Fergus Boyles

ANARCII is a language model that numbers antigen receptor sequences without alignment, achieving high accuracy and efficiency for diverse sequences.

This study evaluates machine learning-based scoring functions by comparing them to baseline models that only learn dataset biases, revealing that the MLBSFs primarily capture biases rather than generalizable physics knowledge.

PLAbDab-nano is a database of 5000 annotated VHH and VNAR nanobody sequences from camelids and sharks, providing a resource for designing libraries and finding similar sequences.

ImmuneBuilder predicts immune protein structures with high accuracy and speed, outperforming AlphaFold2 on specific benchmarks.

The paper introduces Protein-Ligand Interaction Graphs (PLIGs) to improve binding affinity prediction by encoding 3D protein-ligand interactions into GNN node features, achieving high performance on the CASF-2016 benchmark.