{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/molecular-de-novo-design-through-deep","title":"Molecular De Novo Design through Deep Reinforcement Learning","arxiv_id":"1704.07555","date":"2017-04-25","proceeding":null,"authors":["Marcus Olivecrona","Thomas Blaschke","Ola Engkvist","Hongming Chen"],"abstract":"This work introduces a method to tune a sequence-based generative model for\nmolecular de novo design that through augmented episodic likelihood can learn\nto generate structures with certain specified desirable properties. We\ndemonstrate how this model can execute a range of tasks such as generating\nanalogues to a query structure and generating compounds predicted to be active\nagainst a biological target. As a proof of principle, the model is first\ntrained to generate molecules that do not contain sulphur. As a second example,\nthe model is trained to generate analogues to the drug Celecoxib, a technique\nthat could be used for scaffold hopping or library expansion starting from a\nsingle molecule. Finally, when tuning the model towards generating compounds\npredicted to be active against the dopamine receptor type 2, the model\ngenerates structures of which more than 95% are predicted to be active,\nincluding experimentally confirmed actives that have not been included in\neither the generative model nor the activity prediction model.","url_abs":"http://arxiv.org/abs/1704.07555v2","url_pdf":"http://arxiv.org/pdf/1704.07555v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"molecular-de-novo-design-through-deep","repo_url":"https://github.com/MarcusOlivecrona/REINVENT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"molecular-de-novo-design-through-deep","repo_url":"https://github.com/qyuan7/RNN_RL_molecule","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"molecular-de-novo-design-through-deep","repo_url":"https://github.com/qyuan7/RNN_TL_molecule","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"activity-prediction","task_name":"Activity Prediction"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.07555","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}