{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,23]],"date-time":"2026-08-23T17:58:11Z","timestamp":1787507891885,"version":"build-2736575974"},"reference-count":35,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T00:00:00Z","timestamp":1774224000000},"content-version":"vor","delay-in-days":22,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Health Science Reports"],"published-print":{"date-parts":[[2026,3]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background and Aims<\/jats:title>\n                    <jats:p>De novo drug design is the process of generating novel lead compounds that possess desirable pharmacological activities and optimal physicochemical properties for therapeutic development. In recent years, it has evolved into a key computational strategy for discovering and optimizing new therapeutic compounds. Reinforcement learning (RL), a branch of artificial intelligence, has emerged as a powerful tool to address the complex, sequential decision\u2010making processes involved in molecular generation. This study aims to review recent applications of RL in de novo drug design, highlight commonly used algorithms, identify major challenges, and discuss future research directions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A systematic literature review (SLR) was conducted following standard review procedures. Articles published between January 2017 and January 2024 were retrieved from Google Scholar using the keyword \u201cReinforcement Learning Techniques in de novo Drug Design.\u201d Studies were screened based on eligibility criteria, including relevance to RL\u2010based molecular generation, English language, and full\u2010text availability. Selected papers were analyzed to extract information on RL algorithms, design strategies, and application areas.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The reviewed studies demonstrate that RL has been successfully applied to molecular generation, optimization, and drug\u2010target design. Commonly used algorithms include policy\u2010gradient, actor\u2013critic, and value\u2010based methods, often integrated with deep generative models such as recurrent neural networks (RNNs), variational autoencoders (VAEs), generative adversarial networks (GANs), and graph neural networks (GNNs). RL frameworks have optimized properties like binding affinity, solubility, and bioavailability, while promoting molecular diversity. Despite these advances, challenges remain in sample efficiency, reward formulation, and interpretability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Reinforcement learning provides a robust framework for automated drug design, enabling intelligent exploration of chemical space and the generation of novel, bioactive compounds. However, further improvements in multi\u2010objective optimization, computational efficiency, and model transparency are essential for broader clinical applicability. Future research should focus on hybrid RL architectures and explainable AI techniques to bridge computational and experimental drug discovery.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/hsr2.72132","type":"journal-article","created":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T06:53:20Z","timestamp":1774335200000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Application of Reinforcement Learning Techniques in De Novo Drug Design: A Systematic Literature Review"],"prefix":"10.1002","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1707-0227","authenticated-orcid":false,"given":"Masuda Begum","family":"Sampa","sequence":"first","affiliation":[{"name":"Faculty of Engineering and Technology Multimedia University Melaka Malaysia"},{"name":"University of Science and Technology Chittagong (USTC) Chattogram Bangladesh"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nor Hidayati Abdul","family":"Aziz","sequence":"additional","affiliation":[{"name":"Faculty of Engineering and Technology Multimedia University Melaka Malaysia"},{"name":"Centre for Advanced Analytics, CoE for Artificial Intelligence Multimedia University Melaka Malaysia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,3,23]]},"reference":[{"key":"e_1_2_12_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2023.10.001"},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.9b00325"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.aap7885"},{"key":"e_1_2_12_5_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btad693"},{"key":"e_1_2_12_6_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-017-0235-x"},{"key":"e_1_2_12_7_1","first-page":"1","article-title":"De Novo Drug Design Using Reinforcement Learning With Multiple GPT Agents","author":"Hu X.","year":"2023","journal-title":"(NeurIPS)"},{"key":"e_1_2_12_8_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btab301"},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-024-00812-5"},{"key":"e_1_2_12_10_1","unstructured":"M. M.Afsar T.Crump andB.Far(2021). Reinforcement Learning Based Recommender Systems: A Survey. arXiv preprint arXiv:2101.06286."