{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Chen D"],"funding":["Soft Science Research Project of Guangdong Province","Bristol-Myers Squibb","National Institutes of Health (NIH)","Pfizer","Shenzhen Science and Technology Research Grant","Chemistry and Chemical Engineering Guangdong Laboratory","NIGMS NIH HHS","National Science Foundation"],"pagination":["10793-10801"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC9358546"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["12(44)"],"pubmed_abstract":["The construction of appropriate representations remains essential for molecular predictions due to intricate molecular complexity. Additionally, it is often expensive and ethically constrained to generate labeled data for supervised learning in molecular sciences, leading to challenging small and diverse data sets. In this work, we develop a self-supervised learning approach to pretrain models from over 700 million unlabeled molecules in multiple databases. The intrinsic chemical logic learned from this approach enables the extraction of predictive representations from task-specific molecular sequences in a fine-tuned process. To understand the importance of self-supervised learning from unlabeled molecules, we assemble three models with different combinations of databases. Moreover, we pr"],"journal":["The journal of physical chemistry letters"],"pubmed_title":["Extracting Predictive Representations from Hundreds of Millions of Molecules."],"pmcid":["PMC9358546"],"funding_grant_id":["DMS1761320","IIS1900473","No.1922018","DMS-2052983","GM126189","JCYJ20200109140416788","2017B030301013","R01 GM126189"],"pubmed_authors":["Wei GW","Zheng J","Pan F","Chen D"],"additional_accession":[]},"is_claimable":false,"name":"Extracting Predictive Representations from Hundreds of Millions of Molecules.","description":"The construction of appropriate representations remains essential for molecular predictions due to intricate molecular complexity. Additionally, it is often expensive and ethically constrained to generate labeled data for supervised learning in molecular sciences, leading to challenging small and diverse data sets. In this work, we develop a self-supervised learning approach to pretrain models from over 700 million unlabeled molecules in multiple databases. The intrinsic chemical logic learned from this approach enables the extraction of predictive representations from task-specific molecular sequences in a fine-tuned process. To understand the importance of self-supervised learning from unlabeled molecules, we assemble three models with different combinations of databases. Moreover, we pr","dates":{"release":"2021-01-01T00:00:00Z","publication":"2021 Nov","modification":"2025-04-05T08:56:15.284Z","creation":"2025-04-05T08:56:15.284Z"},"accession":"S-EPMC9358546","cross_references":{"pubmed":["34723543"],"doi":["10.1021/acs.jpclett.1c03058"]}}