{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T20:09:42Z","timestamp":1775851782332,"version":"3.50.1"},"reference-count":50,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2023,8,20]],"date-time":"2023-08-20T00:00:00Z","timestamp":1692489600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61972100"],"award-info":[{"award-number":["61972100"]}],"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":["62172300"],"award-info":[{"award-number":["62172300"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,9,20]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Molecular property prediction (MPP) is a crucial and fundamental task for AI-aided drug discovery (AIDD). Recent studies have shown great promise of applying self-supervised learning (SSL) to producing molecular representations to cope with the widely-concerned data scarcity problem in AIDD. As some specific substructures of molecules play important roles in determining molecular properties, molecular representations learned by deep learning models are expected to attach more importance to such substructures implicitly or explicitly to achieve better predictive performance. However, few SSL pre-trained models for MPP in the literature have ever focused on such substructures. To challenge this situation, this paper presents a Chemistry-Aware Fragmentation for Effective MPP (CAFE-MPP in short) under the self-supervised contrastive learning framework. First, a novel fragment-based molecular graph (FMG) is designed to represent the topological relationship between chemistry-aware substructures that constitute a molecule. Then, with well-designed hard negative pairs, a is pre-trained on fragment-level by contrastive learning to extract representations for the nodes in FMGs. Finally, a Graphormer model is leveraged to produce molecular representations for MPP based on the embeddings of fragments. Experiments on 11 benchmark datasets show that the proposed CAFE-MPP method achieves state-of-the-art performance on 7 of the 11 datasets and the second-best performance on 3 datasets, compared with six remarkable self-supervised methods. Further investigations also demonstrate that CAFE-MPP can learn to embed molecules into representations implicitly containing the information of fragments highly correlated to molecular properties, and can alleviate the over-smoothing problem of graph neural networks.<\/jats:p>","DOI":"10.1093\/bib\/bbad296","type":"journal-article","created":{"date-parts":[[2023,8,20]],"date-time":"2023-08-20T21:23:03Z","timestamp":1692566583000},"source":"Crossref","is-referenced-by-count":18,"title":["Self-supervised learning with chemistry-aware fragmentation for effective molecular property prediction"],"prefix":"10.1093","volume":"24","author":[{"given":"Ailin","family":"Xie","sequence":"first","affiliation":[{"name":"Shanghai Key Lab of Intelligent Information Processing , and School of Computer Science, , 200438 Shanghai , China"},{"name":"Fudan University , and School of Computer Science, , 200438 Shanghai , China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ziqiao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Shanghai Key Lab of Intelligent Information Processing , and School of Computer Science, , 200438 Shanghai , China"},{"name":"Fudan University , and School of Computer Science, , 200438 Shanghai , China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jihong","family":"Guan","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tongji University , 201804 Shanghai , China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuigeng","family":"Zhou","sequence":"additional","affiliation":[{"name":"Shanghai Key Lab of Intelligent Information Processing , and School of Computer Science, , 200438 Shanghai , China"},{"name":"Fudan University , and School of Computer Science, , 200438 Shanghai , China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2023,8,20]]},"reference":[{"key":"2023092216550665700_ref1","doi-asserted-by":"crossref","first-page":"103850","DOI":"10.1016\/j.chemolab.2019.103850","article-title":"Artificial intelligence facilitates drug design in the big data era","volume":"194","author":"Wang","year":"2019","journal-title":"Chemom Intel Lab Syst"},{"key":"2023092216550665700_ref2","doi-asserted-by":"crossref","first-page":"606668","DOI":"10.3389\/fphar.2020.606668","article-title":"Improvement of prediction performance with conjoint molecular fingerprint in deep learning","volume":"11","author":"Xie","year":"2020","journal-title":"Front Pharmacol"},{"issue":"8","key":"2023092216550665700_ref3","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":"Yang","year":"2019","journal-title":"J Chem Inf Model"},{"issue":"1","key":"2023092216550665700_ref4","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1021\/ci00057a005","article-title":"Smiles, a chemical language and information system. 