Poster
in
Workshop: 2nd Generative AI for Biology Workshop
Ibex: Pan-immunoglobulin structure prediction
Frederic Dreyer · Karolis Martinkus · Jan Ludwiczak · Brennan Abanades · Robert Alberstein · Pranav Rao · Jae Hyeon Lee · Richard Bonneau · Andrew Watkins · Franziska Seeger
Keywords: [ protein structure ] [ antibody ] [ structure prediction ] [ drug discovery ]
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.