Cryo-EM model validation recommendations based on outcomes of the 2019 EMDataResource challenge

Area of Science:

• Structural Biology
• Biophysics
• Computational Biology

Background:

• Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the 3D structures of biological macromolecules.
• Accurate model building from cryo-EM maps is crucial for understanding molecular function.
• Evaluating the quality and reproducibility of these models is essential for reliable structural data.

Purpose of the Study:

• To assess the quality of models generated from cryo-EM maps using current software.
• To evaluate the reproducibility of modeling results across different software and users.
• To compare the performance of existing model evaluation metrics, especially Fit-to-Map metrics, at near-atomic resolution.

Main Methods:

• 13 teams participated in the 2019 Cryo-EM Model Challenge.
• Four benchmark cryo-EM maps were used, including a series from 1.8 to 3.1 Å resolution.
• Models were generated using various software, and their quality was assessed using multiple metrics.

Main Results:

• Cryo-EM models demonstrated high accuracy and reproducibility across participating teams.
• Current modeling software is capable of producing reliable atomic models from near-atomic resolution cryo-EM maps.
• Fit-to-Map metrics showed varying performance, highlighting the need for comprehensive evaluation.

Conclusions:

• Recommendations are provided for validating near-atomic cryo-EM structures in experiments and databases like the Protein Data Bank.
• Adoption of multiple scoring parameters is advised for objective annotation and assessment of cryo-EM models.
• The study confirms the feasibility of high-quality model building in near-atomic resolution cryo-EM.