{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T22:42:55Z","timestamp":1792795375971,"version":"4.5.0"},"reference-count":30,"publisher":"American Chemical Society (ACS)","issue":"8","license":[{"start":{"date-parts":[[2020,7,23]],"date-time":"2020-07-23T00:00:00Z","timestamp":1595462400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2020,7,23]],"date-time":"2020-07-23T00:00:00Z","timestamp":1595462400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2020,7,23]],"date-time":"2020-07-23T00:00:00Z","timestamp":1595462400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100000268","name":"Biotechnology and Biological Sciences Research Council","doi-asserted-by":"publisher","award":["BB\/S507611\/1"],"award-info":[{"award-number":["BB\/S507611\/1"]}],"id":[{"id":"10.13039\/501100000268","id-type":"DOI","asserted-by":"publisher"}]},{"name":"BenevolentAI"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,8,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Current deep learning methods for structure-based virtual screening take the structures of both the protein and the ligand as input but make little or no use of the protein structure when predicting ligand binding. Here, we show how a relatively simple method of data set augmentation forces such deep learning methods to take into account information from the protein. Models trained in this way are more generalizable (make better predictions on protein\/ligand complexes from a different distribution to the training data). They also assign more meaningful importance to the protein and ligand atoms involved in binding. Overall, our results show that data set augmentation can help deep learning-based virtual screening to learn physical interactions rather than data set biases.<\/jats:p>","DOI":"10.1021\/acs.jcim.0c00263","type":"journal-article","created":{"date-parts":[[2020,7,23]],"date-time":"2020-07-23T11:40:29Z","timestamp":1595504429000},"page":"3722-3730","source":"Crossref","is-referenced-by-count":53,"title":["Data Set Augmentation Allows Deep Learning-Based Virtual\nScreening to Better Generalize to Unseen Target Classes and Highlight\nImportant Binding Interactions"],"prefix":"10.1021","volume":"60","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9877-0107","authenticated-orcid":true,"given":"Jack","family":"Scantlebury","sequence":"first","affiliation":[{"name":"University of Oxford , , 24\u221229 St Giles , ,","place":["Oxford, U.K., OX1 3LB"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9243-8699","authenticated-orcid":true,"given":"Nathan","family":"Brown","sequence":"additional","affiliation":[{"name":"BenevolentAI , 4\u22128 Maple Street , ,","place":["London, U.K., W1T 5HD"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0378-0017","authenticated-orcid":true,"given":"Frank","family":"Von Delft","sequence":"additional","affiliation":[{"name":"University of Oxford , , ,","place":["Oxford, U.K., OX3 7DQ"]},{"name":"Diamond Light Source , Harwell Science and Innovation Campus , ,","place":["Didcot, U.K., OX11 0DE"]},{"name":"University of Johannesburg , , Aukland Park , ,","place":["Johannesburg, South\nAfrica, 2006"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1388-2252","authenticated-orcid":true,"given":"Charlotte M.","family":"Deane","sequence":"additional","affiliation":[{"name":"University of Oxford , , 24\u221229 St Giles , ,","place":["Oxford, U.K., OX1 3LB"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2020,7,23]]},"reference":[{"key":"2026081805075815200_cit1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0220113","article-title":"Hidden bias in the\nDUD-E dataset leads to misleading performance of deep learning in\nstructure-based virtual screening","volume":"14","author":"Chen","year":"2019","journal-title":"PLoS One"},{"key":"2026081805075815200_cit2","doi-asserted-by":"publisher","first-page":"947","DOI":"10.1021\/acs.jcim.8b00712","article-title":"In Need of Bias Control: Evaluating\nChemical Data for Machine Learning in Structure-Based Virtual Screening","volume":"59","author":"Sieg","year":"2019","journal-title":"J. Chem. Inf. Model."},{"key":"2026081805075815200_cit3","doi-asserted-by":"publisher","first-page":"916","DOI":"10.1021\/acs.jcim.7b00403","article-title":"Most Ligand-Based Classification\nBenchmarks Reward\nMemorization Rather than Generalization","volume":"58","author":"Wallach","year":"2018","journal-title":"J.\nChem. Inf. Model."