{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,11]],"date-time":"2026-07-11T21:13:24Z","timestamp":1783804404345,"version":"3.55.0"},"reference-count":46,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T00:00:00Z","timestamp":1754265600000},"content-version":"vor","delay-in-days":34,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Artificial Intelligence and Complex Computational Research Award, University of Florida"},{"DOI":"10.13039\/100000002","name":"NIH","doi-asserted-by":"publisher","award":["NIAID R01 AI170187"],"award-info":[{"award-number":["NIAID R01 AI170187"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,7,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Background: The COVID-19 pandemic has caused over 776 million infections and 7 million deaths globally between December 2019 and November 2024. Since the emergence of the original Wuhan strain, SARS-CoV-2 has evolved into multiple variants\u2014including Alpha, Delta, and Omicron\u2014primarily through mutations in the Spike glycoprotein. The S1 subunit, which binds the human angiotensin-converting enzyme 2 (ACE2) receptor, mutates frequently and plays a key role in infectivity and immune escape, while the more conserved S2 subunit mediates membrane fusion. Anticipating future mutations is essential for guiding vaccine design and therapeutic strategies. Generative Large Language Models (LLMs) have shown promise in protein sequence modeling due to their capacity to produce realistic and functional synthetic sequences. Here, we introduce SARITA, a GPT-3-based LLM with up to 1.2 billion parameters, fine-tuned via continual learning on the protein model RITA trained on 107 017 high-quality SARS-CoV-2 Spike sequences (up to March 1st 2021) to generate high-quality synthetic SARS-CoV-2 Spike S1 subunits. Results: SARITA is able to generate realistic, full-length synthetic S1 subunits starting from a 14-amino-acid prompt. When evaluated on unseen sequences collected between March 2021 and November 2023\u2014including major Variants of Concern (VOCs) such as Delta and Omicron, and Variants of Interest such as Iota\u2014SARITA outperforms baseline and state-of-the-art LLMs in terms of sequence quality, biological plausibility, and similarity to real-world viral evolution. SARITA generates high-quality sequences in over 97% of cases, with markedly lower False Mutation Rate and higher similarity scores (PAM30, Levenshtein distance) compared to alternative approaches. It also accurately reproduces key mutations characteristic of future variants\u2014such as L212I, R158L, T95P, and E406K\u2014which were not present in the training data but emerged later in VOCs like Omicron and Delta. Structure-based analysis confirms the functional plausibility of these substitutions, with \u0394\u0394G values within experimentally supported thresholds for ACE2 and antibody binding. Furthermore, SARITA anticipates immune-evasive mutations and accurately captures the positional and statistical distribution of mutations found in post- March 1st 2021 variants, highlighting its potential as a predictive tool for viral evolution. Conclusion: These results indicate the potential of SARITA to predict future SARS-CoV-2\u00a0S1 evolution, potentially aiding in the development of adaptable vaccines and treatments.<\/jats:p>","DOI":"10.1093\/bib\/bbaf384","type":"journal-article","created":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T07:39:47Z","timestamp":1754293187000},"source":"Crossref","is-referenced-by-count":8,"title":["SARITA: a large language model for generating the S1 subunit of the SARS-CoV-2 spike protein"],"prefix":"10.1093","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-4405-1697","authenticated-orcid":false,"given":"Simone","family":"Rancati","sequence":"first","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]},{"name":"Department of Epidemiology, College of Public Health and Health Professions, University of Florida , 2004 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]},{"name":"Emerging Pathogens Institute, University of Florida , 2055 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Giovanna","family":"Nicora","sequence":"additional","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Laura","family":"Bergomi","sequence":"additional","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tommaso Mario","family":"Buonocore","sequence":"additional","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel M","family":"Czyz","sequence":"additional","affiliation":[{"name":"Emerging Pathogens Institute, University of Florida , 2055 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]},{"name":"Department of Microbiology and Cell Science, University of Florida , 1355 Museum Dr, Gainesville, FL 32603 