1. Introduction to methodology and encoding rules","volume":"28","author":"Weininger","year":"1988","journal-title":"J Chem Inf Comput Sci"},{"issue":"12","key":"2023092216550665700_ref5","doi-asserted-by":"crossref","first-page":"2105","DOI":"10.1351\/pac200476122105","article-title":"Chemical actinometry (IUPAC technical report)","volume":"76","author":"Kuhn","year":"2004","journal-title":"Pure Appl Chem"},{"key":"2023092216550665700_ref6","first-page":"1263","article-title":"Neural message passing for quantum chemistry","volume-title":"Proceedings of the 34th International Conference on Machine Learning (ICML)","author":"","year":"2017"},{"issue":"16","key":"2023092216550665700_ref7","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":"2023092216550665700_ref8","doi-asserted-by":"crossref","first-page":"2831","DOI":"10.24963\/ijcai.2020\/392","article-title":"Communicative representation learning on attributed molecular graphs","volume-title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","author":"","year":"2020"},{"issue":"3\u20134","key":"2023092216550665700_ref9","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1016\/S1359-6446(05)03692-5","article-title":"Optimizing the use of open-source software applications in drug discovery","volume":"11","author":"Geldenhuys","year":"2006","journal-title":"Drug Discov Today"},{"issue":"8","key":"2023092216550665700_ref10","doi-asserted-by":"crossref","first-page":"675","DOI":"10.1007\/s10822-013-9672-4","article-title":"Estimation of the size of drug-like chemical space based on gdb-17 data","volume":"27","author":"Polishchuk","year":"2013","journal-title":"J Comput Aided Mol Des"},{"key":"2023092216550665700_ref11","article-title":"Graph self-supervised learning: a survey","volume":"35","author":"Liu","year":"2023","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"2023092216550665700_ref12","doi-asserted-by":"crossref","first-page":"429","DOI":"10.1145\/3307339.3342186","article-title":"Smiles-bert: large scale unsupervised pre-training for molecular property prediction","volume-title":"Proceedings of the 10th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","author":"","year":"2019"},{"key":"2023092216550665700_ref13","article-title":"Mol-BERT: an effective molecular representation with bert for molecular property prediction","volume":"2021","author":"Li","year":"2021","journal-title":"Wireless Commun Mobile Comput."},{"key":"2023092216550665700_ref14","first-page":"1597","article-title":"A simple framework for contrastive learning of visual representations","volume-title":"International Conference on Machine Learning","author":"","year":"2020"},{"issue":"2\u20133","key":"2023092216550665700_ref15","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1504\/IJCBDD.2014.061649","article-title":"Generic strategies for chemical space exploration","volume":"7","author":"Andersen","year":"2014","journal-title":"Int J Comput Biol Drug Des"},{"issue":"5","key":"2023092216550665700_ref16","doi-asserted-by":"crossref","first-page":"3894","DOI":"10.1021\/acs.energyfuels.6b00303","article-title":"Calculation of average molecular parameters, functional groups, and a surrogate molecule for heavy fuel oils using 1h and 13c nuclear magnetic resonance spectroscopy","volume":"30","author":"Jameel","year":"2016","journal-title":"Energy Fuel"},{"issue":"5","key":"2023092216550665700_ref17","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"},{"issue":"1","key":"2023092216550665700_ref18","first-page":"1","article-title":"Comparing structural fingerprints using a literature-based similarity benchmark","volume":"8","author":"O\u2019Boyle","year":"2016","journal-title":"J