{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T17:03:46Z","timestamp":1775581426737,"version":"3.50.1"},"reference-count":41,"publisher":"Oxford University Press (OUP)","issue":"10","license":[{"start":{"date-parts":[[2025,9,11]],"date-time":"2025-09-11T00:00:00Z","timestamp":1757548800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U23A20321"],"award-info":[{"award-number":["U23A20321"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62272490"],"award-info":[{"award-number":["62272490"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004761","name":"Natural Science Foundation of Hunan Province of China","doi-asserted-by":"crossref","award":["2025JJ20062"],"award-info":[{"award-number":["2025JJ20062"]}],"id":[{"id":"10.13039\/501100004761","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Accurate molecular property prediction remains a central challenge in molecular machine learning, critically dependent on comprehensive molecular representation. Existing methods, however, encounter two major limitations: (i) single-modal learning approaches frequently experience representation bottlenecks, whereas multimodal methods often struggle to effectively leverage complementary information without redundancy across modalities; and (ii) conventional data augmentation techniques typically treat atoms as isolated units, neglecting intrinsic dependencies among atoms within molecular substructures.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Here, we propose MolCL-SP, a substructure-aware multimodal contrastive learning framework specifically designed for molecular property prediction. Our approach integrates molecular representations derived from three complementary modalities using a Transformer-based encoder, followed by modality-specific reconstruction to organically align and fuse cross-modal information. We also introduce a novel substructure-based non-overlapping perturbation strategy for data augmentation, preserving interpretability and effectively enhancing inter-modal interactions. Extensive experimental evaluations demonstrate that MolCL-SP achieves state-of-the-art performance on benchmark datasets for both 2D and 3D molecular property predictions. Additionally, evaluations on drug\u2013drug interaction prediction tasks highlight the model\u2019s strong generalization capabilities. Visualization analyses further indicate that MolCL-SP effectively captures discriminative molecular embeddings even in task-agnostic contexts. Importantly, the model implicitly emphasizes chemically meaningful substructures associated with functional relevance, significantly enhancing interpretability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>Codes and materials are available at https:\/\/github.com\/lylikeeMoon\/MolCL-SP.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaf507","type":"journal-article","created":{"date-parts":[[2025,9,10]],"date-time":"2025-09-10T11:50:21Z","timestamp":1757505021000},"source":"Crossref","is-referenced-by-count":3,"title":["MolCL-SP: a multimodal contrastive learning framework with non-overlapping substructure perturbations for molecular property prediction"],"prefix":"10.1093","volume":"41","author":[{"given":"Yue","family":"Luo","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha 410083,","place":["China"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2869-1619","authenticated-orcid":false,"given":"Lei","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha 410083,","place":["China"]}]}],"member":"286","published-online":{"date-parts":[[2025,9,11]]},"reference":[{"key":"2025102813170236900_btaf507-B1","author":"Ba","year":"2016"},{"key":"2025102813170236900_btaf507-B3","doi-asserted-by":"crossref","first-page":"4316","DOI":"10.1093\/bioinformatics\/btaa501","article-title":"A multimodal deep learning framework for predicting drug\u2013drug interaction events","volume":"36","author":"Deng","year":"2020","journal-title":"Bioinformatics"},{"key":"2025102813170236900_btaf507-B4","author":"Gasteiger"},{"key":"2025102813170236900_btaf507-B5","doi-asserted-by":"crossref","first-page":"6765","DOI":"10.1039\/C3CS60460H","article-title":"Bace1 (\u03b2-secretase) inhibitors for the treatment of Alzheimer\u2019s disease","volume":"43","author":"Ghosh","year":"2014","journal-title":"Chem Soc Rev"},{"key":"2025102813170236900_btaf507-B6","author":"He"},{"key":"2025102813170236900_btaf507-B7","first-page":"1926","author":"Hou"},{"key":"2025102813170236900_btaf507-B8","author":"Hu"},{"key":"2025102813170236900_btaf507-B9","doi-asserted-by":"crossref","first-page":"2416","DOI":"10.1109\/JBHI.2024.3363080","article-title":"Hierarchical and dynamic graph attention network for drug-disease association prediction","author":"Huang","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"2025102813170236900_btaf507-B10","doi-asserted-by":"crossref","first-page":"ii190","DOI":"10.1093\/bioinformatics\/btae386","article-title":"Molmvc: enhancing molecular representations for drug-related tasks through multi-view contrastive learning","volume":"40","author":"Huang","year":"2024","journal-title":"Bioinformatics"},{"key":"2025102813170236900_btaf507-B11","doi-asserted-by":"crossref","first-page":"111329","DOI":"10.1016\/j.knosys.2023.111329","article-title":"Amdgt: attention aware multi-modal fusion using a dual graph transformer for drug\u2013disease associations prediction","volume":"284","author":"Liu","year":"2024","journal-title":"Knowl Based Syst"},{"key":"2025102813170236900_btaf507-B12","first-page":"21497","author":"Liu"},{"key":"2025102813170236900_btaf507-B13","author":"Liu"},{"key":"2025102813170236900_btaf507-B14","author":"Liu"},{"key":"2025102813170236900_btaf507-B15","doi-asserted-by":"crossref","first-page":"e1010812","DOI":"10.1371\/journal.pcbi.1010812","article-title":"A dual graph neural network for drug\u2013drug interactions prediction based on molecular structure and