{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T18:38:29Z","timestamp":1792175909232,"version":"4.4.0"},"reference-count":48,"publisher":"Oxford University Press (OUP)","issue":"6","license":[{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"vor","delay-in-days":34,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002582","name":"Gwangju Institute of Science and Technology","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100002582","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2024-00334990"],"award-info":[{"award-number":["RS-2024-00334990"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Institute of Information & communications Technology Planning & Evaluation","award":["2019\u20130-01842"],"award-info":[{"award-number":["2019\u20130-01842"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Antimicrobial peptides (AMPs) are emerging as promising alternatives to traditional antibiotics, offering solutions to antimicrobial resistance through diverse mechanisms. Despite their potential, current computational approaches for AMP design rarely address strain-specific targeting, limiting their clinical efficacy as bacterial strains exhibit unique membrane compositions and susceptibility profiles requiring tailored interventions. Furthermore, the limited availability of strain-specific training data presents a significant challenge, necessitating parameter-efficient learning approaches that can optimize AMP properties with minimal overfitting. This study introduces a novel AMP generation framework that integrates reinforcement learning with a Generative Pre-trained Transformer (GPT) model enhanced by Low-Rank Adaptation (LoRA) parameter-efficient fine-tuning. This approach enables the design of peptides optimized for multiple objectives, specifically antimicrobial activity and toxicity, tailored to individual pathogen strains. Our framework employs a two-stage learning process: pretraining on a large-scale peptide and AMP database to capture linguistic and contextual features, followed by reinforcement learning that leverages MIC (Minimum Inhibitory Concentration) and hemolysis prediction models to optimize antimicrobial potency and safety profiles. The integration of LoRA is crucial for efficiently adapting the model to strain-specific characteristics while addressing limited training data. The comparative analysis demonstrated our model\u2019s superior performance over existing AMP generation approaches in both activity and hemolytic toxicity metrics. An ablation study confirmed the contributions of reinforcement learning and LoRA. Furthermore, the model can generate peptides satisfying both activity and toxicity conditions for unseen strains, highlighting its capability to design AMPs for emerging pathogens. In addition, molecular dynamics (MDs) simulations confirmed that the generated peptides penetrate bacterial membranes, supporting antimicrobial activity.<\/jats:p>","DOI":"10.1093\/bib\/bbaf641","type":"journal-article","created":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T12:46:05Z","timestamp":1763469965000},"source":"Crossref","is-referenced-by-count":4,"title":["Reinforcement learning with low-rank adaptation for targeted antimicrobial peptide design"],"prefix":"10.1093","volume":"26","author":[{"given":"Juntae","family":"Park","sequence":"first","affiliation":[{"name":"AI Graduate School Gwangju Institute of Science and Technology (GIST), Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daehun","family":"Bae","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST) Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bongsung","family":"Bae","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST) Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5109-9114","authenticated-orcid":false,"given":"Hojung","family":"Nam","sequence":"additional","affiliation":[{"name":"AI Graduate School Gwangju Institute of Science and Technology (GIST), Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]},{"name":"Department of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST) Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]},{"name":"Center for AI-Applied High Efficiency Drug Discovery (AHEDD), Gwangju Institute of Science and Technology (GIST) Buk-gu , Gwangju 61005 ,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,12,5]]},"reference":[{"key":"2026060104575154700_ref1","doi-asserted-by":"publisher","first-page":"115377","DOI":"10.1016\/j.ejmech.2023.115377","article-title":"Antimicrobial peptides: Structure, mechanism, and modification","volume":"255","author":"Chen","year":"2023","journal-title":"Eur J Med Chem"},{"key":"2026060104575154700_ref2","doi-asserted-by":"publisher","first-page":"102733","DOI":"10.1016\/j.sbi.2023.102733","article-title":"Artificial intelligence-driven antimicrobial peptide discovery","volume":"83","author":"Szymczak","year":"2023","journal-title":"Curr Opin Struct Biol"},{"key":"2026060104575154700_ref3","doi-asserted-by":"publisher","first-page":"1750","DOI":"10.1016\/j.jiph.2021.10.020","article-title":"Antibiotic resistance in microbes: History, mechanisms, therapeutic strategies and future prospects","volume":"14","author":"Uddin","year":"2021","journal-title":"J Infect Public Health"},{"key":"2026060104575154700_ref4","first-page":"1","article-title":"Antimicrobial peptides: Mechanism of action, activity and clinical potential","volume":"8","author":"Zhang","year":"2021","journal-title":"Mil Med