Generative Adversarial Networks for Robust Cryo-EM Image Denoising

Hanlin Gu, Poline XIAN, Ilona Christy Unarta, Yuan Yao*

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingChapterpeer-review

2 Citations (Scopus)

Abstract

The cryo-electron microscopy (cryo-EM) becomes popular for macromolecular structure determination. However, the 2D images captured by cryo-EM are of high noise and often mixed with multiple heterogeneous conformations and contamination, imposing a challenge for denoising. Traditional image denoising methods and simple denoising autoencoder cannot work well when the signal-to-noise ratio (SNR) of images is meager and contamination distribution is complex. Thus it is desired to develop new effective denoising techniques to facilitate further research such as 3D reconstruction, 2D conformation classification, and so on. In this chapter, we approach the robust denoising problem for cryo-EM images by introducing a family of generative adversarial networks (GANs), called β-GAN, which is able to achieve robust estimation of certain distributional parameters under Huber contamination model with statistical optimality. To address the denoising challenges, for example, the traditional image generative model might be contaminated by a small portion of unknown outliers, β-GANs are exploited to enhance the robustness of denoising autoencoder. Our proposed method is evaluated by both a simulated dataset on the Thermus aquaticus RNA polymerase (RNAP) and a real-world dataset on the Plasmodium falciparum 80S ribosome dataset (EMPIAR-10028), in terms of mean square error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and 3D reconstruction as well. Quantitative comparisons show that equipped with some designs of β-GANs and the robust ℓ1-autoencoder, one can stabilize the training of GANs and achieve the state-of-the-art performance of robust denoising with low SNR data and against possible information contamination. Our proposed methodology thus provides an effective tool for robust denoising of cryo-EM 2D images and helpful for 3D structure reconstruction.
Original languageEnglish
Title of host publicationHandbook of Mathematical Models and Algorithms in Computer Vision and Imaging
Subtitle of host publicationMathematical Imaging and Vision
EditorsKe Chen, Carola-Bibiane Schönlieb, Xue-Cheng Tai, Laurent Younes
Place of PublicationCham
PublisherSpringer Cham
Chapter26
Pages969–1000
Number of pages32
ISBN (Electronic)973030986612
ISBN (Print)9783030986605
DOIs
Publication statusPublished - 25 Feb 2023

Scopus Subject Areas

  • General Mathematics

User-Defined Keywords

  • Autoencoder
  • Cryo-electron microscopy
  • Denoising
  • Generative adversarial networks
  • Robust statistics

Fingerprint

Dive into the research topics of 'Generative Adversarial Networks for Robust Cryo-EM Image Denoising'. Together they form a unique fingerprint.

Cite this