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 ImmuneBuilder. 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.
Cite
@article{arxiv.2405.20863,
title = {ABodyBuilder3: Improved and scalable antibody structure predictions},
author = {Henry Kenlay and Frédéric A. Dreyer and Daniel Cutting and Daniel Nissley and Charlotte M. Deane},
journal= {arXiv preprint arXiv:2405.20863},
year = {2024}
}
Comments
8 pages, 3 figures, 3 tables, code available at https://github.com/Exscientia/ABodyBuilder3, weights and data available at https://zenodo.org/records/11354577