{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T10:27:51Z","timestamp":1787048871959,"version":"build-2736575974"},"reference-count":28,"publisher":"American Chemical Society (ACS)","issue":"7","license":[{"start":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T00:00:00Z","timestamp":1679616000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T00:00:00Z","timestamp":1679616000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T00:00:00Z","timestamp":1679616000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,4,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>We introduce the AiZynthTrain Python package for training synthesis models in a robust, reproducible, and extensible way. It contains two pipelines that create a template-based one-step retrosynthesis model and a RingBreaker model that can be straightforwardly integrated in retrosynthesis software. We train such models on the publicly available reaction data set from the U.S. Patent and Trademark Office (USPTO), and these are the first retrosynthesis models created in a completely reproducible end-to-end fashion, starting with the original reaction data source and ending with trained machine-learning models. In particular, we show that employing new heuristics implemented in the pipeline greatly improves the ability of the RingBreaker model for disconnecting ring systems. Furthermore, we demonstrate the robustness of the pipeline by training on a more diverse but proprietary data set. We envisage that this framework will be extended with other synthesis models in the future.<\/jats:p>","DOI":"10.1021\/acs.jcim.2c01486","type":"journal-article","created":{"date-parts":[[2023,3,24]],"date-time":"2023-03-24T01:27:20Z","timestamp":1679621240000},"page":"1841-1846","source":"Crossref","is-referenced-by-count":24,"title":["AiZynthTrain: Robust,\nReproducible, and Extensible\nPipelines for Training Synthesis Prediction Models"],"prefix":"10.1021","volume":"63","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7624-7363","authenticated-orcid":true,"given":"Samuel","family":"Genheden","sequence":"first","affiliation":[{"name":"AstraZeneca Gothenburg , , ,","place":["Mo\u0308lndal, Sweden, SE-431 83"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2419-0705","authenticated-orcid":true,"given":"Per-Ola","family":"Norrby","sequence":"additional","affiliation":[{"name":"AstraZeneca Gothenburg , , ,","place":["Mo\u0308lndal, Sweden, SE-431 83"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4970-6461","authenticated-orcid":true,"given":"Ola","family":"Engkvist","sequence":"additional","affiliation":[{"name":"AstraZeneca Gothenburg , , ,","place":["Mo\u0308lndal, Sweden, SE-431 83"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2023,3,24]]},"reference":[{"key":"2026081804441969600_cit1","doi-asserted-by":"publisher","first-page":"100041","DOI":"10.1016\/j.ailsci.2022.100041","article-title":"AI for Drug Design: From Explicit\nRules to Deep Learning","volume":"2","author":"Mervin","year":"2022","journal-title":"Artif. Intell. Life\nSci."},{"key":"2026081804441969600_cit2","doi-asserted-by":"publisher","first-page":"e1604","DOI":"10.1002\/wcms.1604","article-title":"Machine Intelligence for Chemical\nReaction Space","volume":"12","author":"Schwaller","year":"2022","journal-title":"Wiley Interdiscip. Rev. Comput.\nMol. Sci."},{"key":"2026081804441969600_cit3","unstructured":"Lowe, D.\n          \n          Chemical reactions\nfrom U.S. patents (1976-Sep2016), 2017.  https:\/\/figshare.com\/articles\/dataset\/Chemical_reactions_from_US_patents_1976-Sep2016_\/5104873."