Q: How do I go about doing CTF correction in EMAN1
A: Much of this is described in the 4 step tutorial interface, but the description here is more up to date.
CTF correction in EMAN1, is, unfortunately, a somewhat time-consuming process. This has been massively simplified in EMAN2, but as EMAN2 still remains 'experimental' you may wish to stick with EMAN1 for a while longer for serious refinement work. It is a VERY good idea to make sure you are using a current snapshot version of EMAN1, rather than 1.8. There are 3 steps for optimally performing CTF correction in EMAN1:
1. Determine a 1-D Structure Factor from 3-5 of your images
- Use images over a reasonable range of defocuses
- Don't forget to pick a good amplitude contrast (10-20% works best for cryo data)
2. Use fitctf to fit all of your particle stacks automatically
- use the same %AC you used above
- guess at a B-factor based on the fitting above. For a high-end 300 keV instrument this may be as low as 50-100. For a typical 200 FEG scope it will be in the 150-200 range,
and for older/LAB6 scopes it may be as high as ~1000. Note that these are 2 different conventions for B-factor, EMAN1 B-factors are multiplied by 4 to compare with the MRC convention.
- run this program once for each stack of particles (one stack per frame). The particles should be unmasked and unfiltered !
- fitctf <particles from one frame> sf=<sf file from above> apix=<A/pix> V=<voltage> Cs=<spher aber> acfix=<amp contrast, 0.1 is 10%> Bhint=<guess at EMAN1 style B-factor>
3. Go back and manually check your results
- Run 'ctfit *.tnf' This will open the results of the automatic fitting.
- If you need to correct any of the fit values (you very well may):
- select the image in the upper right list in the control panel
- make necessary adjustments
- press the 'set' button below the lower right list in the control panel, this will save your parameters to disk
- Do NOT use the 'phase flip' menu item when working this way, since you loaded '.tnf' files instead of the original particle stacks, it will not work right.
- Instead, after correcting all of the CTF parameters and 'set'ing them, exit ctfit, and use 'applyctf' to flip the phases and store the parameters:
- applyctf will get the CTF parameters from the memories you set in ctfit (ctfparm.txt file)
- applyctf image1.hed image1.fix.hed setparm flipphase
- applyctf image2.hed image2.fix.hed setparm flipphase
- …
Now you can continue with processing. Typically:
mkdir refine cd refine lstcat.py start.hed ../*.fix.hed lstfast.py start.hed cp start.hed start.img (make threed.0a.mrc) refine …
Note that the above commands create a start.hed/img file which is not really in imagic format. Rather it's a text file linking to the *fix.hed particles you created, so it's very important that you not rename or remove those files. EMAN1 will treat these text files as if they were real image files.