DeepEMhancer: a deep learning solution for cryo-EM volume post-processing.
DeepEMhancer: a deep learning solution for cryo-EM volume post-processing.
复制标题
DOI：
10.1038/s42003-021-02399-1
复制
发表时间：
2021-07-15
影响因子：
5.9
通讯作者：
Vargas J
中科院分区：
文献类型：
--
作者：
Sanchez-Garcia R;Gomez-Blanco J;Cuervo A;Carazo JM;Sorzano COS;Vargas J
低温电镜图是蛋白质结构建模的宝贵信息来源。然而，由于在高频率下对比度的损失，它们通常需要后处理以提高其可解释性。基于全局b因子校正的大多数流行方法都存在局限性。例如，他们忽略了重建往往表现出的地图局部质量的异质性。为了克服这些问题，我们提出了DeepEMhancer，这是一种深度学习方法，旨在对冷冻电镜图进行自动后处理。DeepEMhancer在实验地图和使用各自原子模型进行锐化的地图的成对数据集上进行训练，学会了如何在一个步骤中对实验地图进行后期处理，执行类似遮罩和锐化的操作。DeepEMhancer在20张不同的实验地图上进行了测试，显示了它能够降低噪音水平，并获得更详细的实验地图版本。此外，我们还说明了DeepEMhancer对SARS-CoV-2 RNA聚合酶结构的益处。Sanchez-Garcia等人提出了DeepEMhancer，这是一种基于深度学习的方法，可以自动对原始冷冻电子显微镜密度图进行后处理。作者报告说，DeepEMhancer在全球范围内提高了密度图的局部质量，并且可能代表一个有用的工具，用于不容易获得PDB模型的新结构。
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. Sanchez-Garcia et al. present DeepEMhancer, a deep learning-based method that can automatically perform post-processing of raw cryo-electron microscopy density maps. The authors report that DeepEMhancer globally improves local quality of density maps, and may represent a useful tool for novel structures where PDB models are not readily available.
登录
查看更多内容
冷冻电子显微镜中单粒子识别的深度卷积神经网络方法
影响因子：
3
作者：
Zhu Y;Ouyang Q;Mao Y
通讯作者：
Mao Y
DOI：
10.1107/s2059798318004655
发表时间：
2018-06-01
期刊：
Acta crystallographica. Section D, Structural biology
影响因子：
--
作者：
Terwilliger TC;Sobolev OV;Afonine PV;Adams PD
通讯作者：
Adams PD
DOI：
10.1107/s0907444904019158
发表时间：
2004-12-01
影响因子：
2.2
作者：
Emsley, P;Cowtan, K
通讯作者：
Cowtan, K
影响因子：
19.5
作者：
Wu, Yuxin;He, Kaiming
通讯作者：
He, Kaiming
影响因子：
3.9
作者：
Ramirez-Aportela, Erney;Mota, Javier;Sorzano, Carlos Oscar S.
通讯作者：
Sorzano, Carlos Oscar S.