{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T09:18:10Z","timestamp":1788167890199,"version":"build-2803163510"},"reference-count":45,"publisher":"American Chemical Society (ACS)","issue":"16","license":[{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62372342"],"award-info":[{"award-number":["62372342"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003819","name":"Natural Science Foundation of Hubei Province","doi-asserted-by":"publisher","award":["2024AFB865"],"award-info":[{"award-number":["2024AFB865"]}],"id":[{"id":"10.13039\/501100003819","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Drug\u2013drug interactions (DDIs) are modulated not only by structural relations between drugs but are also profoundly influenced by underlying biological mechanisms. Most existing methods fail to adequately consider biologically interpretable signals at multiple biological hierarchies, limiting their ability to model the mechanistic complexity of drug\u2013drug interactions. In this paper, we propose a heterogeneous semantic-aware framework (HSAF-DDI) that integrates molecular motif, protein sequence, and knowledge graph representations for DDI prediction. This framework learns local functional semantic representations of molecular motifs, fine-grained target protein semantics that capture deep biological characteristics, and higher-order associations encoded in knowledge graphs. We design a hierarchical adaptive fusion module that facilitates robust fusion and adaptive representation learning over multisource heterogeneous information. In addition, we introduce a contrastive learning mechanism with adversarial negatives and perturbations to improve the discrimination of DDI types and enhance the discriminability and robustness of learned representations. Experiments demonstrate that HSAF-DDI achieves superior overall performance compared to state-of-the-art methods, indicating the critical role of fine-grained biological features in improving the DDI prediction performance.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c01086","type":"journal-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T14:03:17Z","timestamp":1785852197000},"page":"10379-10395","source":"Crossref","is-referenced-by-count":0,"title":["HSAF-DDI: Heterogeneous\nSemantic-Aware Drug\u2013Drug\nInteraction Prediction with Hierarchical Adaptive Fusion"],"prefix":"10.1021","volume":"66","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5595-4966","authenticated-orcid":true,"given":"Xiaoli","family":"Lin","sequence":"first","affiliation":[{"name":"Wuhan University of Science and Technology , , , ,","place":["Wuhan, Hubei, China, 430065"]},{"name":"Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System , ,","place":["Wuhan, China, 430065"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-0447-842X","authenticated-orcid":true,"given":"Siyuan","family":"Zhang","sequence":"additional","affiliation":[{"name":"Wuhan University of Science and Technology , , , ,","place":["Wuhan, Hubei, China, 430065"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1009-7195","authenticated-orcid":true,"given":"Bo","family":"Li","sequence":"additional","affiliation":[{"name":"Wuhan University of Science and Technology , , , ,","place":["Wuhan, Hubei, China, 430065"]},{"name":"Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System , ,","place":["Wuhan, China, 430065"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5499-4758","authenticated-orcid":true,"given":"Ping","family":"Liang","sequence":"additional","affiliation":[{"name":"Wuhan University of Science and Technology , , , ,","place":["Wuhan, Hubei, China, 430065"]},{"name":"Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System , ,","place":["Wuhan, China, 430065"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8193-6956","authenticated-orcid":true,"given":"Jun","family":"Pang","sequence":"additional","affiliation":[{"name":"Wuhan University of Science and Technology , , , ,","place":["Wuhan, Hubei, China, 430065"]},{"name":"Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System , ,","place":["Wuhan, China, 430065"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,7,31]]},"reference":[{"key":"2026083104345859400_cit1","doi-asserted-by":"publisher","first-page":"1213","DOI":"10.1038\/nbt.3052","article-title":"A community computational challenge to predict the\nactivity of pairs of compounds","volume":"32","author":"Bansal","year":"2014","journal-title":"Nat. Biotechnol."