Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design
- 1AstraZeneca
Abstract
Antibodies recognise their targets through hypervariable complementarity-determining regions (CDRs), which are interleaved with conserved frameworks in sequence space, making de novo CDR design an infilling problem. Autoregressive models generate residues left-to-right, which precludes full framework context during CDR generation and conflates framework and CDR likelihoods, leaving no natural prompt-response interface for feedback to steer generation. We present GenCDR, a family of LLaMa-based autoregressive language models that read all frameworks as a conditioning prompt and generate all CDRs jointly as a variable-length response, making CDR likelihoods a clean, separable target for reward attribution. The family comprises IgGenCDR, p-IgGenCDR, and NanoGenCDR, trained on unpaired, paired, and nanobody chains, respectively. GenCDR achieves the highest CDR recovery among autoregressive models and produces natural, diverse, human-like CDRs whose likelihoods correlate with fitness and developability assays. The prompt-response boundary also enables principled alignment: reward signals for binding affinity, expression, or developability can be composed to steer CDR generation. Over four rounds of alignment against antibody-antigen co-folding and developability objectives, we find that NanoGenCDR, which uses no explicit antigen encoding, can reach in silico structural interface metrics competitive with those of a structure-conditioned diffusion pipeline at roughly half the sampling budget, with more natural, developable designs. The same interface can be extended to integrate experimental feedback, opening a path to closed-loop antibody de novo design.
Related
- Generative Design of New-to-nature Biosynthetic Assembly Lines with Genomic Language Modeling
- Antibody-Antigen Affinity Prediction with Chain-Aware Protein Language Modeling
- Comprehensive evaluation of antibody language models across multiple antibody design tasks
- What Shapes Brain Alignment of Large Language Models during Dialogue? Insights from Language Production and Comprehension
- Generative design of bacteriophages with genome language models
BibTeX
@misc{hernandez2026generative,
title = {{Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design}},
author = {F. Gonzalez Hernandez and O. M. Turnbull and M. Sultana and L. Roldan-Martin and R. J. Kumar and T. Diethe and R. Croasdale-Wood and C. Deane and D. Oglic},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.06.749721},
url = {https://www.biorxiv.org/content/10.64898/2026.09.06.749721v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Generative Language Modeling for Antibody CDR Grafting and Alignment-driven De Novo Design
AU - F. Gonzalez Hernandez
AU - O. M. Turnbull
AU - M. Sultana
AU - L. Roldan-Martin
AU - R. J. Kumar
AU - T. Diethe
AU - R. Croasdale-Wood
AU - C. Deane
AU - D. Oglic
PY - 2026
DA - 2026/09/06
PB - bioRxiv
DO - 10.64898/2026.09.06.749721
UR - https://www.biorxiv.org/content/10.64898/2026.09.06.749721v1
AB - Antibodies recognise their targets through hypervariable complementarity-determining regions (CDRs), which are interleaved with conserved frameworks in sequence space, making de novo CDR design an infilling problem. Autoregressive models generate residues left-to-right, which precludes full framework context during CDR generation and conflates framework and CDR likelihoods, leaving no natural prompt-response interface for feedback to steer generation. We present GenCDR, a family of LLaMa-based autoregressive language models that read all frameworks as a conditioning prompt and generate all CDRs jointly as a variable-length response, making CDR likelihoods a clean, separable target for reward attribution. The family comprises IgGenCDR, p-IgGenCDR, and NanoGenCDR, trained on unpaired, paired, and nanobody chains, respectively. GenCDR achieves the highest CDR recovery among autoregressive models and produces natural, diverse, human-like CDRs whose likelihoods correlate with fitness and developability assays. The prompt-response boundary also enables principled alignment: reward signals for binding affinity, expression, or developability can be composed to steer CDR generation. Over four rounds of alignment against antibody-antigen co-folding and developability objectives, we find that NanoGenCDR, which uses no explicit antigen encoding, can reach in silico structural interface metrics competitive with those of a structure-conditioned diffusion pipeline at roughly half the sampling budget, with more natural, developable designs. The same interface can be extended to integrate experimental feedback, opening a path to closed-loop antibody de novo design.
ER -