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Enea","family":"Parimbelli","sequence":"additional","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6974-9808","authenticated-orcid":false,"given":"Riccardo","family":"Bellazzi","sequence":"additional","affiliation":[{"name":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia , Via Adolfo Ferrata 5, Pavia, 27100 ,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0136-2102","authenticated-orcid":false,"given":"Marco","family":"Salemi","sequence":"additional","affiliation":[{"name":"Emerging Pathogens Institute, University of Florida , 2055 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]},{"name":"Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida , 1600 SW Archer Road, Gainesville, FL 32610 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9021-5595","authenticated-orcid":false,"given":"Mattia","family":"Prosperi","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, College of Public Health and Health Professions, University of Florida , 2004 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]},{"name":"Emerging Pathogens Institute, University of Florida , 2055 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5704-3533","authenticated-orcid":false,"given":"Simone","family":"Marini","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, College of Public Health and Health Professions, University of Florida , 2004 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]},{"name":"Emerging Pathogens Institute, University of Florida , 2055 Mowry Road, Gainesville, FL 32610 ,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,8,4]]},"reference":[{"key":"2025080403394033400_ref1","doi-asserted-by":"publisher","first-page":"1141","DOI":"10.1038\/s41401-020-0485-4","article-title":"Structural and functional properties of SARS-CoV-2 spike protein: Potential antivirus drug development for COVID-19","volume":"41","author":"Huang","year":"2020","journal-title":"Acta Pharmacol Sin"},{"key":"2025080403394033400_ref2","author":"JGraph","year":"2021"},{"key":"2025080403394033400_ref3","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1016\/j.coviro.2021.08.010","article-title":"Structure of SARS-CoV-2 spike protein","volume":"50","author":"Zhang","year":"2021","journal-title":"Curr Opin Virol"},{"key":"2025080403394033400_ref4","doi-asserted-by":"publisher","first-page":"e100643","DOI":"10.1136\/bmjhci-2022-100643","article-title":"Predicting emerging SARS-CoV-2 variants of concern through a one class dynamic anomaly detection algorithm","volume":"29","author":"Nicora","year":"2022","journal-title":"BMJ Health Care Inform"},{"key":"2025080403394033400_ref5","doi-asserted-by":"publisher","first-page":"eabk3445","DOI":"10.1126\/scitranslmed.abk3445","article-title":"Predicting the mutational drivers of future SARS-CoV-2 variants of concern","volume":"14","author":"Maher","year":"2022","journal-title":"Sci Transl Med"},{"key":"2025080403394033400_ref6","doi-asserted-by":"publisher","first-page":"bbae535","DOI":"10.1093\/bib\/bbae535","article-title":"Forecasting dominance of SARS-CoV-2 lineages by anomaly detection using deep AutoEncoders","volume":"25","author":"Rancati","year":"2024","journal-title":"Brief Bioinform"},{"key":"2025080403394033400_ref7","doi-asserted-by":"publisher","first-page":"818","DOI":"10.1038\/s41586-023-06617-0","article-title":"Learning from prepandemic data to forecast viral escape","volume":"622","author":"Thadani","year":"2023","journal-title":"Nature"},{"key":"2025080403394033400_ref8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2108.07435","article-title":"Modeling protein using large-scale Pretrain language model","author":"Xiao","year":"2021"},{"key":"2025080403394033400_ref9","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1186\/s12859-024-05779-6","article-title":"Improvements in viral gene annotation using large language models and soft alignments","volume":"25","author":"Harrigan","year":"2024","journal-title":"BMC Bioinformatics"},{"key":"2025080403394033400_ref10","doi-asserted-by":"publisher","first-page":"pdb.ip59","DOI":"10.1101\/pdb.ip59","article-title":"Comparison of the PAM and BLOSUM amino acid substitution matrices","volume":"2008","author":"Mount","year":"2008","journal-title":"CSH Protoc"},{"key":"2025080403394033400_ref11","doi-asserted-by":"publisher","DOI":"10.1101\/2023.01.17.524472","article-title":"A deep generative model of the SARS-CoV-2 spike protein predicts future variants","author":"Dhodapkar","year":"2023"},{"key":"2025080403394033400_ref12","doi-asserted-by":"publisher","first-page":"e1011790","DOI":"10.1371\/journal.pcbi.1011790","article-title":"PandoGen: Generating complete instances of future SARS-CoV-2 sequences using deep learning","volume":"20","author":"Ramachandran","year":"2024","journal-title":"PLoS Comput Biol"},{"key":"2025080403394033400_ref13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2205.05789","article-title":"RITA: A study on