Chem"},{"issue":"18","key":"2023092216550665700_ref19","doi-asserted-by":"crossref","first-page":"2981","DOI":"10.1093\/bioinformatics\/btab195","article-title":"Fragat: a fragment-oriented multi-scale graph attention model for molecular property prediction","volume":"37","author":"Zhang","year":"2021","journal-title":"Bioinformatics"},{"key":"2023092216550665700_ref20","first-page":"28877","article-title":"Do transformers really perform badly for graph representation?","volume":"34","author":"Ying","year":"2021","journal-title":"Adv Neural Inform Process Syst"},{"key":"2023092216550665700_ref21","first-page":"11773","article-title":"Graph neural networks inspired by classical iterative algorithms","volume-title":"Proceedings of the 38th International Conference on Machine Learning","author":"","year":"2021"},{"key":"2023092216550665700_ref22","article-title":"Chemberta: large-scale self-supervised pretraining for molecular property prediction","author":"Chithrananda","year":"2020","journal-title":"arXiv preprint arXiv:201009885"},{"issue":"2","key":"2023092216550665700_ref23","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1039\/C7SC02664A","article-title":"Moleculenet: a benchmark for molecular machine learning","volume":"9","author":"Zhenqin","year":"2018","journal-title":"Chem Sci"},{"issue":"14","key":"2023092216550665700_ref24","doi-asserted-by":"crossref","first-page":"1753","DOI":"10.4155\/fmc-2016-0095","article-title":"On the origins of three-dimensionality in drug-like molecules","volume":"8","author":"Meyers","year":"2016","journal-title":"Future Med Chem"},{"key":"2023092216550665700_ref25","doi-asserted-by":"crossref","first-page":"2713","DOI":"10.1021\/acs.jcim.2c00495","article-title":"Improving molecular contrastive learning via faulty negative mitigation and decomposed fragment contrast","volume":"62","author":"Wang","year":"2022","journal-title":"J Chem Inf Model"},{"issue":"10","key":"2023092216550665700_ref26","doi-asserted-by":"crossref","first-page":"1503","DOI":"10.1002\/cmdc.200800178","article-title":"On the art of compiling and using\u2019drug-like\u2019chemical fragment spaces","volume":"3","author":"Degen","year":"2008","journal-title":"ChemMedChem: Chemistry Enabling Drug Discovery"},{"key":"2023092216550665700_ref27","first-page":"15870","article-title":"Motif-based graph self-supervised learning for molecular property prediction","volume":"34","author":"Zhang","year":"2021","journal-title":"Adv Neural Inform Process Syst"},{"issue":"3","key":"2023092216550665700_ref28","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":"2023092216550665700_ref29","first-page":"3441","article-title":"Multilingual molecular representation learning via contrastive pre-training","volume-title":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers)","author":"","year":"2022"},{"key":"2023092216550665700_ref30","first-page":"20479","article-title":"3d infomax improves gnns for molecular property prediction","volume-title":"Proceedings of the 39th International Conference on Machine Learning","author":"","year":"2022"},{"key":"2023092216550665700_ref31","first-page":"1558","article-title":"Molecular graph contrastive learning with parameterized explainable augmentations","volume-title":"In: 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","author":"","year":"2021"},{"key":"2023092216550665700_ref32","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbac303","article-title":"Attention-wise masked graph contrastive learning for predicting molecular property","volume":"23","author":"Liu","year":"2022","journal-title":"Briefings in Bioinformatics"},{"key":"2023092216550665700_ref33","doi-asserted-by":"crossref","first-page":"6697","DOI":"10.1021\/acs.chemrev.1c00469","article-title":"Introduction: click chemistry","volume":"121","author":"Devaraj","year":"2021","journal-title":"Chem Rev"},{"issue":"7","key":"2023092216550665700_ref34","doi-asserted-by":"crossref","first-page":"4286","DOI":"10.1021\/acscatal.6b00758","article-title":"Building bridges: biocatalytic c\u2013c-bond formation toward multifunctional products","volume":"6","author":"Schmidt","year":"2016","journal-title":"ACS Catal"},{"issue":"9","key":"2023092216550665700_ref35","doi-asserted-by":"crossref","first-page":"4385","DOI":"10.1021\/acs.jmedchem.6b00153","article-title":"Big data from pharmaceutical