{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T15:15:39Z","timestamp":1777043739908,"version":"3.51.4"},"reference-count":38,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T00:00:00Z","timestamp":1775520000000},"content-version":"vor","delay-in-days":6,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62225209"],"award-info":[{"award-number":["62225209"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004735","name":"Hunan Provincial Natural Science Foundation of China","doi-asserted-by":"crossref","award":["2025JJ30025"],"award-info":[{"award-number":["2025JJ30025"]}],"id":[{"id":"10.13039\/501100004735","id-type":"DOI","asserted-by":"crossref"}]},{"name":"High Performance Computing Center of Central South University"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,4,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Rational molecule generation that balances binding affinity with favorable physicochemical properties remains a formidable challenge in structure-based drug design. The number of functional groups is a key determinant, as over-functionalization compromises physicochemical properties, whereas under-functionalization reduces binding affinity. However, current methods are limited in their capacity to incorporate the constraint.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>To address this, we present NumMolFormer, a Transformer-based framework designed to explicitly model functional group numbers. NumMolFormer adopts a dual-sequence input strategy, integrated with a numerical embedding module and a dual-stream differential attention mechanism, allowing molecular structures and functional group numbers to be encoded separately. This formulation alleviates the inherent limitations of standard Transformer in handling numerical information. In addition, we construct a large-scale dataset of 18 million molecules with functional group annotations for molecular pre-training, and further fine-tune the model using a combination of self-supervised learning and reinforcement learning under protein pocket constraints. The results demonstrate that NumMolFormer effectively leverages functional group information to generate molecules with improved binding affinity, synthetic accessibility, and drug-likeness compared to baseline methods.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>The source code and datasets are available at http:\/\/www.github.com\/zengzhicun\/nummolformer.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag164","type":"journal-article","created":{"date-parts":[[2026,4,5]],"date-time":"2026-04-05T11:27:12Z","timestamp":1775388432000},"source":"Crossref","is-referenced-by-count":0,"title":["NumMolFormer: an explicit functional group number-guided framework for structure-based drug design"],"prefix":"10.1093","volume":"42","author":[{"given":"Zhicun","family":"Zeng","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, Hunan 410083,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yifan","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, Hunan 410083,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhangli","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, Hunan 410083,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0188-1394","authenticated-orcid":false,"given":"Min","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, Hunan 410083,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2026,4,7]]},"reference":[{"key":"2026042409470478300_btag164-B1","author":"Achiam","year":"2023"},{"key":"2026042409470478300_btag164-B2","author":"Adinath","year":"2025"},{"key":"2026042409470478300_btag164-B3","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1038\/nchem.1243","article-title":"Quantifying the chemical beauty of drugs","volume":"4","author":"Bickerton","year":"2012","journal-title":"Nat Chem"},{"key":"2026042409470478300_btag164-B4","doi-asserted-by":"crossref","first-page":"405","DOI":"10.1038\/nm.4306","article-title":"The drug repurposing hub: a next-generation drug library and information resource","volume":"23","author":"Corsello","year":"2017","journal-title":"Nat Med"},{"key":"2026042409470478300_btag164-B5","first-page":"70293","article-title":"Faith and fate: limits of transformers on compositionality","volume":"36","author":"Dziri","year":"2023","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026042409470478300_btag164-B6","doi-asserted-by":"crossref","first-page":"8408","DOI":"10.1021\/acs.jmedchem.0c00754","article-title":"The most common functional groups in bioactive molecules and how their popularity has evolved over time","volume":"63","author":"Ertl","year":"2020","journal-title":"J Med Chem"},{"key":"2026042409470478300_btag164-B7","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1186\/1758-2946-1-8","article-title":"Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions","volume":"1","author":"Ertl","year":"2009","journal-title":"J Cheminform"},{"key":"2026042409470478300_btag164-B8","doi-asserted-by":"crossref","first-page":"412","DOI":"10.3390\/fi17090412","article-title":"Gpt-4.1 sets the standard in automated experiment design using novel python libraries","volume":"17","author":"Fachada","year":"2025","journal-title":"Future Internet"},{"key":"2026042409470478300_btag164-B9","doi-asserted-by":"crossref","first-page":"13384","DOI":"10.3390\/molecules200713384","article-title":"Molecular docking and structure-based drug design strategies","volume":"20","author":"Ferreira","year":"2015","journal-title":"Molecules"},{"key":"2026042409470478300_btag164-B10","doi-asserted-by":"crossref","first-page":"4200","DOI":"10.1021\/acs.jcim.0c00411","article-title":"Three-dimensional convolutional