Catching enhanced flexibility in antibody-antigen recognition: critical evaluation of physics-based and AI-driven methods.
- 1Engineering Research Center of Cell & Therapeutic Antibody (MOE), School of Pharmacy, Shanghai Jiao Tong University, Shanghai, China.
- 2Shanghai Digiwiser Biotechnology, Shanghai, China.
- 3Biomolecular Physics and Modeling Group, School of Physics, Huazhong University of Science and Technology, Wuhan, Hubei, China.
Abstract
Protein conformational dynamics often govern protein-protein interaction mechanisms and function, as flexibility enables adaptation to different environments. Yet, the extent to which current physics‑based and emerging AI‑driven methods can predict flexibility‑dependent physicochemical properties remains to be determined. Here, we studied the HyHEL10 (Hy10) antibody lineage binding to lysozymes with different conformational flexibility. Contrary to traditional models, the affinity-matured Hy104X antibody achieves high affinity by actively destabilizing its CDRH1 loop, enabling the capture of mobile antigen variants via entropy-buffered conformational selection. This structural plasticity presents a significant challenge to standard computational protocols. We found that the polarization effect is enhanced with increasing protein flexibility. Our benchmarking of MM/PBSA and AI-driven structure predictors reveals that both approaches initially fail to accurately rank affinities in more flexible systems, primarily due to sampling limitations and inaccurate structure prediction. However, the inclusion of structural templates in Chai-1 effectively mitigated these limitations, leading to enhanced correlation with experimental affinities. In the future, by coupling AI's quick screening with robust physics-based evaluations, we may have a potential pathway for in silico engineering of therapeutic antibodies against the most elusive, shapeshifting targets in the proteome.
Related
- Antibody CDR-H3 loop flexibility: Insights from X-ray crystallography, structural bioinformatics, and the limits of current deep learning methods
- SMARTFlexDB: a paired apo–holo RNA structure database and analysis platform for RNA conformational flexibility and small-molecule recognition
- Structural basis of diverse antibody recognition of conserved coronavirus spike S2 epitopes that contribute to protective immunity
- Comprehensive evaluation of antibody language models across multiple antibody design tasks
- Peptide Stapling: A Strategy for Enhanced Molecular Recognition of C-Reactive Protein
BibTeX
@article{xu2026catching,
title = {{Catching enhanced flexibility in antibody-antigen recognition: critical evaluation of physics-based and AI-driven methods.}},
author = {Honglin Xu and Ruyue Mi and Ganggang Bai and Yue Kang and Chuance Sun and Yanjing Wang and Changjun Chen and Buyong Ma},
journal = {MAbs},
year = {2026},
month = oct,
doi = {10.1080/19420862.2026.2743404},
url = {https://europepmc.org/article/MED/42859400},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Catching enhanced flexibility in antibody-antigen recognition: critical evaluation of physics-based and AI-driven methods.
AU - Honglin Xu
AU - Ruyue Mi
AU - Ganggang Bai
AU - Yue Kang
AU - Chuance Sun
AU - Yanjing Wang
AU - Changjun Chen
AU - Buyong Ma
PY - 2026
DA - 2026/10/10
JO - MAbs
DO - 10.1080/19420862.2026.2743404
UR - https://europepmc.org/article/MED/42859400
AB - Protein conformational dynamics often govern protein-protein interaction mechanisms and function, as flexibility enables adaptation to different environments. Yet, the extent to which current physics‑based and emerging AI‑driven methods can predict flexibility‑dependent physicochemical properties remains to be determined. Here, we studied the HyHEL10 (Hy10) antibody lineage binding to lysozymes with different conformational flexibility. Contrary to traditional models, the affinity-matured Hy104X antibody achieves high affinity by actively destabilizing its CDRH1 loop, enabling the capture of mobile antigen variants via entropy-buffered conformational selection. This structural plasticity presents a significant challenge to standard computational protocols. We found that the polarization effect is enhanced with increasing protein flexibility. Our benchmarking of MM/PBSA and AI-driven structure predictors reveals that both approaches initially fail to accurately rank affinities in more flexible systems, primarily due to sampling limitations and inaccurate structure prediction. However, the inclusion of structural templates in Chai-1 effectively mitigated these limitations, leading to enhanced correlation with experimental affinities. In the future, by coupling AI's quick screening with robust physics-based evaluations, we may have a potential pathway for in silico engineering of therapeutic antibodies against the most elusive, shapeshifting targets in the proteome.
SN - 1942-0862
SN - 1942-0870
ER -