Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs

TL;DR

This paper introduces a deep learning method using variational auto-encoders to infer a continuous distribution of protein atomic structures directly from cryo-EM images, addressing conformational heterogeneity.

Contribution

It presents a novel VAE-based approach for modeling continuous atomic conformations from cryo-EM data, improving understanding of flexible proteins.

Findings

Effective recovery of continuous atomic structures demonstrated on simulated data.

Method captures conformational heterogeneity better than volume-based models.

Potential to incorporate prior structural information for enhanced modeling.

Abstract

Cryo-electron microscopy (cryo-EM) has revolutionized experimental protein structure determination. Despite advances in high resolution reconstruction, a majority of cryo-EM experiments provide either a single state of the studied macromolecule, or a relatively small number of its conformations. This reduces the effectiveness of the technique for proteins with flexible regions, which are known to play a key role in protein function. Recent methods for capturing conformational heterogeneity in cryo-EM data model it in volume space, making recovery of continuous atomic structures challenging. Here we present a fully deep-learning-based approach using variational auto-encoders (VAEs) to recover a continuous distribution of atomic protein structures and poses directly from picked particle images and demonstrate its efficacy on realistic simulated data. We hope that methods built on this work will allow incorporation of stronger prior information about protein structure and enable better understanding of non-rigid protein structures.