{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T11:59:05Z","timestamp":1781783945434,"version":"3.54.5"},"reference-count":45,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2022,8,6]],"date-time":"2022-08-06T00:00:00Z","timestamp":1659744000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,9,20]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Accurate and efficient prediction of the molecular property is one of the fundamental problems in drug research and development. Recent advancements in representation learning have been shown to greatly improve the performance of molecular property prediction. However, due to limited labeled data, supervised learning-based molecular representation algorithms can only search limited chemical space and suffer from poor generalizability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In this work, we proposed a self-supervised learning method, ATMOL, for molecular representation learning and properties prediction. We developed a novel molecular graph augmentation strategy, referred to as attention-wise graph masking, to generate challenging positive samples for contrastive learning. We adopted the graph attention network as the molecular graph encoder, and leveraged the learned attention weights as masking guidance to generate molecular augmentation graphs. By minimization of the contrastive loss between original graph and augmented graph, our model can capture important molecular structure and higher order semantic information. Extensive experiments showed that our attention-wise graph mask contrastive learning exhibited state-of-the-art performance in a couple of downstream molecular property prediction tasks. We also verified that our model pretrained on larger scale of unlabeled data improved the generalization of learned molecular representation. Moreover, visualization of the attention heatmaps showed meaningful patterns indicative of atoms and atomic groups important to specific molecular property.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bib\/bbac303","type":"journal-article","created":{"date-parts":[[2022,8,8]],"date-time":"2022-08-08T20:08:48Z","timestamp":1659989328000},"source":"Crossref","is-referenced-by-count":47,"title":["Attention-wise masked graph contrastive learning for predicting molecular property"],"prefix":"10.1093","volume":"23","author":[{"given":"Hui","family":"Liu","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology , , 211816, Nanjing , China"},{"name":"Nanjing Tech University , , 211816, Nanjing , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yibiao","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering , ,410075, Changsha, China"},{"name":"Central South University , ,410075, Changsha, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuejun","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology , , 211816, Nanjing , China"},{"name":"Nanjing Tech University , , 211816, Nanjing , China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering , ,410075, Changsha, China"},{"name":"Central South University , ,410075, Changsha, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,8,6]]},"reference":[{"issue":"4","key":"2022092013221740200_ref1","doi-asserted-by":"crossref","first-page":"622","DOI":"10.18632\/oncotarget.938","article-title":"Combinatorial drug screening identifies compensatory pathway interactions and adaptive resistance mechanisms","volume":"4","author":"Axelrod","year":"2013","journal-title":"Oncotarget"},{"issue":"5","key":"2022092013221740200_ref2","doi-asserted-by":"crossref","first-page":"579","DOI":"10.1093\/bib\/bbp023","article-title":"Recent advances in computer-aided drug design","volume":"10","author":"Song","year":"2009","journal-title":"Brief Bioinform"},{"issue":"18","key":"2022092013221740200_ref3","doi-asserted-by":"crossref","DOI":"10.1103\/PhysRevB.87.184115","article-title":"On representing chemical environments","volume":"87","author":"Bartok","year":"2013","journal-title":"Physical Review B"},{"issue":"10","key":"2022092013221740200_ref4","doi-asserted-by":"crossref","DOI":"10.1103\/PhysRevLett.114.105503","article-title":"Big data of materials science: critical role of the descriptor","volume":"114","author":"Ghiringhelli","year":"2015","journal-title":"Phys Rev Lett"},{"issue":"1","key":"2022092013221740200_ref5","article-title":"Molecular representations in AI-driven drug discovery: a review and practical guide","volume":"12","author":"David","year":"2020","journal-title":"J Chem"},{"issue":"12","key":"2022092013221740200_ref6","doi-asserted-by":"crossref","first-page":"5646","DOI":"10.1016\/j.ejmech.2010.09.018","article-title":"Characteristics