patents: a computational analysis of medicinal chemists\u2019 bread and butter","volume":"59","author":"Schneider","year":"2016","journal-title":"J Med Chem"},{"issue":"5","key":"2023092216550665700_ref36","doi-asserted-by":"crossref","first-page":"487","DOI":"10.1016\/j.progpolymsci.2006.03.001","article-title":"Michael addition reactions in macromolecular design for emerging technologies","volume":"31","author":"Mather","year":"2006","journal-title":"Prog Polym Sci"},{"key":"2023092216550665700_ref37","article-title":"The directed aldol reaction","author":"Mukaiyama","year":"1982","journal-title":"Organic React"},{"issue":"22","key":"2023092216550665700_ref38","doi-asserted-by":"crossref","first-page":"2462","DOI":"10.1002\/anie.200200558","article-title":"Modulation of protein-protein interactions with small organic molecules","volume":"42","author":"Berg","year":"2003","journal-title":"Angew Chem Int Ed"},{"issue":"7019","key":"2023092216550665700_ref39","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1038\/nature03196","article-title":"Exploring biology with small organic molecules","volume":"432","author":"Stockwell","year":"2004","journal-title":"Nature"},{"key":"2023092216550665700_ref40","first-page":"203","article-title":"Rdkit: open-source cheminformatics","author":"Landrum","year":"2006"},{"key":"2023092216550665700_ref41","article-title":"Contrastive learning with hard negative samples","author":"Robinson","year":"2020","journal-title":"arXiv preprint arXiv:201004592"},{"issue":"2","key":"2023092216550665700_ref42","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1038\/s42256-021-00438-4","article-title":"Geometry-enhanced molecular representation learning for property prediction","volume":"4","author":"Fang","year":"2022","journal-title":"Nat Mach Intell"},{"key":"2023092216550665700_ref43","first-page":"9323","article-title":"E(n) equivariant graph neural networks","volume-title":"International Conference on Machine Learning","author":"","year":"2021"},{"key":"2023092216550665700_ref44","article-title":"Understanding over-squashing and bottlenecks on graphs via curvature","author":"Topping","year":"2022","journal-title":"International Conference on Learning Representations"},{"issue":"15","key":"2023092216550665700_ref45","doi-asserted-by":"crossref","first-page":"2887","DOI":"10.1021\/jm9602928","article-title":"The properties of known drugs. 1. Molecular frameworks","volume":"39","author":"Bemis","year":"1996","journal-title":"J Med Chem"},{"key":"2023092216550665700_ref46","article-title":"Fast graph representation learning with pytorch geometric","author":"Fey","year":"2019","journal-title":"arXiv preprint arXiv:190302428"},{"key":"2023092216550665700_ref47","article-title":"N-gram graph: simple unsupervised representation for graphs, with applications to molecules","volume":"32","author":"Liu","year":"2019","journal-title":"Adv Neural Inform Process Syst"},{"key":"2023092216550665700_ref48","first-page":"11548","article-title":"Self-supervised graph-level representation learning with local and global structure","volume-title":"Proceedings of the 38th International Conference on Machine Learning","author":"Minghao","year":"2021"},{"key":"2023092216550665700_ref49","first-page":"12559","article-title":"Self-supervised graph transformer on large-scale molecular data","volume":"33","author":"Rong","year":"2020","journal-title":"Adv Neural Inform Process Syst"},{"issue":"86","key":"2023092216550665700_ref50","article-title":"Visualizing data using t-sne","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J Mach Learn Res"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/24\/5\/bbad296\/51710638\/bbad296.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/24\/5\/bbad296\/51710638\/bbad296.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,22]],"date-time":"2023-09-22T17:17:16Z","timestamp":1695403036000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbad296\/7246472"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,20]]},"references-count":50,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2023,9,20]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbad296","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,9]]},"published":{"date-parts":[[2023,8,20]]},"article-number":"bbad296"}}