DeepEMhancer: a deep learning solution for cryo-EM volume post-processing

DeepEMhancer: решение на основе глубокого обучения для постобработки томов криоэлектронной микроскопии
Carlos Óscar S. Sorzano, Javier Vargas, J.M. Carazo, Rubén Sánchez-García, Josué Gómez-Blanco, Ana Cuervo
2021-07-15

DeepEMhancerSARS-CoV-2 RNA polymerasecryo-EM map post-processingdeep learningmap sharpening
Cryo-EM maps are valuable sources of information for protein structure modeling. However, due to the loss of contrast at high frequencies, they generally need to be post-processed to improve their interpretability. Most popular approaches, based on global B-factor correction, suffer from limitations. For instance, they ignore the heterogeneity in the map local quality that reconstructions tend to exhibit. Aiming to overcome these problems, we present DeepEMhancer, a deep learning approach designed to perform automatic post-processing of cryo-EM maps. Trained on a dataset of pairs of experimental maps and maps sharpened using their respective atomic models, DeepEMhancer has learned how to post-process experimental maps performing masking-like and sharpening-like operations in a single step. DeepEMhancer was evaluated on a testing set of 20 different experimental maps, showing its ability to reduce noise levels and obtain more detailed versions of the experimental maps. Additionally, we illustrated the benefits of DeepEMhancer on the structure of the SARS-CoV-2 RNA polymerase.
1
DeepEMhancer is an automated deep-learning method for post-processing cryo-EM maps, addressing limitations of global B-factor correction.
2
Evaluation on 20 experimental maps showed that DeepEMhancer reduces noise and produces more detailed map representations.
3
The method’s benefits were additionally demonstrated on the cryo-EM structure of the SARS-CoV-2 RNA polymerase.
4
Trained on experimental maps paired with atomic-model-sharpened maps, DeepEMhancer performs masking-like and sharpening-like operations simultaneously.

experimental cryo-EM maps of protein structures

automatic post-processing of cryo-EM maps to reduce noise and enhance structural detail while accounting for local map-quality heterogeneity

Publication Details
Publication Date
2021-07-15
Journal
Publisher
ISSN
Cited by
1349
Access Type
Author Information
Authors
Carlos Óscar S. Sorzano
Javier Vargas
J.M. Carazo
Rubén Sánchez-García
Josué Gómez-Blanco
Ana Cuervo
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%