{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T05:45:34Z","timestamp":1785908734044,"version":"3.56.0"},"reference-count":17,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T00:00:00Z","timestamp":1785888000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Immunol."],"abstract":"<jats:p>Antiviral antibody discovery is constrained by the enormous combinatorial space of antibody paratopes and viral epitopes, while novel approaches to search for these vast possibilities such as machine learning models for antibody affinity prediction typically require large target-specific datasets that are rarely available for emerging pathogens. To address this problem, we investigated the ability of transfer learning to reduce the required training data and still create a robust predictive model of antibody affinities. We show that antibody-affinity prediction can be locally transferred from a data-rich viral target to a related data-limited viral target by combining transfer learning, epitope-centered antigen encoding, antibody CDR descriptors, and targeted experimental validation. Using SARS-CoV-2 receptor-binding-domain (RBD) interaction data as the source domain, we developed dual-encoder neural-network models that capture functional determinants of antibody recognition and supports prediction of SARS-CoV-1 variant:antibody affinities despite limited direct training data. High-throughput ELISA measurements generated compact validation datasets spanning heavy-chain-only antibodies (HCAbs) and conventional antibodies, enabling iterative model refinement through similarity-guided, model-informed data expansion. Across expanded withheld-antibody evaluations, the final models showed improved but antibody-dependent predictive performance, with mean Pearson correlations of 0.71 \u00b1 0.12 for SARS-CoV-2 and 0.63 \u00b1 0.09 for SARS-CoV-1, and best-performing holdouts reaching 0.85 and 0.78, respectively. These results demonstrate that existing immunological datasets can support feature-constrained transfer learning for data-efficient prioritization of antibody:antigen interactions across closely related Sarbecoviruses, particularly when conserved epitope regions can be aligned and limited target-specific measurements are available for calibration.<\/jats:p>","DOI":"10.3389\/fimmu.2026.1886934","type":"journal-article","created":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T04:54:50Z","timestamp":1785905690000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Modeling antibody recognition across SARS-CoV-2 and SARS-CoV-1 using epitope-informed transfer learning"],"prefix":"10.3389","volume":"17","author":[{"given":"Jimmy B.","family":"Yuan","sequence":"first","affiliation":[{"name":"Biotechnology and Bioengineering, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ryan C.","family":"Bruneau","sequence":"additional","affiliation":[{"name":"Biotechnology and Bioengineering, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephen","family":"Won","sequence":"additional","affiliation":[{"name":"Biotechnology and Bioengineering, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aidan Elliot","family":"Heller","sequence":"additional","affiliation":[{"name":"Bioresource and Environmental Security, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas","family":"Sheffield","sequence":"additional","affiliation":[{"name":"Biosecurity and Bioassurance, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kenneth L.","family":"Sale","sequence":"additional","affiliation":[{"name":"Biosecurity and Bioassurance, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brooke","family":"Harmon","sequence":"additional","affiliation":[{"name":"Biotechnology and Bioengineering, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Le Thanh Mai","family":"Pham","sequence":"additional","affiliation":[{"name":"Bioresource and Environmental Security, Sandia National Laboratories","place":["Livermore, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,5]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1080\/19420862.2020.1729683","article-title":"Understanding the human antibody repertoire","volume":"12","author":"Rees","year":"2020","journal-title":"MAbs"},{"key":"B2","doi-asserted-by":"publisher","first-page":"787","DOI":"10.1098\/rsif.2007.0229","article-title":"Patterns of antigenic diversity and the mechanisms that maintain them","volume":"4","author":"Lipsitch","year":"2007","journal-title":"J R Soc Interface"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.032405199","article-title":"Protein topology and stability define the space of allowed sequences","volume":"99","author":"Koehl","year":"2002","journal-title":"Proc Natl Acad Sci USA"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1295","DOI":"10.1016\/j.cell.2020.08.012","article-title":"Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding","volume":"182","author":"Starr","year":"2020","journal-title":"Cell"},{"key":"B5","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1016\/j.chom.2021.02.003","article-title":"Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies","volume":"29","author":"Greaney","year":"2021","journal-title":"Cell Host Microbe"},{"key":"B6","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1038\/s41580-021-00407-0","article-title":"A guide to machine learning for biologists","volume":"23","author":"Greener","year":"2022","journal-title":"Nat Rev Mol Cell Biol"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1038\/s41551-021-00699-9","article-title":"Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning","volume":"5","author":"Mason","year":"2021","journal-title":"Nat BioMed Eng"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1038\/s43588-025-00823-8","article-title":"Investigating the volume and diversity of data needed for generalizable antibody-antigen \u0394\u0394G prediction","volume":"5","author":"Hummer","year":"2025","journal-title":"Nat Comput Sci"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1014394","article-title":"Combining machine learning and iterative experiments to keep pace with emerging viral variants of concern","volume":"22","author":"Pham","year":"2026","journal-title":"PloS Comput Biol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"W533","DOI":"10.1093\/nar\/gkae386","article-title":"SEMA 2.0: web-platform for B-cell conformational epitopes prediction using artificial intelligence","volume":"52","author":"Ivanisenko","year":"2024","journal-title":"Nucleic Acids Res"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-5-113","article-title":"MUSCLE: a multiple sequence alignment method with reduced time and space complexity","volume":"5","author":"Edgar","year":"2004","journal-title":"BMC Bioinf"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1008864","article-title":"Using the antibody-antigen binding interface to train image-based deep neural networks for antibody-epitope classification","volume":"17","author":"Ripoll","year":"2021","journal-title":"PloS Comput Biol"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1009380","article-title":"Structural and energetic profiling of SARS-CoV-2 receptor binding domain antibody recognition and the impact of circulating variants","volume":"17","author":"Yin","year":"2021","journal-title":"PloS Comput Biol"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-019-0496-6","article-title":"Machine-learning-guided directed evolution for protein engineering","volume":"16","author":"Yang","year":"2019","journal-title":"Nat Methods"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1011175","article-title":"Enabling interpretable machine learning for biological data with reliability scores","volume":"19","author":"Ahlquist","year":"2023","journal-title":"PloS Comput Biol"},{"key":"B16","doi-asserted-by":"publisher","first-page":"876","DOI":"10.1038\/s42003-023-05133-1","article-title":"Integration of pre-trained protein language models into geometric deep learning networks","volume":"6","author":"Wu","year":"2023","journal-title":"Commun Biol"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1128\/jvi.79.3.1635-1644.2005","article-title":"Molecular and biological characterization of human monoclonal antibodies binding to the spike and nucleocapsid proteins of severe acute respiratory syndrome coronavirus","volume":"79","author":"van den Brink","year":"2005","journal-title":"J Virol"}],"container-title":["Frontiers in Immunology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2026.1886934\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T04:54:55Z","timestamp":1785905695000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2026.1886934\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,5]]},"references-count":17,"alternative-id":["10.3389\/fimmu.2026.1886934"],"URL":"https:\/\/doi.org\/10.3389\/fimmu.2026.1886934","relation":{},"ISSN":["1664-3224"],"issn-type":[{"value":"1664-3224","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,5]]},"article-number":"1886934"}}