Inverse folding for antibody sequence design using deep learning
This work addresses antibody sequence design for drug discovery and binder design, but it is incremental as it builds on previous inverse folding models.
The authors tackled the problem of designing antibody sequences from 3D structural information by proposing a fine-tuned inverse folding model optimized for antibodies, which outperformed generic protein models with improved sequence recovery and structure robustness, particularly on the hypervariable CDR-H3 loop.
We consider the problem of antibody sequence design given 3D structural information. Building on previous work, we propose a fine-tuned inverse folding model that is specifically optimised for antibody structures and outperforms generic protein models on sequence recovery and structure robustness when applied on antibodies, with notable improvement on the hypervariable CDR-H3 loop. We study the canonical conformations of complementarity-determining regions and find improved encoding of these loops into known clusters. Finally, we consider the applications of our model to drug discovery and binder design and evaluate the quality of proposed sequences using physics-based methods.