Differences between revisions 6 and 29 (spanning 23 versions)
Revision 6 as of 2009-03-02 17:58:29
Size: 683
Comment:
Revision 29 as of 2010-06-17 19:10:09
Size: 2512
Editor: SteveLudtke
Comment:
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
attachment:ctf = e2ctf =
||<<TableOfContents>> || {{attachment:ctf.png}} ||
Line 3: Line 4:
= e2ctf = == CTF Correction in EMAN2 ==
CTF correction in EMAN2 is performed on boxed out particles. The entire framework is incompatible with the idea of
doing per-micrograph CTF correction. There is a [[EMAN2/Theory/CTF|theory section]] discussing how this works in
detail. In general, you box out particles with a fairly liberal box size (about 1.5 - 2x the smallest possible box that would
fit your particle). The central part of each box (the particle) contains signal + noise, the edge of the box
contains just background noise. Separating these two regions permits an accurate estimate of the spectral signal to noise
ratio of the particles in each image, and also permits accurate assessment of the defocus and other CTF paramters. This
SSNR is an integral part of later CTF correction procedures, and can ONLY be computed from particles, not entire micrographs.
While the core CTF infrastructure could in principle be used to simply phase-flip a whole micrograph, none of the existing
user-level programs support this operation.
Line 5: Line 15:
To get familiar with how e2ctf.py works, try separating the way you using e2ctf.py into three stages: == Using e2ctf.py ==
Normally this program would be used through the e2workflow.py interface. You can call it manually, from the command-line
as well, of course.
Line 7: Line 19:
== Auto fitting == e2ctf.py is generally used in 3 stages
Line 9: Line 21:
=== Auto fitting ===
Line 10: Line 23:
e2ctf.py  1.img --voltage=200 --apix=2 --cs=3.2 --auto_fit e2ctf.py 1.img --voltage=200 --apix=2 --cs=3.2 --autofit
Line 12: Line 25:
== Fine tuning == This runs auto CTF parameter determination, storing the result in a local database.
Line 14: Line 27:
(save any changes you make with the Save Parms button) === Manual (GUI) parameter check ===
Line 16: Line 29:
e2ctf.py  1.img --gui e2ctf.py 1.img --gui
Line 18: Line 31:
== Write output == Now that you have determined the parameters with the first command, have a look at the results in the interface. Tweak if necessary. Save any changes you make with the Save Parms button
Line 20: Line 33:
=== Write output ===
Line 23: Line 37:
The output will be in the particles directory. Also, you can supply
more than one image as input to e2ctf.py.
Finally generate output. The output will be in the particles directory.
Line 26: Line 39:
You can combine more than command would have worked if you had used a command like this: == Processing multiple images simultaneously ==
e2ctf takes multiple images as input, so you can use commands like this:
Line 28: Line 42:
e2ctf.py 1.img --voltage=200 --apix=2 --cs=3.2 --auto_fit --gui {{{
e2ctf.py *.hdf --voltage=200 --apix=2 --cs=3.2 --autofit
e2ctf.py *.hdf --gui
e2ctf.py *.hdf --phaseflip --wiener
}}}
== Combining stages into a single command ==
You can combine more than one of the e2ctf processing stages into a single command. For instance you can run automated fitting and open the GUI with a single command like this:

{{{
e2ctf.py 1.img --voltage=200 --apix=2 --cs=3.2 --autofit --gui
e2ctf.py *.img --voltage=200 --apix=2 --cs=3.2 --autofit --gui
}}}

e2ctf

ctf.png

CTF Correction in EMAN2

CTF correction in EMAN2 is performed on boxed out particles. The entire framework is incompatible with the idea of doing per-micrograph CTF correction. There is a theory section discussing how this works in detail. In general, you box out particles with a fairly liberal box size (about 1.5 - 2x the smallest possible box that would fit your particle). The central part of each box (the particle) contains signal + noise, the edge of the box contains just background noise. Separating these two regions permits an accurate estimate of the spectral signal to noise ratio of the particles in each image, and also permits accurate assessment of the defocus and other CTF paramters. This SSNR is an integral part of later CTF correction procedures, and can ONLY be computed from particles, not entire micrographs. While the core CTF infrastructure could in principle be used to simply phase-flip a whole micrograph, none of the existing user-level programs support this operation.

Using e2ctf.py

Normally this program would be used through the e2workflow.py interface. You can call it manually, from the command-line as well, of course.

e2ctf.py is generally used in 3 stages

Auto fitting

e2ctf.py 1.img --voltage=200 --apix=2 --cs=3.2 --autofit

This runs auto CTF parameter determination, storing the result in a local database.

Manual (GUI) parameter check

e2ctf.py 1.img --gui

Now that you have determined the parameters with the first command, have a look at the results in the interface. Tweak if necessary. Save any changes you make with the Save Parms button

Write output

e2ctf.py 1.img --phaseflip --wiener

Finally generate output. The output will be in the particles directory.

Processing multiple images simultaneously

e2ctf takes multiple images as input, so you can use commands like this:

e2ctf.py *.hdf --voltage=200 --apix=2 --cs=3.2 --autofit
e2ctf.py *.hdf --gui
e2ctf.py *.hdf --phaseflip --wiener

Combining stages into a single command

You can combine more than one of the e2ctf processing stages into a single command. For instance you can run automated fitting and open the GUI with a single command like this:

e2ctf.py 1.img --voltage=200 --apix=2 --cs=3.2 --autofit --gui
e2ctf.py *.img --voltage=200 --apix=2 --cs=3.2 --autofit --gui

EMAN2/Programs/e2ctf (last edited 2010-06-17 19:16:50 by SteveLudtke)