},{"key":"2026083104345859400_cit2","doi-asserted-by":"publisher","first-page":"1678","DOI":"10.1016\/j.cell.2017.11.009","article-title":"Combination cancer\ntherapy can confer benefit via patient-to-patient\nvariability without drug additivity or synergy","volume":"171","author":"Palmer","year":"2017","journal-title":"Cell"},{"key":"2026083104345859400_cit3","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1021\/acs.jcim.3c01304","article-title":"Comprehensive\nreview of drug\u2013drug interaction prediction based on machine\nlearning: current status, challenges, and opportunities","volume":"64","author":"Wang","year":"2024","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104345859400_cit4","doi-asserted-by":"publisher","first-page":"9800","DOI":"10.1021\/acs.jcim.5c01773","article-title":"MGRL-DDI:\nMultiview Graph Representation Learning for Accurate Drug\u2013Drug\nInteraction Prediction","volume":"65","author":"Xiong","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104345859400_cit5","doi-asserted-by":"publisher","first-page":"961","DOI":"10.1109\/TBDATA.2025.3536924","article-title":"Hierarchical multi-relational\ngraph representation learning for large-scale prediction of drug-drug\ninteractions","volume":"11","author":"Jiang","year":"2025","journal-title":"IEEE Transactions on Big Data"},{"key":"2026083104345859400_cit6","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2023.102640","article-title":"DDI-GCN: Drug-drug\ninteraction prediction via explainable graph convolutional networks","volume":"144","author":"Zhong","year":"2023","journal-title":"Artif. Intell. Med."},{"key":"2026083104345859400_cit7","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-025-06304-z","article-title":"Contrastive learning-based multi-mechanism disentangled\nassessment\nfor drug-drug interaction","volume":"26","author":"Zhang","year":"2025","journal-title":"BMC Bioinf."},{"key":"2026083104345859400_cit8","doi-asserted-by":"publisher","first-page":"8889","DOI":"10.1021\/acs.jcim.5c01171","article-title":"A Molecular Representation\nLearning Model Based on Multidimensional Joint and Cross-Learning\nfor Drug\u2013Drug Interaction Prediction","volume":"65","author":"Chen","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104345859400_cit9","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btad451","article-title":"iGRLDTI: an improved graph representation\nlearning method for predicting\ndrug\u2013target interactions over heterogeneous biological information\nnetwork","volume":"39","author":"Zhao","year":"2023","journal-title":"Bioinformatics"},{"issue":"3","key":"2026083104345859400_cit10","doi-asserted-by":"publisher","DOI":"10.1016\/j.isci.2024.109148","article-title":"Drug-drug\ninteractions prediction based on deep learning and knowledge\ngraph: A review","volume":"27","author":"Luo","year":"2024","journal-title":"Iscience"},{"key":"2026083104345859400_cit11","doi-asserted-by":"publisher","first-page":"2988","DOI":"10.1093\/bioinformatics\/btab207","article-title":"SumGNN: multi-typed\ndrug interaction prediction via efficient knowledge\ngraph summarization","volume":"37","author":"Yu","year":"2021","journal-title":"Bioinformatics"},{"issue":"9","key":"2026083104345859400_cit12","doi-asserted-by":"publisher","DOI":"10.3390\/molecules27093004","article-title":"Prediction of drug-drug\ninteraction using an attention-based\ngraph neural network on drug molecular graphs","volume":"27","author":"Feng","year":"2022","journal-title":"Molecules"},{"key":"2026083104345859400_cit13","doi-asserted-by":"publisher","first-page":"1102","DOI":"10.1609\/aaai.v39i1.32097","article-title":"Motif-oriented representation learning\nwith topology refinement for\ndrug-drug interaction prediction","volume":"39","author":"Zhang","year":"2025","journal-title":"Proc. AAAI\nConf. Artif. Intell."},{"key":"2026083104345859400_cit14","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1089\/cmb.2024.0476","article-title":"BiRNN-DDI: A drug-drug\ninteraction\nevent type prediction model based on bidirectional recurrent neural\nnetwork and Graph2Seq representation","volume":"32","author":"Wang","year":"2025","journal-title":"J. Comput.\nBiol."},{"key":"2026083104345859400_cit15","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1609\/aaai.v38i1.27777","article-title":"Dual-channel learning\nframework for drug-drug interaction prediction via relation-aware\nheterogeneous graph transformer","volume":"38","author":"Su","year":"2024","journal-title":"Proc. AAAI\nConf. Artif. Intell."},{"key":"2026083104345859400_cit16","doi-asserted-by":"publisher","first-page":"329","DOI":"10.1609\/aaai.v38i1.27786","article-title":"Multilevel attention network with\nsemi-supervised domain\nadaptation for drug-target prediction","volume":"38","author":"Xie","year":"2024","journal-title":"Proc.\nAAAI Conf. Artif. Intell."},{"key":"2026083104345859400_cit17","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbab421","article-title":"MDF-SA-DDI: predicting drug\u2013drug\ninteraction events based on multi-source drug fusion, multi-source\nfeature fusion and transformer self-attention mechanism","volume":"23","author":"Lin","year":"2022","journal-title":"Briefings Bioinf."