},{"issue":"45","key":"2026081804441969600_cit4","doi-asserted-by":"publisher","first-page":"18820","DOI":"10.1021\/jacs.1c09820","article-title":"The Open\nReaction Database","volume":"143","author":"Kearnes","year":"2021","journal-title":"J. Am. Chem. Soc."},{"key":"2026081804441969600_cit5","unstructured":"Bai, F.; Ritter, A.; Madrid, P.; Freitag, D.; Niekrasz, J.\n          SynKB:\nSemantic Search for Synthetic Procedures. arXiv\nPreprint, arXiv:2208.07400, 202210.48550\/arxiv.2208.07400"},{"key":"2026081804441969600_cit6","unstructured":"Reaxys. https:\/\/www.reaxys.com\/ (accessed 2022\u201309\u201301)."},{"key":"2026081804441969600_cit7","unstructured":"CAS Content. CAS. https:\/\/www.cas.org\/about\/cas-content (accessed 2022\u201309\u201301)."},{"key":"2026081804441969600_cit8","unstructured":"Pistachio. NextMove Software. https:\/\/www.nextmovesoftware.com\/pistachio.html (accessed 2022\u201309\u201301)."},{"issue":"11","key":"2026081804441969600_cit9","doi-asserted-by":"publisher","first-page":"4819","DOI":"10.1021\/jacs.1c12005","article-title":"Machine Learning May Sometimes Simply\nCapture Literature Popularity Trends: A Case Study of Heterocyclic\nSuzuki-Miyaura Coupling","volume":"144","author":"Beker","year":"2022","journal-title":"J. Am. Chem. Soc."},{"issue":"29","key":"2026081804441969600_cit10","doi-asserted-by":"publisher","first-page":"e202204647","DOI":"10.1002\/anie.202204647","article-title":"Machine Learning for Chemical Reactivity:\nThe Importance of Failed Experiments","volume":"61","author":"Strieth-Kalthoff","year":"2022","journal-title":"Angew.\nChemie - Int. Ed."},{"key":"2026081804441969600_cit11","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1016\/j.ddtec.2020.06.002","article-title":"AI-Assisted Synthesis Prediction","volume":"32-33","author":"Johansson","year":"2019","journal-title":"Drug Discovery Today: Technologies"},{"issue":"6453","key":"2026081804441969600_cit12","doi-asserted-by":"publisher","first-page":"eaax1566","DOI":"10.1126\/science.aax1566","article-title":"A Robotic\nPlatform for Flow Synthesis of Organic Compounds Informed by AI Planning","volume":"365","author":"Coley","year":"2019","journal-title":"Science"},{"issue":"12","key":"2026081804441969600_cit13","doi-asserted-by":"publisher","first-page":"3316","DOI":"10.1039\/C9SC05704H","article-title":"Predicting\nRetrosynthetic Pathways Using Transformer-Based Models and a Hyper-Graph\nExploration Strategy","volume":"11","author":"Schwaller","year":"2020","journal-title":"Chem. Sci."},{"issue":"1","key":"2026081804441969600_cit14","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1186\/s13321-020-00472-1","article-title":"AiZynthFinder: A Fast,\nRobust and Flexible Open-Source\nSoftware for Retrosynthetic Planning","volume":"12","author":"Genheden","year":"2020","journal-title":"J. Cheminform."},{"key":"2026081804441969600_cit15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TCIAIG.2012.2186810","article-title":"A Survey of Monte Carlo Tree Search Methods","volume":"4","author":"Browne","year":"2012","journal-title":"IEEE Transactions on Computational Intelligence and AI in\nGames."},{"issue":"4","key":"2026081804441969600_cit16","doi-asserted-by":"publisher","first-page":"527","DOI":"10.1039\/D2DD00015F","article-title":"PaRoutes: Towards a Framework for Benchmarking Retrosynthesis\nRoute Predictions","volume":"1","author":"Genheden","year":"2022","journal-title":"Digit. Discovery"},{"issue":"1","key":"2026081804441969600_cit17","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1039\/C9SC04944D","article-title":"Datasets\nand Their Influence on the Development of Computer Assisted Synthesis\nPlanning Tools in the Pharmaceutical Domain","volume":"11","author":"Thakkar","year":"2020","journal-title":"Chem. Sci."