APPLE picker: Automatic particle picking, a low-effort cryo-EM framework.

Area of Science:

• Structural biology
• Biophysics
• Microscopy

Background:

• Particle picking is a critical initial step in single-particle cryo-electron microscopy (cryo-EM) data processing.
• Manual picking is time-consuming and prone to bias, especially for small, low-contrast particles.
• Current template-based methods can be slow and introduce user bias.

Purpose of the Study:

• To present APPLE (Automatic Particle Picking with Low user Effort), a novel, template-free particle picking approach.
• To develop a fast, accurate, and automated method for particle selection in cryo-EM.
• To reduce manual effort and potential bias in cryo-EM data processing.

Main Methods:

• Developed the APPLE algorithm for automated particle picking.
• Evaluated APPLE on publicly available cryo-EM datasets.
• Tested on various biological macromolecules including β-galactosidase, T20S proteasome, 70S ribosome, and hemocyanin.

Main Results:

• Demonstrated APPLE's capability for fast and accurate particle picking.
• Showcased the template-free nature of the method, avoiding manual bias.
• Validated performance across diverse sample types and particle characteristics.

Conclusions:

• APPLE provides an efficient and reliable solution for particle picking in cryo-EM.
• The method significantly reduces processing time and manual intervention.
• APPLE is a valuable tool for advancing high-resolution cryo-EM structure determination.