An antigen-specific protein language model guides CDR optimization for antibody design
PLOS Computational Biology · 11.11.2026 · 22 (11) · e1014924
DOI: 10.1371/journal.pcbi.1014924 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 11.11.2026. Verfasst von Faisal Bin Ashraf und 17 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, University of California, Riverside, California, United States of America. Die Arbeit nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Division of Microbiology and Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Science Foundation, Fundação de Amparo à Pesquisa do Estado de São Paulo.
Abstract (Originalsprache)
The capacity of an antibody to recognize and bind an antigen with high specificity and strength (i.e., its binding affinity ) is a key determinant in the design and optimization of neutralizing antibodies against pathogenic targets. Here we introduce Ab-Affinity, a fine-tuned protein language model (PLM) that predicts the binding affinity of an antibody against a target peptide within the SARS-CoV-2 spike protein. Cross-validation results shows that that Ab-Affinity outperforms other PLM-based affinity predictors when evaluated on a held-out validation and independent test sets. When integrated with a genetic algorithm and simulated annealing for sequence diversification and fitness optimization, Ab-Affinity enables the generation of candidate antibodies exhibiting over 100-fold improvement in predicted binding affinity relative to its parental (natural) antibody. More broadly, antibodies designed with Ab-Affinity attain higher predicted binding affinities than those produced by alternative mutational strategies, yielding synthetic antibodies with favorable predicted biophysical characteristics. Molecular docking and molecular dynamics simulations for the strongest synthetic candidates indicate enhanced interactions and increased stability with the target peptide epitope within the spike protein. Experimental validation in vitro for a top-ranked candidate reveals an apparent increase in binding relative to its parental antibody. All model weights, source code, and analysis pipelines are publicly released to allow full reproducibility.
Autor:innen und Institutionen
- Department of Computer Science and Engineering, University of California, Riverside, California, United States of America.
- Henry E. Riggs School of Applied Life Science, Keck Graduate Institute, Claremont, California, United States of America.
- Department of Chemistry, Pomona College, Claremont, California, United States of America.
- Departamento de Ciências Biomoleculares, Universidade de São Paulo, São Paulo, Brazil.
- Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, North Carolina, United States of America.
ORCID-Profile: Faisal Bin Ashraf, Christian J Dávila Ojeda, Fernando L Barroso da Silva, Animesh Ray
Förderung
- Division of Microbiology and Infectious Diseases, National Institute of Allergy and Infectious Diseases
- National Science Foundation
- Fundação de Amparo à Pesquisa do Estado de São Paulo
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
Zitieren
APA
Ashraf, F. B., Paco, K., Zebardast, S., Zhang, Z., Mooney, R., Dávila Ojeda, C. J., Mendivil, M. P., Alam, R. M., Chen, E., Ogu, I., Lay, J. A., Yang, T. Y., Barroso da Silva, F. L., Johal, M., Tolstorukov, I., Sazinsky, M. H., Ray, A., & Lonardi, S. (2026). An antigen-specific protein language model guides CDR optimization for antibody design. PLOS Computational Biology, 22(11), e1014924. https://doi.org/10.1371/journal.pcbi.1014924
Vancouver
Ashraf FB, Paco K, Zebardast S, Zhang Z, Mooney R, Dávila Ojeda CJ, et al. An antigen-specific protein language model guides CDR optimization for antibody design. PLOS Computational Biology. 2026;22(11):e1014924. doi:10.1371/journal.pcbi.1014924
BibTeX
@article{Ashraf2026,
author = {Ashraf, Faisal Bin and Paco, Karen and Zebardast, Sanaz and Zhang, Zihao and Mooney, Ryan and Dávila Ojeda, Christian J and Mendivil, Mariana Paco and Alam, Rashid M and Chen, Eva and Ogu, Iraneus and Lay, Jordan A and Yang, Tristan Y and Barroso da Silva, Fernando L and Johal, Malkiat and Tolstorukov, Ilya and Sazinsky, Matthew H and Ray, Animesh and Lonardi, Stefano},
title = {An antigen-specific protein language model guides CDR optimization for antibody design},
journal = {PLOS Computational Biology},
year = {2026},
volume = {22},
number = {11},
pages = {e1014924},
doi = {10.1371/journal.pcbi.1014924}
}
EUVIMED