No option to correct CTF during ReconstructTomograms in Relion5 GUI

[mvorlander]

Hi,
I want to reconstruct tomograms with CTF correction in RELION 5, but the GUI-generated command adds --noctf by default. Please add an option to perform CTF correction.
Many thanks,
Matthias

[EuanPyle]

Hi,
the option to perform CTF correction was removed (and noctf added by default) as there were some issues with the ctf correction of the tomogram when we were testing it. I don't know if this has been fixed yet. In any case, RELION doesn't use these tomograms for averaging, so it shouldn't matter, unless you want to analyse these tomograms using software outside of the RELION pipeline which I assume is what you're trying?

In the meantime, if you want CTF corrected tomograms, you can if you have used IMOD for the Align Tilt Series. Simply go into AlignTiltSeries/jobXYZ/external/[Tomo]/, open etomo via etomo *.edf and carry out the process in etomo to generate the CTF corrected tomogram. One small note if you do this: After generating the Aligned stack, make sure that the .tlt file contains the exact same angles as the .rawtlt file, if it doesn't just copy the values from .rawtlt into .tlt. As long as the coordinates of your CTF corrected tomogram from IMOD match your RELION ones, you should be ok.

[mvorlander]

Hi Euan,

thanks a lot for the reply. I want to run pytom template matching using per-tilt weights, and for this I need CTF corrected tomos.
The etomo *edf command opens a GUI, which is not ideal when trying to batch process. Is there a way to get the CTF corrected tomograms via the command line?

Many thanks,
Matthias

[EuanPyle]

ahh, I had the exact same problem as you a while back! PyTOM template matching with CTF corrected tomos was amazing for us, worked so well. I actually generated the tomos manually to do this: I wish I knew IMOD well enough to set up a batch CTF correction as it would have saved me a lot of time. It may be worth asking CCPEM mailing list? In the mean time, I'll ask about the internal RELION CTF Correction

[mvorlander]

Great, thanks a lot !
I will look into it and share here if I get something useful :)

[mvorlander]

Hi Euan,

could you expand on why you I should use the .rawtlt angles rather then the the ,tlt angles for reconstructing the tomogram? It seems counterintuitive for reconstructing tomograms ...
Thanks,
Matthias

[EuanPyle]

It's just based off personal experience: when I reconstructed thru etomo I found that the coordinates of the CTF corrected Tomogram didn't perfectly match the RELION one (which is an issue). I found that the .tlt values not matching the .rawtlt was the source of this. Maybe it's not the case anymore? Give both a go and see what happens.
Best
Euan

[EuanPyle]

Hi, the latest version now has CTF correction for tomograms on by default.

[mvorlander]

awesome, thanks a lot for the into!