Related Experiment Video
Updated: Mar 19, 2026
09:49
Routine Collection of High-Resolution cryo-EM Datasets Using 200 KV Transmission Electron Microscope
Published on: March 16, 2022
6.1K
Defining the limits and reliability of rigid-body fitting in cryo-EM maps using multi-scale image pyramids
G C P van Zundert1, A M J J Bonvin1
1Bijvoet Center for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Utrecht 3584 CH, The Netherlands.
Journal of Structural Biology
|June 19, 2016
Summary
Cryo-electron microscopy (cryo-EM) structural data can be objectively interpreted using automated cross-correlation fitting. This method accurately places macromolecular subunits and identifies unreliable fits, improving cryo-EM model building.
Area of Science:
- Structural Biology
- Biophysics
- Computational Biology
Background:
- Cryo-electron microscopy (cryo-EM) offers high-resolution structural insights into macromolecular machines.
- Current limitations in cryo-EM resolution hinder de novo model building, often necessitating manual rigid-body fitting of existing models.
- Manual fitting is subjective and prone to over-interpretation of low-resolution density maps.
Purpose of the Study:
- To develop and validate an objective, automated method for fitting macromolecular subunits into cryo-EM density maps.
- To provide statistical indicators for assessing the reliability and ambiguity of rigid-body fits.
- To determine the resolution requirements for accurate subunit fitting and optimize computational efficiency.
Main Methods:
- Cross-correlation-based rigid-body fitting of subunits into experimental ribosome maps (5.5–6.9Å resolution).
- Utilized Fisher z-transformation and confidence intervals to quantify fit reliability and identify ambiguous regions.
- Implemented multi-scale image pyramids to accelerate the fitting process on CPUs and GPUs.
Main Results:
- Cross-correlation fitting successfully placed subunits correctly in over 90% of cases.
- Developed formal statistical indicators to objectively assess fit quality and detect over-interpretation.
- Quantified resolution requirements, showing unambiguous fitting is possible for large subunits even at 20Å resolution.
- Achieved up to 30-fold (CPU) and 40-fold (GPU) speedup using multi-scale image pyramids with negligible loss in accuracy.
Conclusions:
- Automated cross-correlation fitting provides an objective and reliable alternative to manual model building in cryo-EM.
- Statistical measures enhance the interpretability of cryo-EM data by formally assessing fit confidence.
- The PowerFit software, incorporating these methods, facilitates accurate and efficient model building for cryo-EM structures.
