Atomic-accuracy models from 4.5-Å cryo-electron microscopy data with density-guided iterative local refinement.

Area of Science:

• Structural Biology
• Biophysics
• Computational Biology

Background:

• Cryo-electron microscopy (cryo-EM) enables near-atomic resolution structural determination.
• Accurate protein structure models are crucial for understanding biological function.
• Refining models based on cryo-EM maps remains a challenge.

Purpose of the Study:

• To develop a general and robust method for refining protein structure models using cryo-electron microscopy maps.
• To achieve atomic-level accuracy in refined models, independent of initial model quality.

Main Methods:

• Integration of Monte Carlo sampling with local density-guided optimization.
• Application of Rosetta all-atom refinement protocols.
• Real-space B-factor fitting for model optimization.

Main Results:

• The method successfully refined protein structure models using experimental cryo-EM maps (≥4.5 Å resolution).
• Atomic-level accuracy was consistently achieved across three diverse test systems.
• Performance surpassed the molecular dynamics flexible fitting (MDFF) method.
• Cross-validated model quality statistics showed strong correlation with actual model accuracy.

Conclusions:

• The developed approach provides a reliable strategy for enhancing protein structure models derived from cryo-EM data.
• This method offers a significant advancement in achieving high-accuracy structural models for biological research.