},{"key":"2026081805075815200_cit4","doi-asserted-by":"publisher","first-page":"1375","DOI":"10.3390\/ijms20061375","article-title":"The Light\nand Dark Sides of Virtual Screening: What Is There to Know?","volume":"20","author":"Gimeno","year":"2019","journal-title":"Int. J. Mol. Sci."},{"key":"2026081805075815200_cit5","doi-asserted-by":"publisher","first-page":"2839","DOI":"10.2174\/09298673113209990001","article-title":"Virtual Screening Strategies\nin Drug Discovery: A Critical\nReview","volume":"20","author":"Lavecchia","year":"2013","journal-title":"Curr. Med. Chem."},{"key":"2026081805075815200_cit6","doi-asserted-by":"publisher","first-page":"12964","DOI":"10.1039\/c6cp01555g","article-title":"Comprehensive evaluation\nof ten docking\nprograms on a diverse set of protein\u2013ligand complexes: the\nprediction accuracy of sampling power and scoring power","volume":"18","author":"Wang","year":"2016","journal-title":"Phys. Chem. Chem. Phys."},{"key":"2026081805075815200_cit7","doi-asserted-by":"publisher","first-page":"3347","DOI":"10.2174\/1381612824666180607124038","article-title":"Machine Learning-based\nVirtual Screening and Its Applications to\nAlzheimer\u2019s Drug Discovery: A Review","volume":"24","author":"Carpenter","year":"2018","journal-title":"Curr. Pharm. Des."},{"key":"2026081805075815200_cit8","doi-asserted-by":"publisher","first-page":"253","DOI":"10.1016\/j.compbiomed.2017.09.007","article-title":"Interaction prediction\nin structure-based virtual screening\nusing deep learning","volume":"100","author":"Gonczarek","year":"2018","journal-title":"Comput. Biol. Med."},{"key":"2026081805075815200_cit9","doi-asserted-by":"publisher","first-page":"2495","DOI":"10.1021\/acs.jcim.6b00355","article-title":"Boosting Docking-Based Virtual Screening\nwith Deep Learning","volume":"56","author":"Pereira","year":"2016","journal-title":"J. Chem. Inf. Model."},{"key":"2026081805075815200_cit10","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1038\/s41573-019-0024-5","article-title":"Applications\nof machine learning in drug discovery and development","volume":"18","author":"Vamathevan","year":"2019","journal-title":"Nat. Rev. Drug Discovery"},{"key":"2026081805075815200_cit11","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1007129","article-title":"DeepConv-DTI: Prediction of drug-target\ninteractions via deep learning with convolution on protein sequences","volume":"15","author":"Lee","year":"2019","journal-title":"PLoS Comput. Biol."},{"key":"2026081805075815200_cit12","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1021\/acscentsci.6b00367","article-title":"Low Data Drug Discovery\nwith One-Shot Learning","volume":"3","author":"Altae-Tran","year":"2017","journal-title":"ACS Cent. Sci."},{"key":"2026081805075815200_cit13","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1093\/bioinformatics\/bty535","article-title":"Compound\u2013protein\ninteraction\nprediction with end-to-end learning of neural networks for graphs\nand sequences","volume":"35","author":"Tsubaki","year":"2018","journal-title":"Bioinformatics"},{"key":"2026081805075815200_cit14","doi-asserted-by":"publisher","first-page":"758","DOI":"10.1093\/bioinformatics\/btz665","article-title":"Learning from the\nligand: using ligand-based\nfeatures to improve binding affinity prediction","volume":"36","author":"Boyles","year":"2019","journal-title":"Bioinformatics"},{"key":"2026081805075815200_cit15","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning\napplied to document recognition","volume":"86","author":"Lecun","year":"1998","journal-title":"Proc. IEEE"},{"key":"2026081805075815200_cit16","doi-asserted-by":"publisher","first-page":"942","DOI":"10.1021\/acs.jcim.6b00740","article-title":"Protein\u2013Ligand\nScoring with Convolutional Neural Networks","volume":"57","author":"Ragoza","year":"2017","journal-title":"J. Chem. Inf. Model."},{"key":"2026081805075815200_cit17","doi-asserted-by":"publisher","first-page":"6582","DOI":"10.1021\/jm300687e","article-title":"Directory of Useful\nDecoys, Enhanced (DUD-E): Better Ligands and Decoys for Better Benchmarking","volume":"55","author":"Mysinger","year":"2012","journal-title":"J. Med. Chem."},{"key":"2026081805075815200_cit18","doi-asserted-by":"publisher","first-page":"2319","DOI":"10.1021\/acs.jcim.8b00350","article-title":"Protein Family-Specific\nModels Using Deep Neural Networks and Transfer Learning Improve Virtual\nScreening and Highlight the Need for More Data","volume":"58","author":"Imrie","year":"2018","journal-title":"J. Chem. Inf. Model."