Research map / Image generation / Autoregressive and latent generation
CryoNet.Refine: A One-step Diffusion Model for Rapid Refinement of Structural Models with Cryo-EM Density Map Restraints
Fuyao Huang, Xiaozhu Yu, Kui Xu, Qiangfeng Cliff Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · Protein complexes (110, cryo-EM) | CC_mask: 0.54 -> 0.59 | Improved |
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · Protein complexes (110, cryo-EM) | Rotamer favored (%): 85.42 -> 98.64 | Improved |
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · Protein complexes (110, cryo-EM) | Rotamer outlier (%): 1.15 -> 0.49 | Improved |
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · Protein complexes (110, cryo-EM) | Ramachandran outlier (%): 0.02 -> 0.06 | Hurt |
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · DNA/RNA-protein complexes (10, cryo-EM) | CC_mc: 0.62 -> 0.61 | Hurt |
| use CryoNet.Refine instead of Phenix.real_space_refine | AlphaFold3 models · DNA/RNA-protein complexes (10, cryo-EM) | CC_sc: 0.58 -> 0.67 | Improved |
| use one-step diffusion instead of 200-step classical diffusion (ablation) | CryoNet.Refine · Classical diffusion vs one-step (27 cases) | CC_mask: 0.30 -> 0.59 | Improved |
| use one-step diffusion instead of 200-step classical diffusion (ablation) | CryoNet.Refine · Classical diffusion vs one-step (27 cases) | Angle RMSD (degree): 1.66 -> 0.36 | Improved |
| add more recycling iterations beyond 100 (ablation) | CryoNet.Refine · PDB 6ksw | CC metrics | No clear effect |
| add density loss and geometry loss together (ablation) | CryoNet.Refine · benchmark | CC and geometric metrics | Improved |
| run CryoNet.Refine instead of Phenix on CPU | runtime · 120 complexes | runtime: 65 cases (54.2%) | Mixed |
Closest papers by meaning
- CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM (NeurIPS 2024 · similarity 0.88)
- cryoSPHERE: Single-Particle HEterogeneous REconstruction from cryo EM (ICLR 2025 · similarity 0.88)
- Multiscale guidance of protein structure prediction with heterogeneous cryo-EM data (NeurIPS 2025 · similarity 0.88)
- CryoFM: A Flow-based Foundation Model for Cryo-EM Densities (ICLR 2025 · similarity 0.86)
- Mixture of neural fields for heterogeneous reconstruction in cryo-EM (NeurIPS 2024 · similarity 0.86)
- CryoSPIN: Improving Ab-Initio Cryo-EM Reconstruction with Semi-Amortized Pose Inference (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
