Chapter 11
Signal Optimization in Electron
Tomography
Mauro Maiorca and Peter B. Rosenthal
Abstract Signal optimization is essential for the reliable recognition and interpretation of structural features in volumes obtained by electron tomography.
Optimization may be achieved through image processing algorithms that minimize
noise and artefacts in either raw or reconstructed image signal. Procedures for signal
optimization may be performed in real or Fourier space. We survey procedures and
applications with an emphasis on cryotomography of frozen-hydrated biological
specimens.
11.1 Introduction
Three-dimensional structures of biological specimens from the molecular to cellular
scale may be obtained through volumetric reconstruction from projection images
recorded in the electron microscope. Frozen-hydrated, unstained specimen preparations preserve high-resolution structural features but are sensitive to radiation
damage by the electron beam, requiring minimal electron exposure (low dose)
methods during image acquisition. The projection images therefore have low
contrast and a low signal-to-noise ratio (SNR).
SNR may be increased by image averaging, as exploited in single particle
reconstruction, which requires the identification of randomly oriented particles
followed by coherent averaging. In recent years, atomic structures have been
determined from single particles that are of comparable resolution to those obtained
by X-ray crystallography. However, many specimens are pleomorphic and no two
identical examples of the specimen exist, so that image averaging is not possible. For
three-dimensional reconstruction from projection images, many views must be
recorded of a single, unique example of the specimen, usually by tilting the specimen
M. Maiorca (&) Á P. B. Rosenthal (&)
Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK
e-mail: mauromaiorca@gmail.com
P. B. Rosenthal
e-mail: Peter.Rosenthal@crick.ac.uk
© Springer International Publishing AG 2018
E. Hanssen (ed.), Cellular Imaging, Biological and Medical Physics,
Biomedical Engineering, https://doi.org/10.1007/978-3-319-68997-5_11
283
Signal Optimization in Electron
Tomography
Mauro Maiorca and Peter B. Rosenthal
Abstract Signal optimization is essential for the reliable recognition and interpretation of structural features in volumes obtained by electron tomography.
Optimization may be achieved through image processing algorithms that minimize
noise and artefacts in either raw or reconstructed image signal. Procedures for signal
optimization may be performed in real or Fourier space. We survey procedures and
applications with an emphasis on cryotomography of frozen-hydrated biological
specimens.
11.1 Introduction
Three-dimensional structures of biological specimens from the molecular to cellular
scale may be obtained through volumetric reconstruction from projection images
recorded in the electron microscope. Frozen-hydrated, unstained specimen preparations preserve high-resolution structural features but are sensitive to radiation
damage by the electron beam, requiring minimal electron exposure (low dose)
methods during image acquisition. The projection images therefore have low
contrast and a low signal-to-noise ratio (SNR).
SNR may be increased by image averaging, as exploited in single particle
reconstruction, which requires the identification of randomly oriented particles
followed by coherent averaging. In recent years, atomic structures have been
determined from single particles that are of comparable resolution to those obtained
by X-ray crystallography. However, many specimens are pleomorphic and no two
identical examples of the specimen exist, so that image averaging is not possible. For
three-dimensional reconstruction from projection images, many views must be
recorded of a single, unique example of the specimen, usually by tilting the specimen
M. Maiorca (&) Á P. B. Rosenthal (&)
Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK
e-mail: mauromaiorca@gmail.com
P. B. Rosenthal
e-mail: Peter.Rosenthal@crick.ac.uk
© Springer International Publishing AG 2018
E. Hanssen (ed.), Cellular Imaging, Biological and Medical Physics,
Biomedical Engineering, https://doi.org/10.1007/978-3-319-68997-5_11
283