scaling up generative protein sequence models","author":"Hesslow","year":"2024"},{"key":"2025080403394033400_ref14","doi-asserted-by":"publisher","first-page":"30494","DOI":"10.2807\/1560-7917.ES.2017.22.13.30494","article-title":"GISAID: Global initiative on sharing all influenza data \u2013 From vision to reality","volume":"22","author":"Shu","year":"2017","journal-title":"Euro Surveill"},{"key":"2025080403394033400_ref15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2005.14165","article-title":"Language models are few-shot learners","author":"Brown","year":"2020"},{"key":"2025080403394033400_ref16","doi-asserted-by":"publisher","first-page":"968","DOI":"10.1016\/j.cels.2023.10.002","article-title":"ProGen2: Exploring the boundaries of protein language models","volume":"14","author":"Nijkamp","year":"2023","journal-title":"cels"},{"key":"2025080403394033400_ref17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2104.09864","article-title":"RoFormer: Enhanced transformer with rotary position embedding","author":"Su","year":"2023"},{"key":"2025080403394033400_ref18","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1186\/s12864-022-08358-2","article-title":"Pango lineage designation and assignment using SARS-CoV-2 spike gene nucleotide sequences","volume":"23","author":"O\u2019Toole","year":"2022","journal-title":"BMC Genomics"},{"key":"2025080403394033400_ref19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2203.15556","article-title":"Training compute-optimal large language models","author":"Hoffmann","year":"2022"},{"key":"2025080403394033400_ref20","doi-asserted-by":"publisher","first-page":"851","DOI":"10.4169\/000298910x523344","article-title":"The Gini index and measures of inequality","volume":"117","author":"Farris","year":"2010","journal-title":"Am Math Mon"},{"key":"2025080403394033400_ref21","doi-asserted-by":"publisher","first-page":"4236","DOI":"10.1038\/s41467-025-59422-w","article-title":"A protein language model for exploring viral fitness landscapes","volume":"16","author":"Ito","year":"2025","journal-title":"Nat Commun"},{"key":"2025080403394033400_ref22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1706.03762","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"key":"2025080403394033400_ref23","doi-asserted-by":"publisher","first-page":"284","DOI":"10.1126\/science.abd7331","article-title":"Learning the language of viral evolution and escape","volume":"371","author":"Hie","year":"2021","journal-title":"Science"},{"key":"2025080403394033400_ref24","doi-asserted-by":"publisher","first-page":"4348","DOI":"10.1038\/s41467-022-32007-7","article-title":"ProtGPT2 is a deep unsupervised language model for protein design","volume":"13","author":"Ferruz","year":"2022","journal-title":"Nat Commun"},{"key":"2025080403394033400_ref25","doi-asserted-by":"publisher","first-page":"1282","DOI":"10.1093\/bioinformatics\/btm098","article-title":"UniRef: Comprehensive and non-redundant UniProt reference clusters","volume":"23","author":"Suzek","year":"2007","journal-title":"Bioinformatics"},{"key":"2025080403394033400_ref26","doi-asserted-by":"publisher","first-page":"D570","DOI":"10.1093\/nar\/gkz1035","article-title":"MGnify: The microbiome analysis resource in 2020","volume":"48","author":"Mitchell","year":"2020","journal-title":"Nucleic Acids Res"},{"key":"2025080403394033400_ref27","doi-asserted-by":"publisher","first-page":"2542","DOI":"10.1038\/s41467-018-04964-5","article-title":"Clustering huge protein sequence sets in linear time","volume":"9","author":"Steinegger","year":"2018","journal-title":"Nat Commun"},{"key":"2025080403394033400_ref28","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2211.16742","article-title":"Protein language models and structure prediction: Connection and progression","author":"Hu","year":"2022"},{"key":"2025080403394033400_ref29","doi-asserted-by":"publisher","first-page":"vead022","DOI":"10.1093\/ve\/vead022","article-title":"MutaGAN: A sequence-to-sequence GAN framework to predict mutations of evolving protein populations","volume":"9","author":"Berman","year":"2023","journal-title":"Virus Evolution"},{"key":"2025080403394033400_ref30","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1038\/s41586-020-2179-y","article-title":"Structural basis of receptor recognition by SARS-CoV-2","volume":"581","author":"Shang","year":"2020","journal-title":"Nature"},{"key":"2025080403394033400_ref31","doi-asserted-by":"publisher","first-page":"W417","DOI":"10.1093\/nar\/gkab273","article-title":"mmCSM-PPI: Predicting the effects of multiple point mutations on protein\u2013protein interactions","volume":"49","author":"Rodrigues","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"2025080403394033400_ref32","doi-asserted-by":"publisher","first-page":"eabe5575","DOI":"10.1126\/sciadv.abe5575","article-title":"Conformational dynamics of