{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T17:56:51Z","timestamp":1777485411240,"version":"3.51.4"},"reference-count":32,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T00:00:00Z","timestamp":1747008000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2023YFC2705700"],"award-info":[{"award-number":["2023YFC2705700"]}]},{"DOI":"10.13039\/501100001809","name":"Natural Science Foundation of China","doi-asserted-by":"crossref","award":["No. 62476203"],"award-info":[{"award-number":["No. 62476203"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Artificial Intelligence Innovation Project of Wuhan Science and Technology Bureau","award":["No. 2022010702040070"],"award-info":[{"award-number":["No. 2022010702040070"]}]},{"DOI":"10.13039\/100016804","name":"Natural Science Foundation of Shenzhen City","doi-asserted-by":"crossref","award":["No. JCYJ20230807090211021"],"award-info":[{"award-number":["No. JCYJ20230807090211021"]}],"id":[{"id":"10.13039\/100016804","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Molecular representation learning is pivotal in predicting molecular properties and advancing drug design. Traditional methodologies, which predominantly rely on homogeneous graph encoding, are limited by their inability to integrate external knowledge and represent molecular structures across different levels of granularity. To address these limitations, we propose a paradigm shift by encoding molecular graphs into heterogeneous structures, introducing a novel framework: Knowledge-aware Contrastive Heterogeneous Molecular Graph Learning. This approach leverages contrastive learning to enrich molecular representations with embedded external knowledge. KCHML conceptualizes molecules through three distinct graph views\u2014molecular, elemental, and pharmacological\u2014enhanced by heterogeneous molecular graphs and a dual message-passing mechanism. This design offers a comprehensive representation for property prediction, as well as for downstream tasks such as drug-drug interaction prediction. Extensive benchmarking demonstrates KCHML\u2019s superiority over state-of-the-art molecular property prediction models, underscoring its ability to capture intricate molecular features.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1013008","type":"journal-article","created":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T17:47:13Z","timestamp":1747072033000},"page":"e1013008","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Knowledge-aware contrastive heterogeneous molecular graph learning"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0511-5347","authenticated-orcid":true,"given":"Mukun","family":"Chen","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jia","family":"Wu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shirui","family":"Pan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fu","family":"Lin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bo","family":"Du","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiuwen","family":"Gong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9258-3850","authenticated-orcid":true,"given":"Wenbin","family":"Hu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2025,5,12]]},"reference":[{"key":"pcbi.1013008.ref001","first-page":"2117","article-title":"Zero-shot learning for preclinical drug screening","volume-title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24","author":"K Li","year":"2024"},{"key":"pcbi.1013008.ref002","first-page":"2126","article-title":"Contrastive learning drug response models from natural language supervision","volume-title":"Proceedings of the International Joint Conferences on Artificial Intelligence Organization","author":"K Li","year":"2024"},{"key":"pcbi.1013008.ref003","unstructured":"Kipf T, Welling M. Semi-supervised classification with graph convolutional networks. arXiv. preprint. arXiv. 2016. doi: arXiv:1609.02907"},{"key":"pcbi.1013008.ref004","unstructured":"Xu K, Hu W, Leskovec J, Jegelka S. How powerful are graph neural networks? arXiv. Preprint. 2018; arXiv:1810.00826"},{"issue":"20","key":"pcbi.1013008.ref005","first-page":"10","article-title":"Graph attention networks","volume":"1050","author":"P Velickovic","year":"2017","journal-title":"Statistics"},{"key":"pcbi.1013008.ref006","unstructured":"Li Y, Zemel R, Brockschmidt M, Tarlow D. Gated graph sequence neural networks. In: Proceedings of ICLR\u201916; 2016."},{"key":"pcbi.1013008.ref007","volume-title":"Advances in Neural Information Processing Systems","author":"W Hamilton","year":"2017"},{"key":"pcbi.1013008.ref008","unstructured":"Gilmer J, Schoenholz S, Riley P, Vinyals O, Dahl G. Neural message passing for quantum chemistry. In: Proceedings of the International Conference on Machine Learning. PMLR; 2017. pp. 1263\u201372."},{"issue":"8","key":"pcbi.1013008.ref009","doi-asserted-by":"crossref","first-page":"3370","DOI":"10.1021\/acs.jcim.9b00237","article-title":"Analyzing Learned Molecular Representations for Property Prediction","volume":"59","author":"K Yang","year":"2019","journal-title":"J Chem Inf Model"},{"key":"pcbi.1013008.ref010","first-page":"2831","article-title":"Communicative representation learning on attributed molecular graphs. In: IJCAI International Joint Conference on attributed molecular graphs","author":"Y Song","year":"2020","journal-title":"IJCAI International Joint Conference on Artificial Intelligence"},{"key":"pcbi.1013008.ref011","first-page":"2242","article-title":"Learning attributed graph representation with communicative message passing transformer","volume-title":"Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. International Joint Conferences on Artificial Intelligence Organization","author":"J Chen","year":"2021"},{"key":"pcbi.1013008.ref012","volume-title":"Pre-training graph neural networks. In: International Conference on Learning Representations (ICLR","author":"W Hu","year":"2020"},{"key":"pcbi.1013008.ref013","first-page":"12559","article-title":"Self-supervised graph transformer on large-scale molecular data. In:","author":"Y Rong","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pcbi.1013008.ref014","doi-asserted-by":"crossref","unstructured":"Lin X, Quan Z, Wang Z, Ma T, Zeng X. Kgnn: knowledge graph neural network for drug-drug interaction prediction. In: IJCAI, vol. 380. 