On fine-tuning Boltz-2 for protein-protein affinity prediction
By: James King , Lewis Cornwall , Andrei Cristian Nica and more
Potential Business Impact:
Helps predict how proteins stick together.
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.
Similar Papers
Accurate and generalizable protein-ligand binding affinity prediction with geometric deep learning
Computational Engineering, Finance, and Science
Finds new medicines by predicting how they stick to bodies.
Learning Protein-Ligand Binding in Hyperbolic Space
Machine Learning (CS)
Finds better medicines faster using curved math.
Learning to Align Molecules and Proteins: A Geometry-Aware Approach to Binding Affinity
Machine Learning (CS)
Finds new medicines faster by predicting how they'll work.