SARS-CoV-2 trimeric spike glycoprotein in complex with receptor ACE2 revealed by cryo-EM","volume":"7","author":"Xu","year":"2021","journal-title":"Sci Adv"},{"key":"2025080403394033400_ref33","doi-asserted-by":"publisher","first-page":"5886","DOI":"10.1038\/s41467-022-33309-6","article-title":"DNA-delivered antibody cocktail exhibits improved pharmacokinetics and confers prophylactic protection against SARS-CoV-2","volume":"13","author":"Parzych","year":"2022","journal-title":"Nat Commun"},{"key":"2025080403394033400_ref34","doi-asserted-by":"publisher","first-page":"e1298","DOI":"10.1002\/wcms.1298","article-title":"Using PyMOL as a platform for computational drug design","volume":"7","author":"Yuan","year":"2017","journal-title":"Wiley Interdiscip Rev Comput Mol Sci"},{"key":"2025080403394033400_ref35","doi-asserted-by":"publisher","first-page":"856","DOI":"10.3390\/v15040856","article-title":"Understanding mutations in human SARS-CoV-2 spike glycoprotein: A Systematic Review & Meta-Analysis","volume":"15","author":"Kumar","year":"2023","journal-title":"Viruses"},{"key":"2025080403394033400_ref36","doi-asserted-by":"publisher","first-page":"1195","DOI":"10.3390\/vaccines9101195","article-title":"Emerging SARS-CoV-2 variants: A review of its mutations, its implications and vaccine efficacy","volume":"9","author":"Ramesh","year":"2021","journal-title":"Vaccines (Basel)"},{"key":"2025080403394033400_ref37","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1016\/j.crstbi.2022.01.002","article-title":"Mutations in human SARS-CoV-2 spike proteins, potential drug binding and epitope sites for COVID-19 therapeutics development","volume":"4","author":"Guruprasad","year":"2022","journal-title":"Curr Res Struct Biol"},{"key":"2025080403394033400_ref38","doi-asserted-by":"publisher","first-page":"100682","DOI":"10.1016\/j.genrep.2020.100682","article-title":"Genomic characterization of a novel SARS-CoV-2","volume":"19","author":"Khailany","year":"2020","journal-title":"Gene Reports"},{"key":"2025080403394033400_ref39","doi-asserted-by":"publisher","first-page":"112621","DOI":"10.1016\/j.celrep.2023.112621","article-title":"Structural changes in the SARS-CoV-2 spike E406W mutant escaping a clinical monoclonal antibody cocktail","volume":"42","author":"Addetia","year":"2023","journal-title":"Cell Rep"},{"key":"2025080403394033400_ref40","doi-asserted-by":"publisher","first-page":"384","DOI":"10.1001\/jama.2021.24931","article-title":"Tixagevimab and Cilgavimab (Evusheld) for pre-exposure prophylaxis of COVID-19","volume":"327","year":"2022","journal-title":"JAMA"},{"key":"2025080403394033400_ref41","doi-asserted-by":"publisher","first-page":"4152","DOI":"10.1021\/acs.jcim.1c00990","article-title":"SARS-CoV-2 variants are selecting for spike protein mutations that increase protein stability","volume":"61","author":"Shorthouse","year":"2021","journal-title":"J Chem Inf Model"},{"key":"2025080403394033400_ref42","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1038\/s41586-021-04245-0","article-title":"The N501Y spike substitution enhances SARS-CoV-2 infection and transmission","volume":"602","author":"Liu","year":"2022","journal-title":"Nature"},{"key":"2025080403394033400_ref43","doi-asserted-by":"publisher","first-page":"657","DOI":"10.1038\/s41586-021-04385-3","article-title":"Omicron escapes the majority of existing SARS-CoV-2 neutralizing antibodies","volume":"602","author":"Cao","year":"2022","journal-title":"Nature"},{"key":"2025080403394033400_ref44","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/S1473-3099(22)00733-2","article-title":"Omicron sublineage BQ.1.1 resistance to monoclonal antibodies","volume":"23","author":"Arora","year":"2023","journal-title":"Lancet Infect Dis"},{"key":"2025080403394033400_ref45","doi-asserted-by":"publisher","first-page":"553","DOI":"10.1038\/s41586-022-04594-4","article-title":"Antibody evasion properties of SARS-CoV-2 omicron sublineages","volume":"604","author":"Iketani","year":"2022","journal-title":"Nature"},{"key":"2025080403394033400_ref46","doi-asserted-by":"publisher","first-page":"6183","DOI":"10.1109\/BIBM62325.2024.10822644","volume-title":"2024 IEEE international conference on bioinformatics and biomedicine (BIBM)","author":"Rancati"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/4\/bbaf384\/63926290\/bbaf384.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/4\/bbaf384\/63926290\/bbaf384.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T07:39:49Z","timestamp":1754293189000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf384\/8221601"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":46,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,7,2]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf384","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,7]]},"published":{"date-parts":[[2025,7]]},"article-number":"bbaf384"}}