{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T20:27:45Z","timestamp":1777926465400,"version":"3.51.4"},"reference-count":78,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2021,3,2]],"date-time":"2021-03-02T00:00:00Z","timestamp":1614643200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,3,2]],"date-time":"2021-03-02T00:00:00Z","timestamp":1614643200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2021,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Deep learning methods applied to chemistry can be used to accelerate the discovery of new molecules. This work introduces GraphINVENT, a platform developed for graph-based molecular design using graph neural networks (GNNs). GraphINVENT uses a tiered deep neural network architecture to probabilistically generate new molecules a single bond at a time. All models implemented in GraphINVENT can quickly learn to build molecules resembling the training set molecules without any explicit programming of chemical rules. The models have been benchmarked using the MOSES distribution-based metrics, showing how GraphINVENT models compare well with state-of-the-art generative models. This work compares six different GNN-based generative models in GraphINVENT, and shows that ultimately the gated-graph neural network performs best against the metrics considered here.<\/jats:p>","DOI":"10.1088\/2632-2153\/abcf91","type":"journal-article","created":{"date-parts":[[2020,12,1]],"date-time":"2020-12-01T17:46:49Z","timestamp":1606844809000},"page":"025023","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":115,"title":["Graph networks for molecular design"],"prefix":"10.1088","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6170-6088","authenticated-orcid":false,"given":"Roc\u00edo","family":"Mercado","sequence":"first","affiliation":[]},{"given":"Tobias","family":"Rastemo","sequence":"additional","affiliation":[]},{"given":"Edvard","family":"Lindel\u00f6f","sequence":"additional","affiliation":[]},{"given":"G\u00fcnter","family":"Klambauer","sequence":"additional","affiliation":[]},{"given":"Ola","family":"Engkvist","sequence":"additional","affiliation":[]},{"given":"Hongming","family":"Chen","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1614-7376","authenticated-orcid":false,"given":"Esben","family":"Jannik Bjerrum","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2021,3,2]]},"reference":[{"key":"mlstabcf91bib1","doi-asserted-by":"crossref","first-page":"30033","DOI":"10.1073\/pnas.1907373117","article-title":"The unreasonable effectiveness of deep learning in artificial intelligence","volume":"117","author":"Sejnowski","year":"2020","journal-title":"Proc. Natl. Acad. Sci."},{"key":"mlstabcf91bib2","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"mlstabcf91bib3","article-title":"Relational inductive biases, deep learning, and graph networks","author":"Battaglia","year":"2018"},{"key":"mlstabcf91bib4","doi-asserted-by":"publisher","first-page":"120","DOI":"10.1021\/acscentsci.7b00512","article-title":"Generating focused molecule libraries for drug discovery with recurrent neural networks","volume":"4","author":"Segler","year":"2018","journal-title":"ACS Cent. Sci."},{"key":"mlstabcf91bib5","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1186\/s13321-017-0235-x","article-title":"Molecular de-novo design through deep reinforcement learning","volume":"9","author":"Olivecrona","year":"2017","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib6","article-title":"Molecular generation with recurrent neural networks (RNNS)","author":"Bjerrum","year":"2017"},{"key":"mlstabcf91bib7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-019-0393-0","article-title":"Randomized smiles strings improve the quality of molecular generative models","volume":"11","author":"Ar\u00fas-Pous","year":"2019","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib8","doi-asserted-by":"publisher","DOI":"10.26434\/chemrxiv.5309668.v3","article-title":"Optimizing distributions over molecular space. an objective-reinforced generative adversarial network for inverse-design chemistry (organic)","author":"Sanchez-Lengeling","year":"2017","journal-title":"ChemRxiv"},{"key":"mlstabcf91bib9","article-title":"Learning deep generative models of graphs","author":"Li","year":"2018"},{"key":"mlstabcf91bib10","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1186\/s13321-018-0287-6","article-title":"Multi-objective de novo drug design with conditional graph generative