A robust approach to ab initio cryo-electron microscopy initial volume determination.

Area of Science:

• Structural biology
• Biophysics
• Biochemistry

Background:

• Macromolecular structural information is crucial for understanding biological functions.
• Accurate initial volume determination is a critical, yet challenging, step in cryo-electron microscopy (cryo-EM) reconstruction.
• Current methods for selecting initial volumes are subjective and hinder high-throughput processing.

Purpose of the Study:

• To develop a robust and objective method for selecting the best initial 3D map in cryo-EM.
• To overcome the limitations of subjective and error-prone manual evaluation of candidate initial volumes.
• To facilitate high-throughput data processing in cryo-EM.

Main Methods:

• A novel method evaluating subsets of candidate initial volumes and particle images.
• Utilizing reference-based 3D classification to assess initial map accuracy.
• Identifying majority 3D classes representing accurate initial maps.

Main Results:

• The proposed method objectively identifies the most accurate initial 3D map(s).
• Accurate initial maps lead to particles aligned with high angular accuracy.
• Successful application to both structurally homogeneous and heterogeneous datasets.

Conclusions:

• The developed method provides a robust solution for selecting optimal initial volumes in cryo-EM.
• This approach enhances objectivity and efficiency in cryo-EM data processing.
• The method is applicable to diverse datasets, improving cryo-EM workflow reliability.