Covariance estimation using conjugate gradient for 3D classification in Cryo-EM

TL;DR

This paper introduces an improved covariance estimation method for 3D classification in Cryo-EM that accounts for contrast transfer function and viewing angle distribution, enhancing analysis of biological macromolecules.

Contribution

It extends previous covariance estimation techniques by incorporating real-world imaging factors, improving accuracy in Cryo-EM 3D structure variability analysis.

Findings

Effective on synthetic and experimental datasets

Improves covariance estimation accuracy

Handles contrast transfer function and non-uniform angles

Abstract

Classifying structural variability in noisy projections of biological macromolecules is a central problem in Cryo-EM. In this work, we build on a previous method for estimating the covariance matrix of the three-dimensional structure present in the molecules being imaged. Our proposed method allows for incorporation of contrast transfer function and non-uniform distribution of viewing angles, making it more suitable for real-world data. We evaluate its performance on a synthetic dataset and an experimental dataset obtained by imaging a 70S ribosome complex.