model","volume":"10","author":"Li","year":"2018","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib11","article-title":"Junction tree variational autoencoder for molecular graph generation","author":"Jin","year":"2018"},{"key":"mlstabcf91bib12","first-page":"pp 7795","article-title":"Constrained graph variational autoencoders for molecule design","author":"Liu","year":"2018"},{"key":"mlstabcf91bib13","first-page":"pp 6410","article-title":"Graph convolutional policy network for goal-directed molecular graph generation","author":"You","year":"2018"},{"key":"mlstabcf91bib14","article-title":"Automatic chemical design using a data-driven continuous representation of molecules","author":"Gomez-Bombarelli","year":"2016"},{"key":"mlstabcf91bib15","article-title":"Nevae: a deep generative model for molecular graphs","author":"Samanta","year":"2018"},{"key":"mlstabcf91bib16","first-page":"pp 1263","article-title":"Neural message passing for quantum chemistry","volume":"vol 70","author":"Gilmer","year":"2017"},{"key":"mlstabcf91bib17","article-title":"Gated graph sequence neural networks","author":"Li","year":"2015"},{"key":"mlstabcf91bib18","doi-asserted-by":"publisher","first-page":"595","DOI":"10.1007\/s10822-016-9938-8","article-title":"Molecular graph convolutions: moving beyond fingerprints","volume":"30","author":"Kearnes","year":"2016","journal-title":"J. Comput. Aided Mol. Des."},{"key":"mlstabcf91bib19","first-page":"pp 2224","article-title":"Convolutional networks on graphs for learning molecular fingerprints","author":"Duvenaud","year":"2015"},{"key":"mlstabcf91bib20","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2016"},{"key":"mlstabcf91bib21","article-title":"Generating equilibrium molecules with deep neural networks","author":"Gebauer","year":"2018"},{"key":"mlstabcf91bib22","first-page":"pp 7566","article-title":"Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules","author":"Gebauer","year":"2019"},{"key":"mlstabcf91bib23","article-title":"Generating valid Euclidean distance matrices","author":"Hoffmann","year":"2019"},{"key":"mlstabcf91bib24","article-title":"Molecular sets (MOSES): a benchmarking platform for molecular generation models","author":"Polykovskiy","year":"2018"},{"key":"mlstabcf91bib25","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","article-title":"The graph neural network model","volume":"20","author":"Scarselli","year":"2008","journal-title":"IEEE Trans. Neural Netw."},{"key":"mlstabcf91bib26","article-title":"How powerful are graph neural networks?","author":"Xu","year":"2018"},{"key":"mlstabcf91bib27","article-title":"Spectral networks and locally connected networks on graphs","author":"Bruna","year":"2013"},{"key":"mlstabcf91bib28","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1109\/TNN.2008.2005141","article-title":"Computational capabilities of graph neural networks","volume":"20","author":"Scarselli","year":"2008","journal-title":"IEEE Trans. Neural Netw."},{"key":"mlstabcf91bib29","article-title":"Graph convolutional networks","author":"Kipf","year":"2016"},{"key":"mlstabcf91bib30","doi-asserted-by":"publisher","first-page":"2520","DOI":"10.3390\/molecules23102520","article-title":"Artificial intelligence in drug design","volume":"23","author":"Hessler","year":"2018","journal-title":"Molecules"},{"key":"mlstabcf91bib31","doi-asserted-by":"publisher","first-page":"828","DOI":"10.1039\/C9ME00039A","article-title":"Deep learning for molecular design\u2014a review of the state of the art","volume":"4","author":"Elton","year":"2019","journal-title":"Mol. Syst. Des. Eng."},{"key":"mlstabcf91bib32","article-title":"Generative models for automatic chemical design","author":"Schwalbe-Koda","year":"2019"},{"key":"mlstabcf91bib33","doi-asserted-by":"publisher","first-page":"143","DOI":"10.1038\/s41587-020-0418-2","article-title":"Assessing the impact of generative ai on medicinal chemistry","volume":"38","author":"Walters","year":"2020","journal-title":"Nat. Biotechnol."},{"key":"mlstabcf91bib34","article-title":"Adversarial autoencoders","author":"Makhzani","year":"2015"},{"key":"mlstabcf91bib35","doi-asserted-by":"publisher","DOI":"10.18632\/oncotarget.14073","article-title":"The cornucopia of meaningful leads: applying deep adversarial autoencoders for new molecule development in oncology","volume":"8","author":"Kadurin","year":"2017","journal-title":"Oncotarget"},{"key":"mlstabcf91bib36","doi-asserted-by":"publisher","first-page":"268","DOI":"10.1021\/acscentsci.7b00572","article-title":"Automatic chemical design using a data-driven continuous representation of molecules","volume":"4","author":"G\u00f3mez-Bombarelli","year":"2018","journal-title":"ACS Cent. Sci."