Deep learning methods produce different CDRH3 conformers

Summary

Deep learning methods diverge on predictions of CDRH3 structure. Usually one is correct, but it changes depending on which model is used, i.e., they sample diverse conformations (1). An example is available in (2).

Figures

Ref (2)

See also

• Antibody developability predictions from computational models are sensitive to the structure prediction method

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

2.

Fernández-Quintero ML, Kokot J, Waibl F, Fischer A-LM, Quoika PK, Deane CM, et al. Challenges in antibody structure prediction. mAbs. 2023;15(1). Available from: https://doi.org/10.1080/19420862.2023.2175319