Brennan Abanades

The paper benchmarks structure prediction models on a new antibody dataset to assess their ability to distinguish binders from decoys, finding moderate success but limitations in epitope localization.

MIMOSA improves de novo antibody design by constraining generative models to known binder interfaces, achieving validated binders on targets where unconstrained methods fail.

The paper evaluates TCR structure prediction models using a large dataset, finds comparable structural diversity in alpha and beta chains, and provides over 1.5M predicted structures for further analysis.

This study evaluates deep learning's ability to predict antibody CDR loop structures, finding limited extrapolation beyond training data but recovery with minimal structural examples.

PLAbDab is a database of over 150,000 antibody sequences and structures with literature annotations, enabling searches by sequence, structure, or keyword for functional insights.

SPACE2 improves epitope clustering by using structural models, showing higher coverage and diversity in antibodies binding the same epitope compared to sequence-based methods.

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

KA-Search rapidly and exhaustively searches for similar antibody sequences across defined regions in large antibody repertoires to predict properties like immunogenicity and antigen specificity.

ABlooper predicts antibody CDR loop structures with high accuracy and confidence estimates, achieving an average RMSD of 2.49 Å on the Rosetta Antibody Benchmark.