Boltz2 Co-folding
Co-foldingDeep learning protein-ligand co-folding.
Uses Boltz2, a deep learning model for joint protein-ligand structure prediction (co-folding). Confidence scores from the model's predicted structures are used as binding affinity proxies.
Design Rationale
Co-folding models represent the frontier of ML-based binding prediction. Their performance on this benchmark reveals whether they capture SAR-level resolution.
Evaluation results for this model are not yet available.
Check back after running the benchmark pipeline.