Res"},{"key":"2026060104575154700_ref5","doi-asserted-by":"publisher","first-page":"1573","DOI":"10.1111\/jam.15314","article-title":"Antimicrobial peptides (AMPs): A promising class of antimicrobial compounds","volume":"132","author":"Erdem B\u00fcy\u00fckkiraz","year":"2022","journal-title":"J Appl Microbiol"},{"key":"2026060104575154700_ref6","doi-asserted-by":"publisher","first-page":"316","DOI":"10.1021\/acs.jcim.3c01881","article-title":"Multi-CGAN: Deep generative model-based multiproperty antimicrobial peptide design","volume":"64","author":"Yu","year":"2023","journal-title":"J Chem Inf Model"},{"key":"2026060104575154700_ref7","doi-asserted-by":"publisher","first-page":"e216","DOI":"10.1016\/S1473-3099(20)30327-3","article-title":"The value of antimicrobial peptides in the age of resistance","volume":"20","author":"Magana","year":"2020","journal-title":"Lancet Infect Dis"},{"key":"2026060104575154700_ref8","author":""},{"key":"2026060104575154700_ref9","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1145\/3422622","article-title":"Generative adversarial networks","volume":"63","author":"Goodfellow","year":"2020","journal-title":"Commun ACM"},{"key":"2026060104575154700_ref10","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"2026060104575154700_ref11","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":"2026060104575154700_ref12","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"2026060104575154700_ref13","doi-asserted-by":"publisher","first-page":"472","DOI":"10.1021\/acs.jcim.7b00414","article-title":"Recurrent neural network model for constructive peptide design","volume":"58","author":"Muller","year":"2018","journal-title":"J Chem Inf Model"},{"key":"2026060104575154700_ref14","doi-asserted-by":"publisher","first-page":"22847","DOI":"10.1021\/acsomega.0c02088","article-title":"Generating ampicillin-level antimicrobial peptides with activity-aware generative adversarial networks","volume":"5","author":"Tucs","year":"2020","journal-title":"ACS Omega"},{"key":"2026060104575154700_ref15","doi-asserted-by":"publisher","first-page":"1453","DOI":"10.1038\/s41467-023-36994-z","article-title":"Setny P: Discovering highly potent antimicrobial peptides with deep generative model HydrAMP","volume":"14","author":"Szymczak","year":"2023","journal-title":"Nat Commun"},{"key":"2026060104575154700_ref16","doi-asserted-by":"publisher","author":"Das","DOI":"10.1016\/j.vetimm.2025.111026"},{"key":"2026060104575154700_ref17","article-title":"Flexible and controllable protein design by prefix-tuning large-scale protein language models","volume":"2023","author":"Luo","year":"2012","journal-title":"bioRxiv"},{"key":"2026060104575154700_ref18","author":"Li"},{"key":"2026060104575154700_ref19","doi-asserted-by":"publisher","author":"Wang","DOI":"10.1177\/12034754251396526"},{"key":"2026060104575154700_ref20","author":"Wang"},{"key":"2026060104575154700_ref21","doi-asserted-by":"publisher","first-page":"eads8932","DOI":"10.1126\/sciadv.ads8932","article-title":"Discovery of antimicrobial peptides with notable antibacterial potency by an LLM-based foundation model. Science","volume":"11","author":"Wang","year":"2025","journal-title":"Advances"},{"key":"2026060104575154700_ref22","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"2026060104575154700_ref23","first-page":"3","article-title":"Lora: Low-rank adaptation of large language models","volume":"1","author":"Hu","year":"2022","journal-title":"ICLR"},{"key":"2026060104575154700_ref24","doi-asserted-by":"publisher","first-page":"429","DOI":"10.1038\/embor.2008.56","article-title":"PeptideAtlas: A resource for target selection for emerging targeted proteomics workflows","volume":"9","author":"Deutsch","year":"2008","journal-title":"EMBO Rep"},{"key":"2026060104575154700_ref25","doi-asserted-by":"publisher","first-page":"D288","DOI":"10.1093\/nar\/gkaa991","article-title":"DBAASP v3: Database of antimicrobial\/cytotoxic activity and structure of peptides as a resource for development of new therapeutics","volume":"49","author":"Pirtskhalava","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"2026060104575154700_ref26","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaf343","article-title":"AI-guided discovery and optimization of antimicrobial peptides through species-aware language model","volume":"26","author":"Bae","year":"2025","journal-title":"Brief Bioinform"},{"key":"2026060104575154700_ref27","doi-asserted-by":"publisher","first-page":"115523","DOI":"10.1016\/j.ab.2024.115523","article-title":"HemoDL: Hemolytic peptides prediction by double ensemble engines from rich sequence-derived and transformer-enhanced information","volume":"690","author":"Yang","year":"2024","journal-title":"Anal Biochem"},{"key":"2026060104575154700_ref28","first-page":"37","article-title":"Antibiotic susceptibility testing using minimum inhibitory concentration (MIC) assays. Npj Antimicrob","volume":"2","author":"Kaderabkova","year":"2024","journal-title":"Resist"},{"key":"2026060104575154700_ref29","doi-asserted-by":"publisher","first-page":"389","DOI":"10.1186\/s12859-022-04952-z","article-title":"AMPDeep: Hemolytic activity prediction of antimicrobial peptides using transfer learning","volume":"23","author":"Salem","year":"2022","journal-title":"BMC bioinformatics"},{"key":"2026060104575154700_ref30","doi-asserted-by":"publisher","first-page":"2577","DOI":"10.1002\/bip.360221211","article-title":"Dictionary of protein secondary structure: Pattern recognition of hydrogen-bonded and geometrical