Neural Representations of Cryo-EM Maps and a Graph-Based Interpretation

TL;DR

This paper introduces a neural data format for cryo-EM maps that provides more accurate, continuous representations, enabling detailed graph-based interpretations of protein structures from high-resolution cryo-EM data.

Contribution

The paper presents a novel neural network-based data format for cryo-EM maps that improves interpolation accuracy and facilitates graph-based structural analysis.

Findings

Neural cryo-EM maps achieve less than 0.01 MAE in density interpolation.

High-resolution maps (>1.6 Å) yield over 99% residue coverage.

Graph interpretations from atomic resolution maps have 85% atomic coverage with 0.19 Å RMSD.

Abstract

Advances in imagery at atomic and near-atomic resolution, such as cryogenic electron microscopy (cryo-EM), have led to an influx of high resolution images of proteins and other macromolecular structures to data banks worldwide. Producing a protein structure from the discrete voxel grid data of cryo-EM maps involves interpolation into the continuous spatial domain. We present a novel data format called the neural cryo-EM map, which is formed from a set of neural networks that accurately parameterize cryo-EM maps and provide native, spatially continuous data for density and gradient. As a case study of this data format, we create graph-based interpretations of high resolution experimental cryo-EM maps. Normalized cryo-EM map values interpolated using the non-linear neural cryo-EM format are more accurate, consistently scoring less than 0.01 mean absolute error, than a conventional tri-linear interpolation, which scores up to 0.12 mean absolute error. Our graph-based interpretations of 115 experimental cryo-EM maps from 1.15 to 4.0 Angstrom resolution provide high coverage of the underlying amino acid residue locations, while accuracy of nodes is correlated with resolution. The nodes of graphs created from atomic resolution maps (higher than 1.6 Angstroms) provide greater than 99% residue coverage as well as 85% full atomic coverage with a mean of than 0.19 Angstrom root mean squared deviation (RMSD). Other graphs have a mean 84% residue coverage with less specificity of the nodes due to experimental noise and differences of density context at lower resolutions. This work may be generalized for transforming any 3D grid-based data format into non-linear, continuous, and differentiable format for the downstream geometric deep learning applications.