interactions","volume":"19","author":"Ma","year":"2023","journal-title":"PLoS Comput Biol"},{"key":"2025102813170236900_btaf507-B16","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1186\/s13024-023-00596-6","article-title":"The Alzheimer\u2019s disease-linked protease bace1 modulates neuronal il-6 signaling through shedding of the receptor gp130","volume":"18","author":"M\u00fcller","year":"2023","journal-title":"Mol Neurodegener"},{"key":"2025102813170236900_btaf507-B17","doi-asserted-by":"crossref","first-page":"bbab441","DOI":"10.1093\/bib\/bbab441","article-title":"Drug\u2013drug interaction prediction with learnable size-adaptive molecular substructures","volume":"23","author":"Nyamabo","year":"2022","journal-title":"Brief Bioinform"},{"key":"2025102813170236900_btaf507-B18","doi-asserted-by":"crossref","first-page":"bbab133","DOI":"10.1093\/bib\/bbab133","article-title":"SSI\u2013DDI: substructure\u2013substructure interactions for drug\u2013drug interaction prediction","volume":"22","author":"Nyamabo","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025102813170236900_btaf507-B19","doi-asserted-by":"crossref","first-page":"140022","DOI":"10.1038\/sdata.2014.22","article-title":"Quantum chemistry structures and properties of 134 kilo molecules","volume":"1","author":"Ramakrishnan","year":"2014","journal-title":"Sci Data"},{"key":"2025102813170236900_btaf507-B20","doi-asserted-by":"crossref","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-connectivity fingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J Chem Inf Model"},{"key":"2025102813170236900_btaf507-B21","first-page":"12559","article-title":"Self-supervised graph transformer on large-scale molecular data","volume":"33","author":"Rong","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025102813170236900_btaf507-B22","first-page":"992","article-title":"Schnet: a continuous-filter convolutional neural network for modeling quantum interactions","volume":"30","author":"Sch\u00fctt","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025102813170236900_btaf507-B23","first-page":"10435","author":"Scott"},{"key":"2025102813170236900_btaf507-B24","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1007\/s12031-007-0055-y","article-title":"The role of size and charge for blood\u2013brain barrier permeation of drugs and fatty acids","volume":"33","author":"Seelig","year":"2007","journal-title":"J Mol Neurosci"},{"key":"2025102813170236900_btaf507-B25","doi-asserted-by":"crossref","first-page":"bbae256","DOI":"10.1093\/bib\/bbae256","article-title":"Complementary multi-modality molecular self-supervised learning via non-overlapping masking for property prediction","volume":"25","author":"Shen","year":"2024","journal-title":"Brief Bioinform"},{"key":"2025102813170236900_btaf507-B26","first-page":"20479","author":"St\u00e4rk"},{"key":"2025102813170236900_btaf507-B27","author":"Sun"},{"key":"2025102813170236900_btaf507-B28","author":"Vaswani"},{"key":"2025102813170236900_btaf507-B29","first-page":"429","author":"Wang"},{"key":"2025102813170236900_btaf507-B30","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1038\/s42256-022-00447-x","article-title":"Molecular contrastive learning of representations via graph neural networks","volume":"4","author":"Wang","year":"2022","journal-title":"Nat Mach Intell"},{"key":"2025102813170236900_btaf507-B31","doi-asserted-by":"crossref","first-page":"D1074","DOI":"10.1093\/nar\/gkx1037","article-title":"Drugbank 5.0: a major update to the drugbank database for 2018","volume":"46","author":"Wishart","year":"2018","journal-title":"Nucleic Acids Res"},{"key":"2025102813170236900_btaf507-B32","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1039\/C7SC02664A","article-title":"Moleculenet: a benchmark for molecular machine learning","volume":"9","author":"Wu","year":"2018","journal-title":"Chem Sci"},{"key":"2025102813170236900_btaf507-B33","author":"Xia"},{"key":"2025102813170236900_btaf507-B2"},{"key":"2025102813170236900_btaf507-B34","first-page":"5339","author":"Xiong","year":"2023"},{"key":"2025102813170236900_btaf507-B35","doi-asserted-by":"crossref","first-page":"8749","DOI":"10.1021\/acs.jmedchem.9b00959","article-title":"Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism","volume":"63","author":"Xiong","year":"2020","journal-title":"J Med Chem"},{"key":"2025102813170236900_btaf507-B36","author":"Xu"},{"key":"2025102813170236900_btaf507-B37","author":"Yu"},{"key":"2025102813170236900_btaf507-B38","doi-asserted-by":"crossref","first-page":"5191","DOI":"10.1093\/bioinformatics\/btz418","article-title":"Deepdr: a network-based deep learning approach to in silico drug repositioning","volume":"35","author":"Zeng","year":"2019","journal-title":"Bioinformatics"},{"key":"2025102813170236900_btaf507-B39","doi-asserted-by":"crossref","first-page":"147937","DOI":"10.1016\/j.brainres.2022.147937","article-title":"Factors influencing the blood-brain barrier permeability","volume":"1788","author":"Zhao","year":"2022","journal-title":"Brain Res"},{"key":"2025102813170236900_btaf507-B40","first-page":"3615","author":"Zhu"},{"key":"2025102813170236900_btaf507-B41","first-page":"4757","author":"Zhu"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/10\/btaf507\/64245681\/btaf507.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/10\/btaf507\/64245681\/btaf507.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T17:17:21Z","timestamp":1761671841000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btaf507\/8251523"}},"subtitle":[],"editor":[{"given":"Jianlin","family":"Cheng","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2025,9,11]]},"references-count":41,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2025,10,2]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaf507","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,10]]},"published":{"date-parts":[[2025,9,11]]},"article-number":"btaf507"}}