},{"key":"mlstabcf91bib37","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1186\/s13321-019-0397-9","article-title":"A de novo molecular generation method using latent vector based generative adversarial network","volume":"11","author":"Prykhodko","year":"2019","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib38","doi-asserted-by":"publisher","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."},{"key":"mlstabcf91bib39","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1126\/science.aat2663","article-title":"Inverse molecular design using machine learning: generative models for matter engineering","volume":"361","author":"Sanchez-Lengeling","year":"2018","journal-title":"Science"},{"key":"mlstabcf91bib40","article-title":"Defactor: differentiable edge factorization-based probabilistic graph generation","author":"Assouel","year":"2018"},{"key":"mlstabcf91bib41","article-title":"Molgan: an implicit generative model for small molecular graphs","author":"De Cao","year":"2018"},{"key":"mlstabcf91bib42","article-title":"Learning multimodal graph-to-graph translation for molecular optimization","author":"Jin","year":"2018"},{"key":"mlstabcf91bib43","first-page":"pp 412","article-title":"Graphvae: towards generation of small graphs using variational autoencoders","author":"Simonovsky","year":"2018"},{"key":"mlstabcf91bib44","article-title":"Graphrnn: generating realistic graphs with deep auto-regressive models","author":"You","year":"2018"},{"key":"mlstabcf91bib45","doi-asserted-by":"publisher","first-page":"4451","DOI":"10.1021\/acs.molpharmaceut.9b00500","article-title":"Deep convolutional generative adversarial network (DCGAN) models for screening and design of small molecules targeting cannabinoid receptors","volume":"16","author":"Bian","year":"2019","journal-title":"Mol. Pharm."},{"key":"mlstabcf91bib46","article-title":"A two-step graph convolutional decoder for molecule generation","author":"Bresson","year":"2019"},{"key":"mlstabcf91bib47","article-title":"Tiered latent representations and latent spaces for molecular graphs","author":"Chang","year":"2019"},{"key":"mlstabcf91bib48","first-page":"747","article-title":"Bradshaw: a system for automated molecular design","author":"Green","year":"2019","journal-title":"J. Comput. Aided Mol. Des."},{"key":"mlstabcf91bib49","doi-asserted-by":"crossref","DOI":"10.26434\/chemrxiv.8266745.v1","article-title":"Multi-resolution autoregressive graph-to-graph translation for molecules","author":"Jin","year":"2019"},{"key":"mlstabcf91bib50","article-title":"Decoding molecular graph embeddings with reinforcement learning","author":"Kearnes","year":"2019"},{"key":"mlstabcf91bib51","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1186\/s13321-019-0396-x","article-title":"Efficient learning of non-autoregressive graph variational autoencoders for molecular graph generation","volume":"11","author":"Kwon","year":"2019","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib52","first-page":"pp 4257","article-title":"Efficient graph generation with graph recurrent attention networks","author":"Liao","year":"2019"},{"key":"mlstabcf91bib53","article-title":"Scaffold-based molecular design using graph generative model","author":"Lim","year":"2019"},{"key":"mlstabcf91bib54","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-019-56773-5","article-title":"Molecular geometry prediction using a deep generative graph neural network","volume":"9","author":"Mansimov","year":"2019","journal-title":"Sci. Rep."},{"key":"mlstabcf91bib55","article-title":"Graphnvp: an invertible flow model for generating molecular graphs","author":"Madhawa","year":"2019"},{"key":"mlstabcf91bib56","article-title":"Likelihood-free inference and generation of molecular graphs","author":"P\u00f6lsterl","year":"2019"},{"key":"mlstabcf91bib57","article-title":"Molecularrnn: generating realistic molecular graphs with optimized properties","author":"Popova","year":"2019"},{"key":"mlstabcf91bib58","article-title":"Hierarchical generation of molecular graphs using structural motifs","author":"Jin","year":"2020"},{"key":"mlstabcf91bib59","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-019-0404-1","article-title":"Mol-cyclegan: a generative model for molecular optimization","volume":"12","author":"Maziarka","year":"2020","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib60","article-title":"Graphaf: a flow-based autoregressive model for molecular graph generation","author":"Shi","year":"2020"},{"key":"mlstabcf91bib61","doi-asserted-by":"publisher","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":"mlstabcf91bib62","doi-asserted-by":"publisher","first-page":"1096","DOI":"10.1021\/acs.jcim.8b00839","article-title":"Guacamol: benchmarking models for de novo molecular design","volume":"59","author":"Brown","year":"2019","journal-title":"J. Chem. Inf. Model."