},{"key":"2026083104345859400_cit18","doi-asserted-by":"publisher","DOI":"10.1038\/s43856-024-00486-y","article-title":"Accurate and\ninterpretable drug-drug\ninteraction prediction enabled by knowledge subgraph learning","volume":"4","author":"Wang","year":"2024","journal-title":"Commun. Med."},{"key":"2026083104345859400_cit19","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.10718","article-title":"Multitask dyadic prediction\nand its application in prediction of\nadverse drug-drug interaction","volume":"31","author":"Jin","year":"2017","journal-title":"Proc. AAAI Conf.\nArtif. Intell."},{"key":"2026083104345859400_cit20","doi-asserted-by":"publisher","first-page":"E4304","DOI":"10.1073\/pnas.1803294115","article-title":"Deep learning\nimproves prediction\nof drug\u2013drug and drug\u2013food interactions","volume":"115","author":"Ryu","year":"2018","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"key":"2026083104345859400_cit21","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1007\/978-981-97-5575-2_16","article-title":"Hierarchical Cross-Level Graph Contrastive Learning for Drug-Drug\nInteraction Prediction","volume-title":"International Conference\non Database Systems for Advanced Applications","author":"Ye","year":"2024"},{"key":"2026083104345859400_cit22","doi-asserted-by":"publisher","first-page":"13927","DOI":"10.1609\/aaai.v34i10.7236","article-title":"Structure-based drug-drug\ninteraction detection via expressive graph convolutional networks\nand deep sets (student abstract)","volume":"34","author":"Sun","year":"2020","journal-title":"Proc. AAAI\nConf. Artif. Intell."},{"key":"2026083104345859400_cit23","unstructured":"Feeney, A.; Gupta, R.; Thost, V.; Angell, R.; Chandu, G.; Adhikari, Y.; Ma, T.\n          Relation matters in sampling: a scalable multi-relational\ngraph neural network for drug-drug interaction prediction. 2021https:\/\/arxiv.org\/abs\/2105.13975. arXiv:2105.13975. arXiv.org\ne-Print archive."},{"key":"2026083104345859400_cit24","first-page":"2739","article-title":"KGNN: Knowledge\ngraph neural network for drug-drug interaction prediction","volume":"380","author":"Lin","year":"2020","journal-title":"IJCAI"},{"key":"2026083104345859400_cit25","doi-asserted-by":"crossref","unstructured":"Wang, H.; Lian, D.; Zhang, Y.; Qin, L.; Lin, X.\n          Gognn: Graph of\ngraphs neural network for predicting\nstructured entity interactions. 2020, https:\/\/arxiv.org\/abs\/2005.05537. arXiv:2005.05537. arXiv.org e-Print archive.","DOI":"10.24963\/ijcai.2020\/183"},{"key":"2026083104345859400_cit26","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbab511","article-title":"Predicting drug\u2013drug interactions\nby graph convolutional network with multi-kernel","volume":"23","author":"Wang","year":"2022","journal-title":"Briefings Bioinf."},{"key":"2026083104345859400_cit27","doi-asserted-by":"publisher","first-page":"5339","DOI":"10.1609\/aaai.v37i4.25665","article-title":"Multi-relational\ncontrastive learning graph neural network for drug-drug\ninteraction event prediction","volume":"37","author":"Xiong","year":"2023","journal-title":"Proc. AAAI Conf.\nArtif. Intell."},{"key":"2026083104345859400_cit28","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad385","article-title":"HetDDI: a pre-trained\nheterogeneous\ngraph neural network model for drug\u2013drug interaction prediction","volume":"24","author":"Li","year":"2023","journal-title":"Briefings Bioinf."},{"key":"2026083104345859400_cit29","doi-asserted-by":"crossref","unstructured":"Zhang, R.; Wang, X.; Wang, S.; Liu, K.; Zhou, Y.; Wang, P.; et al. In  Proceedings of the 33rd ACM International\nConference\non Information and Knowledge Management, H2D: hierarchical\nheterogeneous graph learning framework for drug-drug interaction prediction; ACM, 2024; pp 4283\u20134287.","DOI":"10.1145\/3627673.3679936"},{"key":"2026083104345859400_cit30","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-017-0200-8","article-title":"Predicting drug\u2013drug\ninteractions through drug structural similarities and interaction\nnetworks incorporating pharmacokinetics and pharmacodynamics knowledge","volume":"9","author":"Takeda","year":"2017","journal-title":"J. Cheminf."},{"key":"2026083104345859400_cit31","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2021.103696","article-title":"Drug repurposing\nfor COVID-19 via knowledge graph completion","volume":"115","author":"Zhang","year":"2021","journal-title":"J. Biomed. Inf."},{"key":"2026083104345859400_cit32","doi-asserted-by":"publisher","first-page":"2651","DOI":"10.1093\/bioinformatics\/btab169","article-title":"MUFFIN: multi-scale\nfeature fusion for drug\u2013drug interaction prediction","volume":"37","author":"Chen","year":"2021","journal-title":"Bioinformatics"},{"key":"2026083104345859400_cit33","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2023.111337","article-title":"SRR-DDI: a drug\u2013drug\ninteraction prediction model with substructure refined representation\nlearning based on self-attention mechanism","volume":"285","author":"Niu","year":"2024","journal-title":"Knowl.