I.; Hutter, F.\n          Sgdr:\nStochastic Gradient Descent with Warm Restarts\n          arXiv, preprint arXiv:1608.03983, 2016."},{"issue":"3","key":"2026092810071875200_cit45","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0118432","article-title":"The Precision-Recall Plot Is More\nInformative than\nthe ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets","volume":"10","author":"Saito","year":"2015","journal-title":"PLoS One"},{"issue":"8","key":"2026092810071875200_cit46","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."},{"key":"2026092810071875200_cit47","doi-asserted-by":"crossref","unstructured":"Xia, J.; Zhao, C.; Hu, B.; et al.\n          Mole-Bert:\nRethinking Pre-Training\nGraph Neural Networks for Molecules; The Eleventh International\nConference on Learning Representations, ICLR: Kigali, Rwanda, 2023.","DOI":"10.26434\/chemrxiv-2023-dngg4"},{"key":"2026092810071875200_cit48","unstructured":"Ahmad, W.; Simon, E.; Chithrananda, S.; et al.\n          Chemberta-2: Towards\nChemical Foundation Models\n          arXiv,\npreprint arXiv:2209.01712, 2022."},{"issue":"6","key":"2026092810071875200_cit49","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbac408","article-title":"FP-GNN: A Versatile\nDeep Learning Architecture for Enhanced Molecular Property Prediction","volume":"23","author":"Cai","year":"2022","journal-title":"Briefings Bioinf."},{"key":"2026092810071875200_cit50","doi-asserted-by":"crossref","unstructured":"McInnes, L.; Healy, J.; Melville, J.\n          Umap: Uniform Manifold Approximation\nand Projection for Dimension Reduction\n          arXiv, preprint arXiv:1802.03426, 2018.","DOI":"10.21105\/joss.00861"}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c02659","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/18\/11793\/68474529\/acs.jcim.6c02659.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/18\/11793\/68474529\/acs.jcim.6c02659.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T14:29:37Z","timestamp":1790605777000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/18\/11793\/5427833\/A-Task-Adaptive-Multimodal-Pretrained-Framework"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,10]]},"references-count":50,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9,10]]},"published-print":{"date-parts":[[2026,9,28]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c02659","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,10]]}}}