Pre-pro is a fast pre-processor for single-particle cryo-EM by enhancing 2D classification.

Area of Science:

• Structural Biology
• Biophysics
• Microscopy Techniques

Background:

• 2D classification is crucial for analyzing single particle cryo-electron microscopy (cryo-EM) images.
• Current methods can be computationally intensive and may not always yield optimal results for heterogeneous datasets.

Purpose of the Study:

• To introduce and evaluate a novel, loss-less pre-processing method to enhance 2D classification in cryo-EM.
• To assess the impact of the proposed pre-processor on classification speed, particle yield, and the quality of initial models for 3D reconstruction.

Main Methods:

• Development of a simple and loss-less pre-processor incorporating a fast dimension-reduction (2SDR) de-noiser.
• Integration of the 2SDR pre-processor with established classification algorithms (RELION, ISAC).
• Comparative performance analysis using multiple cryo-EM experimental datasets, including a challenging nanodisc-embedded TRPV1 dataset.

Main Results:

• The 2SDR pre-processor significantly accelerates 2D classification.
• Improved yield of relevant particles and an increased number of class-average images were observed.
• Application to a heterogeneous TRPV1 dataset improved the final 3D reconstruction resolution to 2.82 Å.

Conclusions:

• The 2SDR pre-processor offers a low-cost, widely applicable solution for enhancing 2D classification in cryo-EM.
• The method facilitates the generation of better initial models for 3D reconstruction, leading to improved resolutions.
• Future generalization to neural network de-noisers is envisioned for broader applicability.