-Based Syst."},{"key":"2026083104345859400_cit34","doi-asserted-by":"publisher","first-page":"3784","DOI":"10.1021\/acs.jcim.2c00457","article-title":"Hotspot identification\nand drug design of protein\u2013protein\ninteraction modulators using the fragment molecular orbital method","volume":"62","author":"Monteleone","year":"2022","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104345859400_cit35","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.124647","article-title":"Drug\u2013target\nbinding affinity prediction model based on multi-scale diffusion and\ninteractive learning","volume":"255","author":"Zhu","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"2026083104345859400_cit36","doi-asserted-by":"publisher","first-page":"2102","DOI":"10.1093\/bioinformatics\/btac020","article-title":"ProteinBERT: a universal deep-learning\nmodel of protein\nsequence and function","volume":"38","author":"Brandes","year":"2022","journal-title":"Bioinformatics"},{"key":"2026083104345859400_cit37","unstructured":"Ioannidis, V. N.; Song, X.; Manchanda, S.; Li, M.; Pan, X.; Zheng, D.; Ning, X.; Zeng, X.; Karypis, G.\n          Drkg-drug\nrepurposing knowledge graph for covid-19. 2020, https:\/\/arxiv.org\/pdf\/2010.09600. arXiv:2010.09600. arXiv.org e-Print archive."},{"key":"2026083104345859400_cit38","doi-asserted-by":"publisher","first-page":"4316","DOI":"10.1093\/bioinformatics\/btaa501","article-title":"A multimodal\ndeep learning framework for predicting drug\u2013drug interaction\nevents","volume":"36","author":"Deng","year":"2020","journal-title":"Bioinformatics"},{"key":"2026083104345859400_cit39","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbab133","article-title":"SSI\u2013DDI:\nsubstructure\u2013substructure interactions\nfor drug\u2013drug interaction prediction","volume":"22","author":"Nyamabo","year":"2021","journal-title":"Briefings Bioinf."},{"key":"2026083104345859400_cit40","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbac597","article-title":"DSN-DDI:\nan accurate and generalized framework for drug\u2013drug interaction\nprediction by dual-view representation learning","volume":"24","author":"Li","year":"2023","journal-title":"Briefings Bioinf."},{"key":"2026083104345859400_cit41","doi-asserted-by":"publisher","first-page":"1004","DOI":"10.1038\/s42256-022-00557-6","article-title":"Accurate prediction\nof molecular\nproperties and drug targets using a self-supervised image representation\nlearning framework","volume":"4","author":"Zeng","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"2026083104345859400_cit42","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad404","article-title":"Chemical structure-aware\nmolecular image representation learning","volume":"24","author":"Xiang","year":"2023","journal-title":"Briefings\nBioinf."},{"key":"2026083104345859400_cit43","doi-asserted-by":"crossref","unstructured":"Du, H.; Yao, Q.; Zhang, J.; Liu, Y.; Wang, Z. In  Customized subgraph selection and encoding\nfor drug-drug\ninteraction prediction, Advances in Neural Information Processing\nSystems 37; NeurIPS, 2024; pp 109582\u2013109608.","DOI":"10.52202\/079017-3478"},{"key":"2026083104345859400_cit44","doi-asserted-by":"crossref","unstructured":"Ma, T.; Chen, K.; Zang, Y.; Chen, Y.; Ren, X.; Song, B.; Xiang, H.; Liu, Y.; Zeng, X. In  Self-supervised Blending\nStructural\nContext of Visual Molecules for Robust Drug Interaction Prediction, Thirty-ninth Annual Conference on Neural Information Processing\nSystems; NeurIPS, 2025.","DOI":"10.52202\/085713-4435"},{"key":"2026083104345859400_cit45","doi-asserted-by":"publisher","first-page":"9999","DOI":"10.1021\/acs.jcim.5c01467","article-title":"Predicting Drug-Drug\nInteraction via Dual\u2013Drug Visual Representation","volume":"65","author":"Xie","year":"2025","journal-title":"J. Chem. Inf. Model."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c01086","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/16\/10379\/66346811\/acs.jcim.6c01086.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/16\/10379\/66346811\/acs.jcim.6c01086.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T08:35:45Z","timestamp":1788165345000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/16\/10379\/5237653\/HSAF-DDI-Heterogeneous-Semantic-Aware-Drug-Drug"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,31]]},"references-count":45,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,7,31]]},"published-print":{"date-parts":[[2026,8,24]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c01086","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,31]]}}}