of known drug space. natural products, their derivatives and synthetic drugs","volume":"45","author":"Bade","year":"2010","journal-title":"Eur J Med Chem"},{"key":"2022092013221740200_ref7","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1016\/j.ymeth.2014.08.005","article-title":"Molecular fingerprint similarity search in virtual screening","volume":"71","author":"Cereto-Massague","year":"2015","journal-title":"Methods"},{"issue":"5","key":"2022092013221740200_ref8","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":"2","key":"2022092013221740200_ref9","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1109\/TNN.2010.2091281","article-title":"Domain adaptation via transfer component analysis","volume":"22","author":"Pan","year":"2011","journal-title":"IEEE Trans Neural Netw"},{"key":"2022092013221740200_ref10","first-page":"9729","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"He","year":"2020"},{"key":"2022092013221740200_ref11","article-title":"Semi-supervised classification with graph convolutional networks","volume-title":"International Conference on Learning Representations (ICLR 2017)","author":"Welling","year":"2016"},{"key":"2022092013221740200_ref12","article-title":"Convolutional networks on graphs for learning molecular fingerprints","volume":"28","author":"Duvenaudt","year":"2015","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2022092013221740200_ref13","first-page":"70","article-title":"Neural message passing for quantum chemistry","volume":"70","author":"Gilmer","year":"2017","journal-title":"International Conference on Machine Learning"},{"issue":"9","key":"2022092013221740200_ref14","doi-asserted-by":"crossref","DOI":"10.1103\/PhysRevMaterials.4.093801","article-title":"Orbital graph convolutional neural network for material property prediction","volume":"4","author":"Karamad","year":"2020","journal-title":"Physical Review Materials"},{"key":"2022092013221740200_ref15","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-018-06169-2","article-title":"Towards exact molecular dynamics simulations with machine-learned force fields","volume":"9","author":"Chmiela","year":"2018","journal-title":"Nat Commun"},{"issue":"11","key":"2022092013221740200_ref16","doi-asserted-by":"crossref","first-page":"2879","DOI":"10.1021\/acs.jpclett.8b00902","article-title":"Realistic atomistic structure of amorphous silicon from machine-learning-driven molecular dynamics","volume":"9","author":"Deringer","year":"2018","journal-title":"Journal of Physical Chemistry Letters"},{"issue":"1","key":"2022092013221740200_ref17","doi-asserted-by":"crossref","DOI":"10.1038\/s41524-019-0261-5","article-title":"Coarse-graining auto-encoders for molecular dynamics","volume":"5","author":"Wang","year":"2019","journal-title":"Npj Computational Materials"},{"issue":"4","key":"2022092013221740200_ref18","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1021\/acscentsci.6b00367","article-title":"Low data drug discovery with one-shot learning","volume":"3","author":"Altae-Tran","year":"2017","journal-title":"ACS Cent Sci"},{"issue":"6","key":"2022092013221740200_ref19","doi-asserted-by":"crossref","first-page":"1241","DOI":"10.1016\/j.drudis.2018.01.039","article-title":"The rise of deep learning in drug discovery","volume":"23","author":"Chen","year":"2018","journal-title":"Drug Discov Today"},{"key":"2022092013221740200_ref20","doi-asserted-by":"crossref","first-page":"731","DOI":"10.1145\/3394486.3403117","volume-title":"Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","author":"Hao","year":"2020"},{"key":"2022092013221740200_ref21","first-page":"2014","article-title":"How transferable are features in deep neural networks","volume":"27","author":"Yosinski","journal-title":"Advances in Neural Information Processing Systems 27 (Nips 2014)"},{"key":"2022092013221740200_ref22","first-page":"22243","article-title":"Big self-supervised models are strong semi-supervised learners","volume":"33","author":"Chen","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"2022092013221740200_ref23","first-page":"119","article-title":"A simple framework for contrastive learning of visual representations","volume":"119","author":"Chen","year":"2020","journal-title":"International Conference on Machine Learning"},{"key":"2022092013221740200_ref24","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"issue":"6","key":"2022092013221740200_ref25","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab152","article-title":"MG-BERT: leveraging unsupervised atomic representation learning for molecular property prediction","volume":"22","author":"Zhang","year":"2021","journal-title":"Brief