Evaluation of automated particle picking for cryogenic electron microscopy using high-precision transmission electron...

Area of Science:

• Structural Biology
• Biophysics
• Microscopy

Background:

• Automated particle picking is crucial for single-particle analysis in cryo-electron microscopy (cryo-EM).
• Challenges include low particle contrast under small defocus and preferred particle orientation, hindering accurate selection.
• Existing methods may struggle with these specific conditions.

Purpose of the Study:

• To develop and evaluate GRIPS, an automated particle-picking software for cryo-EM.
• To address challenges of particle picking under small defocus and preferred orientation.
• To validate the software's performance using a high-precision transmission electron microscope (TEM) image simulator.

Main Methods:

• GRIPS software employs iterative extraction, 2D classification, and positioning.
• Mask processing is introduced to exclude already picked particles.
• High-precision TEM image simulation (elbis) was used for evaluation, generating images with particles and amorphous ice.

Main Results:

• GRIPS successfully picks particles under small defocus conditions where visibility is limited.
• Mask processing effectively handles particles exhibiting preferred orientation.
• The software demonstrated applicability to real experimental cryo-EM data.

Conclusions:

• GRIPS provides a robust solution for automated particle picking in cryo-EM.
• The method enhances particle selection accuracy, particularly in challenging imaging conditions.
• GRIPS is a valuable tool for advancing structural determination using cryo-EM.