{"ID":2832338,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.06592","arxiv_id":"2512.06592","title":"On fine-tuning Boltz-2 for protein-protein affinity prediction","abstract":"Accurate prediction of protein-protein binding affinity is vital for understanding molecular interactions and designing therapeutics. We adapt Boltz-2, a state-of-the-art structure-based protein-ligand affinity predictor, for protein-protein affinity regression and evaluate it on two datasets, TCR3d and PPB-affinity. Despite high structural accuracy, Boltz-2-PPI underperforms relative to sequence-based alternatives in both small- and larger-scale data regimes. Combining embeddings from Boltz-2-PPI with sequence-based embeddings yields complementary improvements, particularly for weaker sequence models, suggesting different signals are learned by sequence- and structure-based models. Our results echo known biases associated with training with structural data and suggest that current structure-based representations are not primed for performant affinity prediction.","short_abstract":"Accurate prediction of protein-protein binding affinity is vital for understanding molecular interactions and designing therapeutics. We adapt Boltz-2, a state-of-the-art structure-based protein-ligand affinity predictor, for protein-protein affinity regression and evaluate it on two datasets, TCR3d and PPB-affinity. D...","url_abs":"https://arxiv.org/abs/2512.06592","url_pdf":"https://arxiv.org/pdf/2512.06592v1","authors":"[\"James King\",\"Lewis Cornwall\",\"Andrei Cristian Nica\",\"James Day\",\"Aaron Sim\",\"Neil Dalchau\",\"Lilly Wollman\",\"Joshua Meyers\"]","published":"2025-12-06T23:07:10Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"q-bio.BM\"]","methods":"[\"Generative Adversarial Network\"]","has_code":false}