期刊
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
卷 221, 期 -, 页码 -出版社
ELSEVIER IRELAND LTD
DOI: 10.1016/j.cmpb.2022.106871
关键词
Cryo-ET; Sub-tomogram averaging; Particle localization; Particle classification; Convolutional neural networks
类别
资金
- National Key Research and Development Program of China [2021YFF0704300, 2017YFA0504702]
- Strategic Priority Research Program of the Chinese Academy of Sciences [XDA16021400]
- NSFC [61932018, 62072441, 62072280, 62072283]
This study presents a precise macromolecule localization and classification method called VP-Detector, which utilizes a 3D multiscale dense convolutional neural network and weighted focal loss to achieve high accuracy particle detection and classification under challenging conditions such as low signal-to-noise ratio, missing wedge artifacts, and diverse macromolecule shapes and sizes. Experimental results demonstrate that VP-Detector outperforms other state-of-the-art methods, and it can replace manual particle picking in real-world tomograms.
Background and objective: Cryo-electron tomography (cryo-ET) with subtomogram averaging (STA) is indispensable when studying macromolecule structures and functions in their native environments. Due to the low signal-to-noise ratio, the missing wedge artifacts in tomographic reconstructions, and multiple macromolecules of varied shapes and sizes, macromolecule localization and classification remain challenging. To tackle this bottleneck problem for structural determination by STA, we design an accurate macromolecule localization and classification method named voxelwise particle detector (VP-Detector). Methods: VP-Detector is a two-stage particle detection method based on a 3D multiscale dense convolutional neural network (3D MSDNet). The proposed network uses 3D hybrid dilated convolution (3D HDC) to avoid the resolution loss caused by scaling operations. Meanwhile, it uses 3D dense connectivity to encourage the reuse of feature maps to reduce trainable parameters. In addition, the weighted focal loss is proposed to focus more attention on difficult samples and rare classes, which relieves the class imbalance caused by multiple particles of various sizes. The performance of VP-Detector is evaluated on both simulated and real-world tomograms, and it shows that VP-Detector outperforms state-of-the-art methods. Results: The experiments show that VP-Detector outperforms the state-of-the-art methods on particle localization with an F1-score of 0.951 and a precision of 0.978. In addition, VP-Detector can replace manual particle picking in experiment on the real-world tomograms. Furthermore, it performs well in classifying large-, medium-, and small-weight proteins with accuracies of 1, 0.95, and 0.82, respectively. Finally, ablation studies demonstrate the effectiveness of 3D HDC, 3D dense connectivity, weighted focal loss, and training on small training sets. Conclusions: VP-Detector can achieve high accuracy in particle detection with few trainable parameters and support training on small datasets. It can also relieve the class imbalance caused by multiple particles with various shapes and sizes. (C) 2022 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据