Summary

CDR embeddings generated by Protein language models segregate into conformational clusters when mapped using t-SNE. This is NOT true for CDRH3, which instead clusters by loop length. This was observed from AntiBERTy embeddings.

Figures

Ref (1)

See also

1.
Ruffolo JA, Chu L-S, Mahajan SP, Gray JJ. Fast, accurate antibody structure prediction from deep learning on massive set of natural antibodies. Nature Communications. 2023;14(1). Available from: https://doi.org/10.1038/s41467-023-38063-x