Ibex: Pan-immunoglobulin structure prediction

Frederic A Dreyer, Karolis Martinkus, Jan Ludwiczak, Brennan Abanades, Robert G Alberstein, Pranav Rao, Jae Hyeon Lee, Richard Bonneau, Andrew Martin Watkins, Franziska Seeger

Keywords: structure prediction, protein structure, antibody, drug discovery

TL;DR: Fast state-of-the-art antibody and TCR structure predictions incorporating conformation changes of the binding domain

Abstract: We introduce Ibex, an immunoglobulin protein structure prediction model that achieves state-of-the-art accuracy in modeling the binding domains of antibodies, nanobodies and T-cell receptors. Ibex can model both bound and unbound conformations of the protein, having been trained on labeled apo and holo pairs. Using a private dataset of hundreds of antibody structures, we evaluate the out-of-distribution performance of common structure prediction tools, showing improved robustness of Ibex compared to existing specialized structure prediction models.

Submission Number: 41