Hostname: page-component-586b7cd67f-dlnhk Total loading time: 0 Render date: 2024-11-29T01:43:55.220Z Has data issue: false hasContentIssue false
  • English
  • Français

Automated Filament Finding and Selection from Cryo Electron Micrographs

Published online by Cambridge University Press:  02 July 2020

Y. Zhu ,
B. Carragher  and
C.S. Potter
Y. Zhu
Affiliation:
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, EL, 61801
B. Carragher
Affiliation:
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, EL, 61801
C.S. Potter
Affiliation:
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, EL, 61801
Get access

Abstract

For several years, we have been involved in developing a completely automated system for cryoelectron microscopy, which will allow a 3D electron density map of a macromolecular structure to be generated from a specimen inserted into the microscope with no manual intervention required by an operator. Towards this goal, we have developed a system, called Leginon, to automatically acquire images from the electron microscope. In order to integrate the 3D reconstruction process with the data acquisition step we need then to automatically identify and segment objects from the electron micrographs. This presents a challenge, particularly in the case when micrographs are acquired very close to focus, which results in an image with extremely low contrast. We present here the results of a two-stage automated process for detecting and selecting filamentous structures from images of this kind. Images are acquired in defocus pairs; a near-to-focus (NTF) image is acquired first, followed by a far-from-focus (FFF) image. In the first stage of the process, a threelevel perceptual organization algorithm is used to identify and segment filaments in the FFF image. In the second stage the NTF image is aligned to the FFF image using phase correlation techniques and the filaments are selected at the same coordinates as were identified in the FFF image.

In humans, perceptual organization is the ability to immediately detect structural relationships such as co-linearity, parallelism, and connectivity among image elements. Researchers in human vision perception have long recognized the importance of the common rules by which our visual system attempts to group information. Some of these rules include proximity, similarity, common fate, continuation and closure.

Type
Instrument Automation (Organized by W. Deruijter and C. Potter)
Information
Microscopy and Microanalysis , Volume 7 , Issue S2: Proceedings: Microscopy & Microanalysis 2001, Microscopy Society of America 59th Annual Meeting, Microbeam Analysis Society 35th Annual Meeting, Long Beach, California August 5-9, 2001 , August 2001 , pp. 986 - 987
Copyright
Copyright © Microscopy Society of America 2001

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1.Potter, C.S., et al. (1999) Ultramicroscopy, 77, 153–161.CrossRefGoogle Scholar
2.Carragher, et al. (2000) JSB. 132, 33–45.Google Scholar
3.Kuglin, C. D. and Hines, D. C. (1975) Proc. of Int’l Conf. Cybernetics and Society, 163–165.Google Scholar
4.Sarkar, S. and Boyer, K.L. (1993) Computing Perceptual Organization in Computer Vision, Singapore: World Scientific.Google Scholar
5.Canny, J. (1986) IEEE Trans. Pattern Analysis and Machine Intelligence, PAMI-8, 679–698.CrossRefGoogle Scholar
6. This research is supported by the National Science Foundation (DBI-9730056, DBI-9904547) and the National Institutes of Health (GM61939-01).Google Scholar