},{"key":"mlstabcf91bib63","article-title":"Deep learning for drug discovery: property prediction with neural networks on raw molecular graphs","author":"Lindel\u00f6f","year":"2019"},{"key":"mlstabcf91bib64","article-title":"Order matters: sequence to sequence for sets","author":"Vinyals","year":"2015"},{"key":"mlstabcf91bib65","doi-asserted-by":"publisher","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."},{"key":"mlstabcf91bib66","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13321-019-0407-y","article-title":"Building attention and edge message passing neural networks for bioactivity and physical\u2013chemical property prediction","volume":"12","author":"Withnall","year":"2020","journal-title":"J. Cheminformatics"},{"key":"mlstabcf91bib67","article-title":"Graph neural networks for drug discovery","author":"Lindel\u00f6f","year":"2020"},{"key":"mlstabcf91bib68","first-page":"pp 5998","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"key":"mlstabcf91bib69","doi-asserted-by":"publisher","DOI":"10.1002\/ail2.18","article-title":"Practical notes on building molecular graph generative models","author":"Mercado","year":"2020","journal-title":"AAIL"},{"key":"mlstabcf91bib70","doi-asserted-by":"publisher","first-page":"8732","DOI":"10.1021\/ja902302h","article-title":"970 million druglike small molecules for virtual screening in the chemical universe database gdb-13","volume":"131","author":"Blum","year":"2009","journal-title":"J. Am. Chem. Soc."},{"key":"mlstabcf91bib71","doi-asserted-by":"crossref","DOI":"10.3389\/fphar.2020.565644","article-title":"Molecular sets (moses): a benchmarking platform for molecular generation models","author":"Polykovskiy","year":"2020"},{"key":"mlstabcf91bib72","doi-asserted-by":"publisher","first-page":"2324","DOI":"10.1021\/acs.jcim.5b00559","article-title":"Zinc 15\u2013ligand discovery for everyone","volume":"55","author":"Sterling","year":"2015","journal-title":"J. Chem. Inf. Model."},{"key":"mlstabcf91bib73","article-title":"Open-source cheminformatics","author":"","year":"2020"},{"key":"mlstabcf91bib74","doi-asserted-by":"publisher","first-page":"1736","DOI":"10.1021\/acs.jcim.8b00234","article-title":"Fr\u00e9chet chemnet distance: a metric for generative models for molecules in drug discovery","volume":"58","author":"Preuer","year":"2018","journal-title":"J. Chem. Inf. Model."},{"key":"mlstabcf91bib75","first-page":"pp 8024","article-title":"Pytorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"key":"mlstabcf91bib76","doi-asserted-by":"publisher","DOI":"10.1002\/minf.201700123","article-title":"Application of generative autoencoder in de novo molecular design","volume":"37","author":"Blaschke","year":"2018","journal-title":"Mol. Inform."},{"key":"mlstabcf91bib77","doi-asserted-by":"publisher","first-page":"688","DOI":"10.1016\/j.cell.2020.01.021","article-title":"A deep learning approach to antibiotic discovery","volume":"180","author":"Stokes","year":"2020","journal-title":"Cell"},{"key":"mlstabcf91bib78","first-page":"pp 971","article-title":"Self-normalizing neural networks","author":"Klambauer","year":"2017"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,6]],"date-time":"2022-01-06T08:31:44Z","timestamp":1641457904000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/abcf91"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,3,2]]},"references-count":78,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2021,3,2]]},"published-print":{"date-parts":[[2021,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/abcf91","relation":{"has-preprint":[{"id-type":"doi","id":"10.26434\/chemrxiv.12843137.v1","asserted-by":"object"}]},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,3,2]]},"assertion":[{"value":"Graph networks for molecular design","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2021 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2020-08-25","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2020-12-01","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2021-03-02","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}