},{"key":"2026081804441969600_cit18","doi-asserted-by":"crossref","unstructured":"Kannas, C.; Thakkar, A.; Bjerrum, E.; Genheden, S.\n          rxnutils\n\u2013 A Cheminformatics Python Library\nfor Manipulating Chemical Reaction Data. ChemRxiv, 2022. 10.26434\/CHEMRXIV-2022-WT440-V2.","DOI":"10.26434\/chemrxiv-2022-wt440-v2"},{"issue":"15","key":"2026081804441969600_cit19","doi-asserted-by":"publisher","first-page":"eabe4166","DOI":"10.1126\/sciadv.abe4166","article-title":"Extraction\nof Organic Chemistry Grammar from Unsupervised Learning of Chemical\nReactions","volume":"7","author":"Schwaller","year":"2021","journal-title":"Sci. Adv."},{"key":"2026081804441969600_cit20","unstructured":"NameRxn. NextMove Software. https:\/\/www.nextmovesoftware.com\/namerxn.html (accessed 2022\u201309\u201301)."},{"key":"2026081804441969600_cit21","unstructured":"Pipeline\nPilot v20.1.0.2208. Biovia. https:\/\/www.3ds.com\/products-services\/biovia\/products\/data-science\/pipeline-pilot\/"},{"key":"2026081804441969600_cit22","unstructured":"RDKit. https:\/\/github.com\/rdkit (accessed 2022\u201309\u201301)."},{"issue":"16","key":"2026081804441969600_cit23","doi-asserted-by":"publisher","first-page":"8791","DOI":"10.1021\/acs.jmedchem.9b01919","article-title":"\u201cRing\nBreaker\u201d: Neural Network Driven Synthesis Prediction of the\nRing System Chemical Space","volume":"63","author":"Thakkar","year":"2020","journal-title":"J. Med. Chem."},{"issue":"6","key":"2026081804441969600_cit24","doi-asserted-by":"publisher","first-page":"2529","DOI":"10.1021\/acs.jcim.9b00286","article-title":"RDChiral: An RDKit Wrapper for Handling\nStereochemistry in Retrosynthetic Template Extraction and Application","volume":"59","author":"Coley","year":"2019","journal-title":"J. Chem. Inf. Model."},{"issue":"D1","key":"2026081804441969600_cit25","doi-asserted-by":"publisher","first-page":"D1100","DOI":"10.1093\/nar\/gkr777","article-title":"ChEMBL:\nA Large-Scale Bioactivity Database for Drug Discovery","volume":"40","author":"Gaulton","year":"2012","journal-title":"Nucleic Acids Res."},{"issue":"12","key":"2026081804441969600_cit26","doi-asserted-by":"publisher","first-page":"5918","DOI":"10.1021\/acs.jcim.0c00915","article-title":"REINVENT 2.0: An AI Tool for de Novo Drug Design","volume":"60","author":"Blaschke","year":"2020","journal-title":"J. Chem. Inf. Model."},{"issue":"11","key":"2026081804441969600_cit27","doi-asserted-by":"publisher","first-page":"2324","DOI":"10.1021\/acs.jcim.5b00559","article-title":"ZINC 15 - Ligand\nDiscovery for Everyone","volume":"55","author":"Sterling","year":"2015","journal-title":"J. Chem. Inf. Model."},{"key":"2026081804441969600_cit28","unstructured":"eMolecules. https:\/\/www.emolecules.com\/ (accessed 2022\u201302\u201301)."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.2c01486","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/63\/7\/1841\/8909206\/ci2c01486.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/63\/7\/1841\/8909206\/ci2c01486.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T09:51:24Z","timestamp":1787046684000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/63\/7\/1841\/850322\/AiZynthTrain-Robust-Reproducible-and-Extensible"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,24]]},"references-count":28,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2023,3,24]]},"published-print":{"date-parts":[[2023,4,10]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.2c01486","relation":{"has-preprint":[{"id-type":"doi","id":"10.26434\/chemrxiv-2022-kls5q","asserted-by":"object"}]},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,24]]}}}