Repurposing AlphaFold3-like Protein Folding Models for Antibody Sequence and Structure Co-design

Junchi Yan (Shanghai Jiao Tong University) · Shuangjia Zheng (Shanghai Jiaotong University) · Nianzu Yang (Shanghai Jiao Tong University) · Songlin Jiang (Shanghai Jiao Tong University) · Jian Ma (Shanghai Jiaotong University) · Huaijin Wu (Shanghai Jiao Tong University) · Wengong Jin (Northeastern University)

alphafold3antibody designantibody-antigen complexesantigen-conditioned modelbenchmark resultsbiomolecular interactioncdr-h3 recovery ratediffusion models fine-tuningnanobodiesprotein folding modelssequence diffusion headsequence-structure co-diffusionstate-of-the-art methodsstructure diffusion module

Diffusion models hold great potential for accelerating antibody design, but their performance is so far limited by the number of antibody-antigen complexes used for model training. Meanwhile, AlphaFold3-like protein folding models, pre-trained on a large corpus of crystal structures, have acquired a broad understanding of biomolecular interaction. Based on this insight, we develop a new antigen-conditioned antibody design model by adapting the diffusion module of AlphaFold3-like models for sequence-structure co-diffusion. Specifically, we extend their structure diffusion module with a sequence diffusion head and fine-tune the entire protein folding model for antibody sequence-structure co-design. Our benchmark results show that sequence-structure co-diffusion models not only surpass state-of-the-art antibody design methods in performance but also maintain structure prediction accuracy comparable to the original folding model. Notably, in the antibody co-design task, our method achieves a CDR-H3 recovery rate of 65% for typical antibodies, outperforming the baselines by 87%, and attains a remarkable 63% recovery rate for nanobodies.