Data-driven regularization lowers the size barrier of cryo-EM structure determination

TL;DR

This paper introduces a new method using deep learning to improve cryo-EM imaging of small biological molecules.

Contribution

A novel regularization technique called Blush is introduced, using a pre-trained neural network to enhance cryo-EM reconstructions.

Findings

Blush regularization improves image alignment in cryo-EM for low signal-to-noise data.

The method successfully reconstructs a 40 kDa complex previously considered intractable.

Denoising neural networks expand the range of biological macromolecules suitable for cryo-EM.

Abstract

Macromolecular structure determination by electron cryo-microscopy (cryo-EM) is limited by the alignment of noisy images of individual particles. Because smaller particles have weaker signals, alignment errors impose size limitations on its applicability. Here, we explore how image alignment is improved by the application of deep learning to exploit prior knowledge about biological macromolecular structures that would otherwise be difficult to express mathematically. We train a denoising convolutional neural network on pairs of half-set reconstructions from the electron microscopy data bank (EMDB) and use this denoiser as an alternative to a commonly used smoothness prior. We demonstrate that this approach, which we call Blush regularization, yields better reconstructions than do existing algorithms, in particular for data with low signal-to-noise ratios. The reconstruction of a protein–nucleic acid complex with a molecular weight of 40 kDa, which was previously intractable, illustrates that denoising neural networks will expand the applicability of cryo-EM structure determination for a wide range of biological macromolecules.

Blush regularization makes use of a neural network pre-trained on a diverse set of high-resolution cryo-EM half-maps to improve image alignment, effectively lowering the size barrier, during cryo-EM structure determination.

Figures