Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Dec:12304:82-94.
doi: 10.1007/978-3-030-57821-3_8. Epub 2020 Aug 18.

Dilated-DenseNet For Macromolecule Classification In Cryo-electron Tomography

Shan Gao  1   2   3 , Renmin Han  4 , Xiangrui Zeng  3 , Xuefeng Cui  5 , Zhiyong Liu  1 , Min Xu  3 , Fa Zhang  1
Affiliations

Dilated-DenseNet For Macromolecule Classification In Cryo-electron Tomography

Shan Gao et al. Bioinform Res Appl. 2020 Dec.

Abstract

Cryo-electron tomography (cryo-ET) combined with subtomogram averaging (STA) is a unique technique in revealing macromolecule structures in their near-native state. However, due to the macromolecular structural heterogeneity, low signal-to-noise-ratio (SNR) and anisotropic resolution in the tomogram, macromolecule classification, a critical step of STA, remains a great challenge. In this paper, we propose a novel convolution neural network, named 3D-Dilated-DenseNet, to improve the performance of macromolecule classification in STA. The proposed 3D-Dilated-DenseNet is challenged by the synthetic dataset in the SHREC contest and the experimental dataset, and compared with the SHREC-CNN (the state-of-the-art CNN model in the SHREC contest) and the baseline 3D-DenseNet. The results showed that 3D-Dilated-DenseNet significantly outperformed 3D-DenseNet but 3D-DenseNet is well above SHREC-CNN. Moreover, in order to further demonstrate the validity of dilated convolution in the classification task, we visualized the feature map of 3D-Dilated-DenseNet and 3D-DenseNet. Dilated convolution extracts a much more representative feature map.

Keywords: Convolutional Neural Network; Cryo-electron Tomography; Object Classification; Subtomogram Averaging.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
The architecture of 3D-Dilated-DenseNet. (A) The model framework of 3D-Dilated-DenseNet. (B) The composite function of each layer in dilated dense block. (C) The composite function of transition block in 3D-Dilated-DenseNet.
Fig. 2.
Fig. 2.
A 2D example of dilated convolution layers with 3 x 3 kernel, and the dilated ration is 1, 2, 3.
Fig. 3.
Fig. 3.
The example of synthetic data. (A) The middle slice of one 512 x 512 x 512 tomogram. The right 32 x 32 slices are the consecutive slices of a subtomogram with PDB ID 4d8q. The number of right corner are their slice index. (B) Ground truth corresponding to Fig.(A).
Fig. 4.
Fig. 4.
Dilated-DenseNet performance on synthetic data. (A) Average F1 value on macromolecules according to different size of 3D-DenseNet and 3D-Dilated-DenseNet. (B) The relationship between F1 value and training epoch of 3D-DenseNet and 3D-Dilated-DenseNet.
Fig. 5.
Fig. 5.
Class active mapping image of 3D-DenseNet and 3D-Dilated-DenseNet. Each row represents one macromolecule. And the column images are raw input data, ground truth, CAM image of 3D-DenseNet and CAM image of 3D-Dilated-DenseNet
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Grünewald K, Medalia O, Gross A, Steven AC, Baumeister W: Prospects of electron cryotomography to visualize macromolecular complexes inside cellular compartments: implications of crowding. Biophysical chemistry 100(1-3) (2002) 577–591 - PubMed
    1. Han R, Wan X, Wang Z, Hao Y, Zhang J, Chen Y, Gao X, Liu Z, Ren F, Sun F, et al.: Autom: a novel automatic platform for electron tomography reconstruction. Journal of structural biology 199(3) (2017) 196–208 - PubMed
    1. Han R, Wang L, Liu Z, Sun F, Zhang F: A novel fully automatic scheme for fiducial marker-based alignment in electron tomography. Journal of structural biology 192(3) (2015) 403–417 - PubMed
    1. Wan W, Briggs J: Cryo-electron tomography and subtomogram averaging. In: Methods in enzymology. Volume 579. Elsevier; (2016) 329–367 - PubMed
    1. Ortiz JO, Förster F, Kürner J, Linaroudis AA, Baumeister W: Mapping 70s ribosomes in intact cells by cryoelectron tomography and pattern recognition. Journal of structural biology 156(2) (2006) 334–341 - PubMed

LinkOut - more resources