neural networks and a cross-docked data set for structure-based drug design","volume":"60","author":"Francoeur","year":"2020","journal-title":"J Chem Inf Model"},{"key":"2026042409470478300_btag164-B11","author":"Guan"},{"key":"2026042409470478300_btag164-B12","first-page":"11827","author":"Guan","year":"2023"},{"key":"2026042409470478300_btag164-B13","doi-asserted-by":"crossref","first-page":"856","DOI":"10.1021\/ci000403i","article-title":"Molecular complexity and its impact on the probability of finding leads for drug discovery","volume":"41","author":"Hann","year":"2001","journal-title":"J Chem Inf Comput Sci"},{"key":"2026042409470478300_btag164-B14","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1021\/ci049714+","article-title":"Zinc- a free database of commercially available compounds for virtual screening","volume":"45","author":"Irwin","year":"2005","journal-title":"J Chem Inf Model"},{"key":"2026042409470478300_btag164-B15","first-page":"89069","article-title":"Molecule design by latent prompt transformer","volume":"37","author":"Kong","year":"2024","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026042409470478300_btag164-B16","first-page":"5281","article-title":"Rdkit: a software suite for cheminformatics, computational chemistry, and predictive modeling","volume":"8","author":"Landrum","year":"2013","journal-title":"Greg Landrum"},{"key":"2026042409470478300_btag164-B17","doi-asserted-by":"crossref","first-page":"180","DOI":"10.3389\/fmolb.2020.00180","article-title":"Application of fragment-based drug discovery to versatile targets","volume":"7","author":"Li","year":"2020","journal-title":"Front Mol Biosci"},{"key":"2026042409470478300_btag164-B18","doi-asserted-by":"crossref","first-page":"403","DOI":"10.1016\/S0031-6997(24)01340-1","article-title":"Role of pharmacokinetics and metabolism in drug discovery and development","volume":"49","author":"Lin","year":"1997","journal-title":"Pharmacol Rev"},{"key":"2026042409470478300_btag164-B19","author":"Lin","year":"2022"},{"key":"2026042409470478300_btag164-B20","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/S0169-409X(96)00423-1","article-title":"Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings","volume":"23","author":"Lipinski","year":"1997","journal-title":"Adv Drug Deliv Rev"},{"key":"2026042409470478300_btag164-B21","author":"Liu","year":"2024"},{"key":"2026042409470478300_btag164-B22","author":"Liu","year":"2022"},{"key":"2026042409470478300_btag164-B23","author":"Liu","year":"2022"},{"key":"2026042409470478300_btag164-B24","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1021\/acs.accounts.6b00491","article-title":"Forging the basis for developing protein\u2013ligand interaction scoring functions","volume":"50","author":"Liu","year":"2017","journal-title":"Acc Chem Res"},{"key":"2026042409470478300_btag164-B25","first-page":"6229","article-title":"A 3d generative model for structure-based drug design","volume":"34","author":"Luo","year":"2021","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026042409470478300_btag164-B26","doi-asserted-by":"crossref","first-page":"75","DOI":"10.3390\/molecules21010075","article-title":"Chemical structure-related drug-like criteria of global approved drugs","volume":"21","author":"Mao","year":"2016","journal-title":"Molecules"},{"key":"2026042409470478300_btag164-B27","volume-title":"Chem Sci","author":"Ragoza"},{"key":"2026042409470478300_btag164-B28","first-page":"17644","author":"Peng","year":"2022"},{"key":"2026042409470478300_btag164-B29","first-page":"644","author":"Thawani","year":"2021"},{"key":"2026042409470478300_btag164-B30","author":"Touvron","year":"2023"},{"key":"2026042409470478300_btag164-B31","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1042\/EBC20170052","article-title":"Structure-based drug design: aiming for a perfect fit","volume":"61","author":"Van Montfort","year":"2017","journal-title":"Essays Biochem"},{"key":"2026042409470478300_btag164-B32","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026042409470478300_btag164-B33","doi-asserted-by":"crossref","first-page":"2391","DOI":"10.1038\/s41467-025-56349-0","article-title":"Accelerating discovery of bioactive ligands with pharmacophore-informed generative models","volume":"16","author":"Xie","year":"2025","journal-title":"Nat Commun"},{"key":"2026042409470478300_btag164-B34","first-page":"41382","author":"Zhang","year":"2023"},{"key":"2026042409470478300_btag164-B35","author":"Zhang","year":"2023"},{"key":"2026042409470478300_btag164-B36","author":"Zhou","year":"2023"},{"key":"2026042409470478300_btag164-B37","doi-asserted-by":"crossref","first-page":"6234","DOI":"10.1038\/s41467-023-41454-9","article-title":"A pharmacophore-guided deep learning approach for bioactive molecular generation","volume":"14","author":"Zhu","year":"2023","journal-title":"Nat Commun"},{"key":"2026042409470478300_btag164-B38","author":"Ziegler","year":"2019"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btag164\/67937229\/btag164.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/4\/btag164\/67937229\/btag164.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/4\/btag164\/67937229\/btag164.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T13:47:15Z","timestamp":1777038435000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag164\/8627007"}},"subtitle":[],"editor":[{"given":"Xin","family":"Gao","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":38,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,4,7]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag164","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,4]]},"published":{"date-parts":[[2026,4]]},"article-number":"btag164"}}