EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Nucleotide context models outperform protein language models for predicting antibody affinity maturation

Mackenzie M Johnson iD, Kevin Sung, Hugh K Haddox iD, Ashni A Vora, Tatsuya Araki iD, Gabriel D Victora, Yun S Song, Julia Fukuyama, Frederick A Matsen Iv iD

PLOS Computational Biology · 01.12.2025 · 21 (12) · e1013758

DOI: 10.1371/journal.pcbi.1013758 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 01.12.2025. Verfasst von Mackenzie M Johnson und 8 weiteren Autor:innen. Beteiligte Einrichtung: Computational Biology Program, Fred Hutchinson Cancer Center, Seattle, Washington, United States of America. Hauptthemen laut MEDLINE: Antibody Affinity, Nucleotides. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIAID NIH HHS, NHGRI NIH HHS, NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Antibodies play a crucial role in adaptive immunity. They develop as B cell receptors (BCRs): membrane-bound forms of antibodies that are expressed on the surfaces of B cells. BCRs are refined through affinity maturation, a process of somatic hypermutation (SHM) and natural selection, to improve binding to an antigen. Computational models of affinity maturation have developed from two main perspectives: molecular evolution and language modeling. The molecular evolution perspective focuses on nucleotide sequence context to describe mutation and selection; the language modeling perspective involves learning patterns from large data sets of protein sequences. In this paper, we compared models from both perspectives on their ability to predict the course of antibody affinity maturation along phylogenetic trees of BCR sequences. This included models of SHM, models of SHM combined with an estimate of selection, and protein language models. We evaluated these models for large human BCR repertoire data sets, as well as an antigen-specific mouse experiment with a pre-rearranged cognate naive antibody. We demonstrated that precise modeling of SHM, which requires the nucleotide context, provides a substantial amount of predictive power for predicting the course of affinity maturation. Notably, a simple nucleotide-based convolutional neural network modeling SHM outperformed state-of-the-art protein language models, including one trained exclusively on antibody sequences. Furthermore, incorporating estimates of selection based on a custom deep mutational scanning experiment brought only modest improvement in predictive power. To support further research, we introduce EPAM (Evaluating Predictions of Affinity Maturation), a benchmarking framework to integrate evolutionary principles with advances in language modeling, offering a road map for understanding antibody evolution and improving predictive models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Antibody AffinityNucleotidesAnimalsHumansMiceReceptors, Antigen, B-CellComputational BiologyB-LymphocytesEvolution, MolecularSomatic Hypermutation, ImmunoglobulinPhylogeny

Substanzen

Receptors, Antigen, B-CellNucleotides

Autor:innen und Institutionen

  • Computational Biology Program, Fred Hutchinson Cancer Center, Seattle, Washington, United States of America.
  • Laboratory of Lymphocyte Dynamics, The Rockefeller University, New York, New York, United States of America.
  • Howard Hughes Medical Institute, Chevy Chase, Maryland, United States of America.
  • Computer Science Division, University of California, Berkeley, Berkeley, California, United States of America.
  • Department of Statistics, University of California, Berkeley, Berkeley, California, United States of America.
  • Department of Statistics, Indiana University Bloomington, Bloomington, Indiana, United States of America.

ORCID-Profile: Mackenzie M Johnson, Hugh K Haddox, Tatsuya Araki, Frederick A Matsen Iv

Förderung

  • NIAID NIH HHS
  • NHGRI NIH HHS
  • NIH HHS
  • National Institutes of Health
  • Howard Hughes Medical Institute

Zitieren

APA

Johnson, M. M., Sung, K., Haddox, H. K., Vora, A. A., Araki, T., Victora, G. D., Song, Y. S., Fukuyama, J., & Matsen Iv, F. A. (2025). Nucleotide context models outperform protein language models for predicting antibody affinity maturation. PLOS Computational Biology, 21(12), e1013758. https://doi.org/10.1371/journal.pcbi.1013758

Vancouver

Johnson MM, Sung K, Haddox HK, Vora AA, Araki T, Victora GD, et al. Nucleotide context models outperform protein language models for predicting antibody affinity maturation. PLOS Computational Biology. 2025;21(12):e1013758. doi:10.1371/journal.pcbi.1013758

BibTeX
@article{Johnson2025,
  author = {Johnson, Mackenzie M and Sung, Kevin and Haddox, Hugh K and Vora, Ashni A and Araki, Tatsuya and Victora, Gabriel D and Song, Yun S and Fukuyama, Julia and Matsen Iv, Frederick A},
  title = {Nucleotide context models outperform protein language models for predicting antibody affinity maturation},
  journal = {PLOS Computational Biology},
  year = {2025},
  volume = {21},
  number = {12},
  pages = {e1013758},
  doi = {10.1371/journal.pcbi.1013758}
}
Kostenlos starten →