Research map / Probabilistic and scientific modeling / Molecules and proteins
Sparse Fourier Backpropagation in Cryo-EM Reconstruction
Dari Kimanius, Kiarash Jamali, Sjors HW Scheres
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 sparse Fourier-space decoder instead of cryoDRGN-small | cryo-EM VAE · simulated spike | training time for 50 epochs (min): 213 -> 40.3 | Improved |
| decode a full volume in one pass versus cryoDRGN-small | cryo-EM VAE · simulated spike | decoder time per structure: 250 ms -> 5 ms | Improved |
| sparse Fourier method versus cryoDRGN default and large architectures | cryo-EM VAE · simulated spike | reconstruction quality (FSC to ground truth) | Improved |
| sparse Fourier method versus cryoDRGN on conformation separation | cryo-EM VAE · simulated spike | separated conformations | Improved |
| use larger cryoDRGN network (ablation) | cryoDRGN · simulated spike | average FSC per epoch | Hurt |
| increase batch size (ablation) | cryoDRGN · simulated spike | reconstruction quality | Hurt |
| sparse Fourier method versus cryoDRGN-large | cryo-EM VAE · 80S ribosome | training time (min): 1638 -> 210 | Improved |
| sparse Fourier method versus cryoDRGN | cryo-EM VAE · 80S ribosome | decoder time per structure: 480 ms -> 10 ms | Improved |
| sparse Fourier method versus cryoDRGN-large | cryo-EM VAE · 80S ribosome | reconstruction and clustering quality | No clear effect |
| use cryoDRGN-small instead of large (ablation) | cryoDRGN · 80S ribosome | reconstruction quality | Hurt |
Closest papers by meaning
- cryoSPHERE: Single-Particle HEterogeneous REconstruction from cryo EM (ICLR 2025 · similarity 0.90)
- Mixture of neural fields for heterogeneous reconstruction in cryo-EM (NeurIPS 2024 · similarity 0.89)
- Reconstructing Heterogeneous Biomolecules via Hierarchical Gaussian Mixtures and Part Discovery (NeurIPS 2025 · similarity 0.88)
- CryoSPIN: Improving Ab-Initio Cryo-EM Reconstruction with Semi-Amortized Pose Inference (NeurIPS 2024 · similarity 0.88)
- CryoFM: A Flow-based Foundation Model for Cryo-EM Densities (ICLR 2025 · similarity 0.88)
- CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM (NeurIPS 2024 · similarity 0.88)
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.