features","volume":"22","author":"Kabsch","year":"1983","journal-title":"Biopolymers: Original Research on Biomolecules"},{"key":"2026060104575154700_ref31","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction of biomolecular interactions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"2026060104575154700_ref32","doi-asserted-by":"publisher","first-page":"229","DOI":"10.1023\/A:1022672621406","article-title":"Simple statistical gradient-following algorithms for connectionist reinforcement learning","volume":"8","author":"Williams","year":"1992","journal-title":"Mach Learn"},{"key":"2026060104575154700_ref33","doi-asserted-by":"publisher","first-page":"2317","DOI":"10.1007\/s00018-011-0721-7","article-title":"Proline-rich antimicrobial peptides: Converging to a non-lytic mechanism of action","volume":"68","author":"Scocchi","year":"2011","journal-title":"Cell Mol Life Sci"},{"key":"2026060104575154700_ref34","doi-asserted-by":"publisher","first-page":"184260","DOI":"10.1016\/j.bbamem.2023.184260","article-title":"Tryptophan-and arginine-rich antimicrobial peptides: Anti-infectives with great potential","volume":"1866","author":"Straus","year":"2024","journal-title":"Biochimica et Biophysica Acta (BBA)-Biomembranes"},{"key":"2026060104575154700_ref35","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1016\/S0005-2736(99)00201-1","article-title":"Structural features of helical antimicrobial peptides: Their potential to modulate activity on model membranes and biological cells","volume":"1462","author":"Dathe","year":"1999","journal-title":"Biochim Biophys Acta (BBA)-Biomembranes"},{"key":"2026060104575154700_ref36","doi-asserted-by":"publisher","first-page":"116088","DOI":"10.1016\/j.biopha.2023.116088","article-title":"Sequential rearrangement and stereochemical reorganization to design an antimicrobial peptide with enhanced stability","volume":"170","author":"Hsu","year":"2024","journal-title":"Biomed Pharmacother"},{"key":"2026060104575154700_ref37","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1002\/psc.560","article-title":"Structure\u2013activity relationship of an antibacterial peptide, maculatin 1.1, from the skin glands of the tree frog, Litoria genimaculata","volume":"10","author":"Niidome","year":"2004","journal-title":"Journal of peptide science: an official publication of the European Peptide Society"},{"key":"2026060104575154700_ref38","doi-asserted-by":"crossref","first-page":"bbae644","DOI":"10.1093\/bib\/bbae644","article-title":"TG-CDDPM: Text-guided antimicrobial peptides generation based on conditional denoising diffusion probabilistic model","volume":"26","author":"Cao","year":"2025","journal-title":"Brief Bioinform"},{"key":"2026060104575154700_ref39","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","volume":"45","author":"Breiman","year":"2001","journal-title":"Random forests Machine learning"},{"key":"2026060104575154700_ref40","doi-asserted-by":"publisher","first-page":"275","DOI":"10.4155\/fmc-2016-0188","article-title":"HemoPred: A web server for predicting the hemolytic activity of peptides","volume":"9","author":"Win","year":"2017","journal-title":"Future Med Chem"},{"key":"2026060104575154700_ref41","doi-asserted-by":"publisher","first-page":"D377","DOI":"10.1093\/nar\/gkac933","article-title":"CAMPR4: A database of natural and synthetic antimicrobial peptides","volume":"51","author":"Gawde","year":"2023","journal-title":"Nucleic Acids Res"},{"key":"2026060104575154700_ref42","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2026060104575154700_ref43","doi-asserted-by":"publisher","first-page":"D444","DOI":"10.1093\/nar\/gkt1008","article-title":"Hemolytik: A database of experimentally determined hemolytic and non-hemolytic peptides","volume":"42","author":"Gautam","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2026060104575154700_ref44","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1002\/1097-0282(2000)55:1&lt;4::AID-BIP30&gt;3.0.CO;2-M","article-title":"Amphipathic, \u03b1-helical antimicrobial peptides","volume":"55","author":"Tossi","year":"2000","journal-title":"Pept Sci"},{"key":"2026060104575154700_ref45","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction with AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2026060104575154700_ref46","doi-asserted-by":"publisher","first-page":"1859","DOI":"10.1002\/jcc.20945","article-title":"CHARMM-GUI: A web-based graphical user interface for CHARMM","volume":"29","author":"Jo","year":"2008","journal-title":"J Comput Chem"},{"key":"2026060104575154700_ref47","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1016\/0263-7855(96)00018-5","article-title":"VMD: Visual molecular dynamics","volume":"14","author":"Humphrey","year":"1996","journal-title":"J Mol Graph"},{"key":"2026060104575154700_ref48","doi-asserted-by":"publisher","volume-title":"Sci. Rep.","author":"","DOI":"10.1038\/srep22843"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/6\/bbaf641\/65779748\/bbaf641.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/6\/bbaf641\/65779748\/bbaf641.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T08:58:01Z","timestamp":1780304281000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf641\/8371890"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,1]]},"references-count":48,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf641","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,11]]},"published":{"date-parts":[[2025,11,1]]},"article-number":"bbaf641"}}