},{"key":"e_1_2_12_11_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42004-022-00733-0"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-019-0355-6"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.9b00325"},{"key":"e_1_2_12_14_1","doi-asserted-by":"crossref","unstructured":"J.Born M.Manica A.Oskooei J.Cadow andM.Rodr\u00edguez Mart\u00ednez(2019). Reinforcement Learning\u2010driven De\u2010novo Design of Anticancer Compounds Conditioned on Biomolecular Profiles https:\/\/doi.org\/10.1007\/978-3-030-45257-5_18.","DOI":"10.1007\/978-3-030-45257-5_18"},{"key":"e_1_2_12_15_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.1c01319"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1039\/D3SC04091G"},{"key":"e_1_2_12_17_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.2c00838"},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2205.10473"},{"key":"e_1_2_12_19_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-021-00561-9"},{"key":"e_1_2_12_20_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.0c00915"},{"key":"e_1_2_12_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.isci.2021.102269"},{"key":"e_1_2_12_22_1","doi-asserted-by":"crossref","unstructured":"Q.Wang X.Hu andH.Liu(2022). Deep Reinforcement Learning and Docking Simulations for Autonomous Molecule Generation in De Novo Drug Design. In MMAsia '21: Proceedings of the 3rd ACM International Conference on Multimedia in Asia.","DOI":"10.1145\/3469877.3497694"},{"key":"e_1_2_12_23_1","doi-asserted-by":"publisher","DOI":"10.1021\/acsomega.2c06653"},{"key":"e_1_2_12_24_1","unstructured":"S. K.Gottipati B.Sattarov S.Niu et al. (2020). Learning to Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning. In Proceedings of the 37th International Conference on Machine Learning PMLR."},{"key":"e_1_2_12_25_1","doi-asserted-by":"publisher","DOI":"10.1039\/C8SC05372C"},{"key":"e_1_2_12_26_1","doi-asserted-by":"crossref","unstructured":"A.Schneuing C.Harris Y.Du et al. \u201cStructure\u2011Based Drug Design With Equivariant Diffusion Models \u201d arXiv 2024.","DOI":"10.1038\/s43588-024-00737-x"},{"key":"e_1_2_12_27_1","unstructured":"E.Bengio M.Jain M.Korablyov D.Precup andY.Bengio \u201cFlow Network Based Generative Models for Non\u2010iterative Diverse Candidate Generation \u201d arXiv 2021 https:\/\/arxiv.org\/abs\/2106.04399."},{"key":"e_1_2_12_28_1","doi-asserted-by":"crossref","unstructured":"W.Gao T.Fu J.Sun andC. W.Coley(2022). Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization. In Advances in Neural Information Processing Systems (Vol. 35). NeurIPS https:\/\/arxiv.org\/abs\/2206.12411.","DOI":"10.52202\/068431-1551"},{"key":"e_1_2_12_29_1","unstructured":"W.Jin R.Barzilay andT.Jaakkola(2020). Multi\u2010Objective Molecule Generation Using Interpretable Substructures. In International Conference on Machine Learning pp.4849\u20134859. PMLR."},{"key":"e_1_2_12_30_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.8b00839"},{"key":"e_1_2_12_31_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.0c00174"},{"key":"e_1_2_12_32_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-00236-4"},{"key":"e_1_2_12_33_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-020-0418-2"},{"key":"e_1_2_12_34_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41573-019-0024-5"},{"key":"e_1_2_12_35_1","article-title":"Graph Convolutional Policy Network for Goal\u2010Directed Molecular Graph Generation","author":"You J.","year":"2019","journal-title":"arXiv"},{"key":"e_1_2_12_36_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aat2663"}],"container-title":["Health Science Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/hsr2.72132","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/hsr2.72132","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/hsr2.72132","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T17:50:23Z","timestamp":1787248223000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/hsr2.72132"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3]]},"references-count":35,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,3]]}},"alternative-id":["10.1002\/hsr2.72132"],"URL":"https:\/\/doi.org\/10.1002\/hsr2.72132","archive":["Portico"],"relation":{},"ISSN":["2398-8835","2398-8835"],"issn-type":[{"value":"2398-8835","type":"print"},{"value":"2398-8835","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3]]},"assertion":[{"value":"2024-08-18","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-12","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e72132"}}