WikiSysop: /* Abstract */

2021-01-13T14:01:02Z

Abstract

← Older revision		Revision as of 14:01, 13 January 2021
Line 5:		Line 5:
	== Abstract ==		== Abstract ==

	Single-particle cryo-electron microscopy (cryo-EM) is an emerging imaging modality~~<br>~~capable of visualizing proteins and macro-molecular complexes at near-atomic~~<br>~~resolution. The low electron-doses used to prevent sample radiation damage, result~~<br>~~in images where the power of the noise is 100 times greater than the power of the~~<br>~~signal. To overcome the low-SNRs, hundreds of thousands of particle projections~~<br>~~acquired over several days of data collection are averaged in 3D to determine the~~<br>~~structure of interest. Meanwhile, recent image super-resolution (SR) techniques~~<br>~~based on neural networks have shown state of the art performance on natural images.~~<br>~~Building on these advances, we present a multiple-image SR algorithm based~~<br>~~on deep internal learning designed specifically to work under low-SNR conditions.~~<br>~~Our approach leverages the internal image statistics of cryo-EM movies and does~~<br>~~not require training on ground-truth data. When applied to a single-particle dataset~~<br>~~of apoferritin, we show that the resolution of 3D structures obtained from SR~~<br>~~micrographs can surpass the limits imposed by the imaging system. Our results~~<br>~~indicate that the combination of low magnification imaging with image SR has the~~<br>~~potential to accelerate cryo-EM data collection without sacrificing resolution.		Single-particle cryo-electron microscopy (cryo-EM) is an emerging imaging modality capable of visualizing proteins and macro-molecular complexes at near-atomic resolution. The low electron-doses used to prevent sample radiation damage, result in images where the power of the noise is 100 times greater than the power of the signal. To overcome the low-SNRs, hundreds of thousands of particle projections acquired over several days of data collection are averaged in 3D to determine the structure of interest. Meanwhile, recent image super-resolution (SR) techniques based on neural networks have shown state of the art performance on natural images. Building on these advances, we present a multiple-image SR algorithm based on deep internal learning designed specifically to work under low-SNR conditions. Our approach leverages the internal image statistics of cryo-EM movies and does not require training on ground-truth data. When applied to a single-particle dataset of apoferritin, we show that the resolution of 3D structures obtained from SR micrographs can surpass the limits imposed by the imaging system. Our results indicate that the combination of low magnification imaging with image SR has the potential to accelerate cryo-EM data collection without sacrificing resolution.

	== Keywords ==		== Keywords ==

WikiSysop: Created page with "== Citation == Huang, Q.; Zhou, Y.; Du, X.; Chen, R.; Wang, J.; Rudin, C.; Bartesaghi, A. Cryo-ZSSR: multiple-image super-resolution basedon deep internal learning. Proc. 34t..."

2021-01-13T13:58:44Z

Created page with "== Citation == Huang, Q.; Zhou, Y.; Du, X.; Chen, R.; Wang, J.; Rudin, C.; Bartesaghi, A. Cryo-ZSSR: multiple-image super-resolution basedon deep internal learning. Proc. 34t..."

New page

== Citation ==

Huang, Q.; Zhou, Y.; Du, X.; Chen, R.; Wang, J.; Rudin, C.; Bartesaghi, A. Cryo-ZSSR: multiple-image super-resolution basedon deep internal learning. Proc. 34th Conf. on Neural Information Processing Systems, 2020

== Abstract ==

Single-particle cryo-electron microscopy (cryo-EM) is an emerging imaging modality capable of visualizing proteins and macro-molecular complexes at near-atomic resolution. The low electron-doses used to prevent sample radiation damage, result in images where the power of the noise is 100 times greater than the power of the signal. To overcome the low-SNRs, hundreds of thousands of particle projections acquired over several days of data collection are averaged in 3D to determine the structure of interest. Meanwhile, recent image super-resolution (SR) techniques based on neural networks have shown state of the art performance on natural images. Building on these advances, we present a multiple-image SR algorithm based on deep internal learning designed specifically to work under low-SNR conditions. Our approach leverages the internal image statistics of cryo-EM movies and does not require training on ground-truth data. When applied to a single-particle dataset of apoferritin, we show that the resolution of 3D structures obtained from SR micrographs can surpass the limits imposed by the imaging system. Our results indicate that the combination of low magnification imaging with image SR has the potential to accelerate cryo-EM data collection without sacrificing resolution.

== Keywords ==

== Links ==

https://www.mlsb.io/papers/MLSB2020_Cryo-ZSSR:_multiple-image_super-resolution_based.pdf

== Related software ==

== Related methods ==

== Comments ==

2020Huang SuperResolution - Revision history

WikiSysop: /* Abstract */

WikiSysop: Created page with "== Citation == Huang, Q.; Zhou, Y.; Du, X.; Chen, R.; Wang, J.; Rudin, C.; Bartesaghi, A. Cryo-ZSSR: multiple-image super-resolution basedon deep internal learning. Proc. 34t..."