# TomoScore: A Neural Network Approach for Quality Assessment of Cellular cryo-ET

## TL;DR

TomoScore is a deep-learning tool that helps assess the quality of cellular cryo-ET images for annotation tasks, reducing the need for expert judgment.

## Contribution

The novel contribution is a deep-learning model that quantifies tomogram quality for subcellular annotation and identifies optimal electron dose ranges.

## Key findings

- TomoScore provides a quantitative measure of tomogram quality for cellular annotation.
- The study identifies an optimal electron dose range for cryo-ET data collection.
- The tool reduces the need for human involvement in tomogram pre-selection.

## Abstract

Electron cryo-tomography (cryo-ET) is a powerful imaging tool that allows three-dimensional visualization of subcellular and molecular architecture without chemical fixation. Tomogram quality varies widely, particularly during large high-throughput data collections, and the most common strategy for initial quality assessment is empirical judgment by an expert. Tomograms may be collected for two distinct purposes: annotation of subcellular features and cellular morphology, typically performed at lower magnifications and higher defocus, and subtomogram averaging, at high magnifications, closer to focus. For the first purpose, contrast and the ability to distinguish cellular features of interest are key, whereas for subtomogram averaging, recoverable signal at high resolution is the key factor. We have developed “TomoScore” a deep-learning based tomogram screening tool targeting cellular annotation. This tool provides a single quantitative measure of the suitability of a tomogram for annotation of subcellular features, in terms of the scale of features that can be readily distinguished. We further explore the relationship between accumulated electron dose and resulting quality, suggesting an optimum dose range for cryo-ET data collection. Overall, our study streamlines data processing and reduces the need for human involvement during pre-selection for tomogram segmentation.