Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DCFold: Efficient Protein Structure Generation with Single Forward Pass

About

AlphaFold3 introduces a diffusion-based architecture that elevates protein structure prediction to all-atom resolution with improved accuracy. This state-of-the-art performance has established AlphaFold3 as a foundation model for diverse generation and design tasks. However, its iterative design substantially increases inference time, limiting practical deployment in downstream settings such as virtual screening and protein design. We propose DCFold, a single-step generative model that attains AlphaFold3-level accuracy. Our Dual Consistency training framework, which incorporates a novel Temporal Geodesic Matching (TGM) scheduler, enables DCFold to achieve a 15x acceleration in inference while maintaining predictive fidelity. We validate its effectiveness across both structure prediction and binder design benchmarks.

Zhe Zhang, Yuanning Feng, Yuxuan Song, Keyue Qiu, Hao Zhou, Wei-Ying Ma• 2026

Related benchmarks

TaskDatasetResultRank
Protein-ligand complex foldingPoseBusters v2
Average Inference Time (s)3.76
14
Protein Structure PredictionHomology Recent PDB (PL-complex)
TM-score0.824
3
Protein Structure PredictionHomology Recent PDB (Monomer)
TM-score0.85
3
Protein Structure PredictionHomology Recent PDB (PP-complex)
TM-score0.8
3
Protein Structure PredictionPoseBusters v2
Best % RMSD < 1Å58.1
3
Binder DesignSix targets for binder design (IL-2Rα, TrkA, H3, VirB8, ALK, LTK)
Physics SR (IL-2Rα)37
2
Showing 6 of 6 rows

Other info

Follow for update