SimBinder-IF: structure-aware antibody affinity optimization via efficient preference learning
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag548 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Xinyan Zhao und 9 weiteren Autor:innen. Beteiligte Einrichtung: McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, Houston, TX 77030, United States. Hauptthemen laut MEDLINE: Antibody Affinity, Antibodies, Computational Biology. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIA NIH HHS, National Institute on Aging of the National Institutes of Health.
Abstract (Originalsprache)
Motivation Antibody therapeutic efficacy depends on high-affinity target engagement, yet laboratory affinity maturation is slow and costly. Most protein language models (PLMs) lack explicit training for high affinity, and current preference optimization methods introduce computational overhead without clear affinity improvements. Therefore, structure-aware and parameter-efficient approaches for antibody affinity optimization are urgently needed. Results We propose SimBinder-IF, a structure-aware antibody optimization model trained by freezing the Evolutionary Scale Modeling inverse folding (ESM-IF) structure encoder and fine-tuning only its decoder via Simple Preference Optimization (SimPO) to prefer stronger binders. In generalization tests across seven held-out complexes (93 477 mutants), SimBinder-IF shows a 22% numerical increase in average Spearman correlation (0.26 to 0.32) compared with vanilla ESM-IF and outperforms ESM-IF on five of seven assays; however, the assay-level paired Wilcoxon test does not reach statistical significance (p=0.188). For seed-averaged top-ranking enrichment, SimBinder-IF has the highest 20-fold improvement@20 mean (0.453), whereas ESM-IF has the highest 10-fold improvement@10 mean (0.520 vs. 0.487 for SimBinder-IF). In a case study redesigning antibody F045-092 to target the A/California/04/2009 pandemic H1N1 (pdmH1N1) strain—a target the original antibody fails to bind—SimBinder-IF generates variants with markedly lower predicted binding free energy than ESM-IF (mean ΔΔG: −68.56 vs. −49.77 kcal/mol). Notably, SimBinder-IF updates only 18% of the ESM-IF parameters. Regarding generic protein–protein affinity modelling, SimBinder-IF shows class-dependent preservation with trade-offs, improving performance on antibody–antigen (AB/AG) and protease–protein inhibitor (Pr/PI) complexes while yielding slightly lower aggregate mean correlations than ESM-IF and ProteinDPO. Availability and implementation The source code and model weights for SimBinder-IF are available at https://github.com/MSBMI-SAFE/SimBinder-IF.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, Houston, TX 77030, United States.
- Department of Computer Science, The University of Texas at Austin, Austin, TX 78712, United States.
- Texas Therapeutics Institute, Brown Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston, Houston, TX 77030, United States.
- Computer Science Department, Worcester Polytechnic Institute, Worcester, MA 01609, United States.
ORCID-Profile: Xinyan Zhao, Xiaozhong Liu, Zhiqiang An, Xiaoqian Jiang, Yejin Kim
Förderung
- NIA NIH HHS
- National Institute on Aging of the National Institutes of Health
Zitieren
APA
Zhao, X., Tang, Y. C., Monsia, R., Cantu, V. J., Ramesh, A. K., Yang, S., Liu, X., An, Z., Jiang, X., & Kim, Y. (2026). SimBinder-IF: structure-aware antibody affinity optimization via efficient preference learning. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag548
Vancouver
Zhao X, Tang YC, Monsia R, Cantu VJ, Ramesh AK, Yang S, et al. SimBinder-IF: structure-aware antibody affinity optimization via efficient preference learning. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag548
BibTeX
@article{Zhao2026,
author = {Zhao, Xinyan and Tang, Yi-Ching and Monsia, Rivaaj and Cantu, Victor J and Ramesh, Ashwin Kumar and Yang, Siyu and Liu, Xiaozhong and An, Zhiqiang and Jiang, Xiaoqian and Kim, Yejin},
title = {SimBinder-IF: structure-aware antibody affinity optimization via efficient preference learning},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag548}
}
EUVIMED