Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction
PLOS Computational Biology · 31.03.2025 · 21 (3) · e1012153
DOI: 10.1371/journal.pcbi.1012153 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 31.03.2025. Verfasst von Meng Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Program in Computational Biology and Bioinformatics, Yale University, New Haven, Connecticut, United States of America. Hauptthemen laut MEDLINE: Antibodies, Viral, Models, Immunological, Antibody Specificity. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NIAID NIH HHS, NIGMS NIH HHS.
Abstract (Originalsprache)
Antibodies play a crucial role in the adaptive immune response, with their specificity to antigens being a fundamental determinant of immune function. Accurate prediction of antibody-antigen specificity is vital for understanding immune responses, guiding vaccine design, and developing antibody-based therapeutics. In this study, we present a method of supervised fine-tuning for antibody language models, which improves on pre-trained antibody language model embeddings in binding specificity prediction to SARS-CoV-2 spike protein and influenza hemagglutinin. We perform supervised fine-tuning on four pre-trained antibody language models to predict specificity to these antigens and demonstrate that fine-tuned language model classifiers exhibit enhanced predictive accuracy compared to classifiers trained on pre-trained model embeddings. Additionally, we investigate the change of model attention activations after supervised fine-tuning to gain insights into the molecular basis of antigen recognition by antibodies. Furthermore, we apply the supervised fine-tuned models to BCR repertoire data related to influenza and SARS-CoV-2 vaccination, demonstrating their ability to capture changes in repertoire following vaccination. Overall, our study highlights the effect of supervised fine-tuning on pre-trained antibody language models as valuable tools to improve antigen specificity prediction.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Program in Computational Biology and Bioinformatics, Yale University, New Haven, Connecticut, United States of America.
- Program in Applied Mathematics, Yale University, New Haven, Connecticut, United States of America.
- Department of Pathology, Yale School of Medicine, New Haven, Connecticut, United States of America.
- Department of Immunobiology, Yale School of Medicine, New Haven, Connecticut, United States of America.
ORCID-Profile: Meng Wang, Steven H Kleinstein
Förderung
- NCI NIH HHS
- NIAID NIH HHS
- NIGMS NIH HHS
- NIH
Zitieren
APA
Wang, M., Patsenker, J., Li, H., Kluger, Y., & Kleinstein, S. H. (2025). Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction. PLOS Computational Biology, 21(3), e1012153. https://doi.org/10.1371/journal.pcbi.1012153
Vancouver
Wang M, Patsenker J, Li H, Kluger Y, Kleinstein SH. Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction. PLOS Computational Biology. 2025;21(3):e1012153. doi:10.1371/journal.pcbi.1012153
BibTeX
@article{Wang2025,
author = {Wang, Meng and Patsenker, Jonathan and Li, Henry and Kluger, Yuval and Kleinstein, Steven H},
title = {Supervised fine-tuning of pre-trained antibody language models improves antigen specificity prediction},
journal = {PLOS Computational Biology},
year = {2025},
volume = {21},
number = {3},
pages = {e1012153},
doi = {10.1371/journal.pcbi.1012153}
}
EUVIMED