ABodyBuilder3: Improved and scalable antibody structure predictions

Henry Kenlay, Frederic A Dreyer, Daniel Cutting, Daniel Allen Nissley, Charlotte Deane

Keywords: antibody, structure prediction, machine learning

TL;DR: Improved antibody structure predictions leveraging language model embeddings, with better uncertainty estimates

Abstract:

Accurate prediction of antibody structure is a central task in the design and development of monoclonal antibodies, notably to understand both their developability and their binding properties.
In this article, we introduce ABodyBuilder3, an improved and scalable antibody structure prediction model based on ABodyBuilder2.
We achieve a new state-of-the-art accuracy in the modelling of CDR loops by leveraging language model embeddings, and show how predicted structures can be further improved through careful relaxation strategies.
Finally, we incorporate a predicted Local Distance Difference Test into the model output to allow for a more accurate estimation of uncertainties.

Submission Number: 2