Bioinform"},{"issue":"9","key":"2022092013221740200_ref26","doi-asserted-by":"crossref","DOI":"10.1021\/acs.jcim.1c00600","article-title":"Molgpt: Molecular generation using a transformer-decoder model","volume":"62","author":"Bagal","year":"2021","journal-title":"J Chem Inf Model"},{"key":"2022092013221740200_ref27","article-title":"Molclr: molecular contrastive learning of representations via graph neural networks","author":"Wang","year":"2021"},{"key":"2022092013221740200_ref28","volume-title":"International Conference on Database Systems for Advanced Applications","author":"Song","year":"2022"},{"issue":"6","key":"2022092013221740200_ref29","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab109","article-title":"An effective self-supervised framework for learning expressive molecular global representations to drug discovery","volume":"22","author":"Li","year":"2021","journal-title":"Brief Bioinform"},{"key":"2022092013221740200_ref30","doi-asserted-by":"crossref","DOI":"10.1101\/2021.01.13.426608","article-title":"MoTSE: an interpretable task similarity estimator for small molecular property prediction tasks","author":"Li","year":"2021"},{"key":"2022092013221740200_ref31","article-title":"Multi-view graph neural networks for molecular property prediction","author":"Ma","year":"2020"},{"key":"2022092013221740200_ref32","article-title":"RDKit: A software suite for cheminformatics, computational chemistry, and predictive modeling","author":"Landrum","year":"2013","journal-title":"Greg Landrum"},{"issue":"2","key":"2022092013221740200_ref33","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":"2022092013221740200_ref34","volume-title":"Deep learning for the life sciences: applying deep learning to genomics, microscopy, drug discovery, and more","author":"Ramsundar","year":"2019"},{"key":"2022092013221740200_ref35","first-page":"1074","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Hu","year":"2021"},{"key":"2022092013221740200_ref36","volume-title":"Advances in Neural Information Processing Systems","author":"Suresh","year":"2021"},{"key":"2022092013221740200_ref37","article-title":"Strategies for pre-training graph neural networks","author":"Hu","year":"2019"},{"key":"2022092013221740200_ref38","volume-title":"Advances in neural information processing systems","author":"Liu","year":"2019"},{"key":"2022092013221740200_ref39","first-page":"12559","article-title":"Self-supervised graph transformer on large-scale molecular data","volume":"33","author":"Rong","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2022092013221740200_ref40","article-title":"Motif-based graph self-supervised learning for molecular property prediction","volume":"34","author":"Zhang","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2022092013221740200_ref41","article-title":"Umap: uniform manifold approximation and projection for dimension reduction","author":"McInnes","year":"2018"},{"issue":"6","key":"2022092013221740200_ref42","doi-asserted-by":"crossref","first-page":"1686","DOI":"10.1021\/ci300124c","article-title":"A bayesian approach to in silico blood-brain barrier penetration modeling","volume":"52","author":"Martins","year":"2012","journal-title":"J Chem Inf Model"},{"issue":"10","key":"2022092013221740200_ref43","doi-asserted-by":"crossref","first-page":"1936","DOI":"10.1021\/acs.jcim.6b00290","volume":"56","author":"Subramanian","year":"2016","journal-title":"J Chem Inf Model"},{"key":"2022092013221740200_ref44","first-page":"3968","article-title":"Molecular contrastive learning with chemical element knowledge graph","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Fang","year":"2022"},{"key":"2022092013221740200_ref45","first-page":"110","article-title":"Fragment-based drug discovery by NMR. Where are the successes and where cans it be improved","author":"Mureddu","year":"2022","journal-title":"Front Mol Biosci"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/23\/5\/bbac303\/45937770\/bbac303.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/23\/5\/bbac303\/45937770\/bbac303.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,20]],"date-time":"2022-09-20T13:56:45Z","timestamp":1663682205000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbac303\/6657662"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,6]]},"references-count":45,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,9,20]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbac303","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2022.05.08.491075","asserted-by":"object"}]},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,9]]},"published":{"date-parts":[[2022,8,6]]},"article-number":"bbac303"}}