},{"key":"2026081805075815200_cit19","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1038\/s42004-020-0261-x","article-title":"Elucidating the\nmultiple roles of hydration for accurate protein-ligand binding prediction\nvia deep learning","volume":"3","author":"Mahmoud","year":"2020","journal-title":"Chem. Commun."},{"key":"2026081805075815200_cit20","doi-asserted-by":"publisher","first-page":"D945","DOI":"10.1093\/nar\/gkw1074","article-title":"The ChEMBL\ndatabase\nin 2017","volume":"45","author":"Gaulton","year":"2016","journal-title":"Nucleic Acids Res."},{"key":"2026081805075815200_cit21","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1016\/j.jmgm.2018.06.005","article-title":"Visualizing\nconvolutional neural network protein-ligand scoring","volume":"84","author":"Hochuli","year":"2018","journal-title":"J. Mol. Graphics Modell."},{"key":"2026081805075815200_cit22","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1186\/1758-2946-5-17","article-title":"The influence of\nthe inactives subset\ngeneration on the performance of machine learning methods","volume":"5","author":"Smusz","year":"2013","journal-title":"J. Cheminf."},{"key":"2026081805075815200_cit23","doi-asserted-by":"crossref","DOI":"10.26434\/chemrxiv.8139194.v1","article-title":"Debiasing Algorithms\nfor Protein Ligand Binding Data do not Improve Generalisation","author":"Sundar","year":"2019"},{"key":"2026081805075815200_cit24","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1002\/jcc.21334","article-title":"AutoDock Vina: Improving the speed and accuracy of\ndocking with a new scoring function, efficient optimization, and multithreading","volume":"31","author":"Trott","year":"2010","journal-title":"J. Comput. Chem."},{"key":"2026081805075815200_cit25","doi-asserted-by":"crossref","unstructured":"Zaman, A. B.; Shehu, A.\n          Equipping Decoy\nGeneration Algorithms for Template-free Protein Structure Prediction\nwith Maps of the Protein Conformation Space. Proceedings of 11th International Conference on Bioinformatics and\nComputational Biology, 2019; pp 161\u2013169.","DOI":"10.29007\/j5p9"},{"key":"2026081805075815200_cit26","doi-asserted-by":"publisher","first-page":"1831","DOI":"10.1021\/ci200199u","article-title":"Large-Scale Similarity Search Profiling of ChEMBL Compound\nData Sets","volume":"51","author":"Heikamp","year":"2011","journal-title":"J. Chem. Inf. Model."},{"key":"2026081805075815200_cit27","doi-asserted-by":"publisher","first-page":"1079","DOI":"10.1021\/acs.jcim.9b01145","article-title":"libmolgrid: Graphics Processing Unit Accelerated Molecular\nGridding for Deep Learning Applications","volume":"60","author":"Sunseri","year":"2020","journal-title":"J.\nChem. Inf. Model."},{"key":"2026081805075815200_cit28","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1007\/s00894-016-3119-5","article-title":"A method\nfor predicting individual residue contributions to enzyme specificity\nand binding-site energies, and its application to MTH1","volume":"22","author":"Stewart","year":"2016","journal-title":"J. Mol. Model."},{"key":"2026081805075815200_cit29","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1093\/nar\/28.1.235","article-title":"The Protein\nData Bank","volume":"28","author":"Berman","year":"2000","journal-title":"Nucleic Acids Res."},{"key":"2026081805075815200_cit30","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0118432","article-title":"The Precision-Recall Plot Is More Informative than\nthe ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets","volume":"10","author":"Saito","year":"2015","journal-title":"PLoS One"}],"container-title":["Journal of Chemical Information and Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.0c00263","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/60\/8\/3722\/9372789\/ci0c00263.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/60\/8\/3722\/9372789\/ci0c00263.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T10:23:28Z","timestamp":1787048608000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/60\/8\/3722\/874154\/Data-Set-Augmentation-Allows-Deep-Learning-Based"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,7,23]]},"references-count":30,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2020,7,23]]},"published-print":{"date-parts":[[2020,8,24]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.0c00263","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2020.03.06.979625","asserted-by":"object"}]},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,7,23]]}}}