Flashcards in Electron Microscopy 6 Deck (15):
Describe briefly the principle of tomography.
Image single molecules in 3D, but tilting and capturing 2D images. Project 2D images together to form 3D image as you know the relative orientations from the tilt angles.
Name two algorithms that can be used improve your model iteratively.
1. Weighted back projection (WBP).
2. Simultaneous iterated reconstruction technique.
Why are gold beads sometimes introduced into the sample?
Helps with working out the orientations of each image and also determining the CTF.
Why is there missing information in tomography?
Different tilt angles results in the loss of certain spatial frequencies, additionally there is a missing cone of information in the z axis.
What is different about the CTF in tomography?
The defocussing (CTF graph) is different depending on the tilt angle chosen, therefore so will the CTF.
What is energy filtering?
The electron beam passes through an electron lens that alters the beam by 90˚ which directs it towards a slit. Any electrons of a different wavelength will be bent differently and not pass through the slit.
Why are energy filters needed?
Thick samples result in more inelastic scattering - filtering out the inelastically scattered electrons gives amplitude contrast.
How is resolution determined in tomography?
Cannot use Fourier shell correlation as cannot image the same cell twice. Instead use known features in the image - e.g. membrane distance is 4.5nm (this is resolution).
Describe sub-tomogram averaging.
This is the use of a tilt series to image thousands of the particle and then averages them based on orientation (3D image).
What are the limitations to resolution in sub-tomogram averaging?
1. Radiation sensitivity - have to collect multiple images from same specimen so have to use a lower electron dosage.
2. Sample movement during tilt series, makes it harder to determine the CTF.
3. Accurately determining the CTF.
What are the general challenges in tomography?
1. Sample too thick
2. Difficulty identifying particles.
3. No dynamics
How can the issue a thick sample be resolved?
1. Cryo-sectioning - flash freeze the sample and use a diamond knife to cut thinner slices.
2. Ion beam milling - use ion beam to mill away at top and bottom layers of the cell.
3. Genetic manipulation - grow mini cells.
What can be done to improve identification of particles?
1. Correlate light microscopy with tomography - use GFP to tag the protein of interest.
2. Use genetically encoded tag on the protein e.g. ferritin which forms a cage that encloses iron ions. Very obvious on the image.
3. Template matching - computationally search for match.
4. Remove proteins from the genome sequentially and see what disappears.
Describe the Zernike phase plate.
This is a metal plate with a hole in the middle. The unscattered electrons travel through the hole. The scattered electrons pass through the metal resulting in a 90˚ phase shift.