2020. pp. 2739\u201345.","DOI":"10.24963\/ijcai.2020\/380"},{"key":"pcbi.1013008.ref015","doi-asserted-by":"crossref","unstructured":"Lyu T, Gao J, Tian L, Li Z, Zhang P, Zhang J. MDNN: A multimodal deep neural network for predicting drug-drug interaction events. In: IJCAI, 2021, pp. 3536\u201342.","DOI":"10.24963\/ijcai.2021\/487"},{"key":"pcbi.1013008.ref016","doi-asserted-by":"crossref","first-page":"10216","DOI":"10.1609\/aaai.v38i9.28887","article-title":"MKG-FENN: A Multimodal Knowledge Graph Fused End-to-End Neural Network for Accurate Drug\u2013Drug Interaction Prediction","author":"D Wu","year":"2024","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"pcbi.1013008.ref017","first-page":"5812","article-title":"Graph contrastive learning with augmentations","author":"Y You","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pcbi.1013008.ref018","first-page":"3585","article-title":"MoCL: Data-driven Molecular Fingerprint via Knowledge-aware Contrastive Learning from Molecular Graph","author":"M Sun","year":"2021","journal-title":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining"},{"issue":"1","key":"pcbi.1013008.ref019","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1038\/s42004-024-01109-2","article-title":"Molecular fragmentation as a crucial step in the AI-based drug development pathway","volume":"7","author":"S Jinsong","year":"2024","journal-title":"Commun Chem"},{"key":"pcbi.1013008.ref020","article-title":"Molecule attention transformer","author":"\u0141 Maziarka","year":"2020"},{"issue":"11","key":"pcbi.1013008.ref021","doi-asserted-by":"crossref","first-page":"2324","DOI":"10.1021\/acs.jcim.5b00559","article-title":"ZINC 15--Ligand Discovery for Everyone","volume":"55","author":"T Sterling","year":"2015","journal-title":"J Chem Inf Model"},{"key":"pcbi.1013008.ref022","article-title":"Drkg - drug repurposing knowledge graph for covid-19","author":"V Ioannidis","year":"2020"},{"issue":"2","key":"pcbi.1013008.ref023","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1039\/C7SC02664A","article-title":"MoleculeNet: a benchmark for molecular machine learning","volume":"9","author":"Z Wu","year":"2017","journal-title":"Chem Sci"},{"issue":"1","key":"pcbi.1013008.ref024","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab441","article-title":"Drug-drug interaction prediction with learnable size-adaptive molecular substructures","volume":"23","author":"AK Nyamabo","year":"2022","journal-title":"Brief Bioinform"},{"issue":"16","key":"pcbi.1013008.ref025","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":"Z Xiong","year":"2020","journal-title":"J Med Chem"},{"key":"pcbi.1013008.ref026","article-title":"N-gram graph: simple unsupervised representation for graphs, with applications to molecules","author":"S Liu","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pcbi.1013008.ref027","volume-title":"Strategies for pre-training graph neural networks. In: International Conference on Learning Representations (ICLR","author":"W Hu","year":"2020"},{"key":"pcbi.1013008.ref028","doi-asserted-by":"crossref","first-page":"3968","DOI":"10.1609\/aaai.v36i4.20313","article-title":"Molecular Contrastive Learning with Chemical Element Knowledge Graph","author":"Y Fang","year":"2022","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"issue":"18","key":"pcbi.1013008.ref029","article-title":"Deep learning improves prediction of drug-drug and drug-food interactions","volume":"115","author":"JY Ryu","year":"2018","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"29","key":"pcbi.1013008.ref030","doi-asserted-by":"crossref","first-page":"8693","DOI":"10.1039\/D2SC02023H","article-title":"Learning size-adaptive molecular substructures for explainable drug-drug interaction prediction by substructure-aware graph neural network","volume":"13","author":"Z Yang","year":"2022","journal-title":"Chem Sci"},{"issue":"1","key":"pcbi.1013008.ref031","article-title":"A dual graph neural network for drug-drug interactions prediction based on molecular structure and interactions","volume":"19","author":"M Ma","year":"2023","journal-title":"PLoS Comput Biol"},{"issue":"1","key":"pcbi.1013008.ref032","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1609\/aaai.v38i1.27777","article-title":"Dual-Channel Learning Framework for Drug-Drug Interaction Prediction via Relation-Aware Heterogeneous Graph Transformer","volume":"38","author":"X Su","year":"2024","journal-title":"AAAI"}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013008","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T17:47:25Z","timestamp":1747072045000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013008"}},"subtitle":[],"editor":[{"given":"Piero","family":"Fariselli","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2025,5,12]]},"references-count":32,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2025,5,12]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1013008","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,12]]}}}