I divided my big data set into two groups, now i have two refined particle stacks. I want to combine them together so I can run 3D Ab-initio. Is there a way to combine them cisTEM?
There is not an easy way to do this from within the GUI. One way you can do it is to export the two refinement packages you want to combine, then manually combine the stacks and parameter / star files, then reimport.
Can anyone tell me if this approach worked? I seem to be missing something.
I have 1 dataset processed to 2d classes with one refinement package, and a second dataset processed to 2d classes with a different refinement package.
Once I have picked the appropriate classes, I want to combine the particles associated and start an ab ignition job.
I exported refinement packages from two cisTEM datasets in Frealign format and would like to combine the mrc and par files from the two sets to reimport. I was thinking of using 'cat' to combine them, but wondered if I need to get rid of the header on one or both of the mrc files first in order to do so. If so, is there anything similar that would need to be removed from the par files before combining?
if you the classifications are in the same project, then you can combine them by selecting more than one class selection in the new refinement package wizard.
If they are not, then you will have to export them by clicking on the export button in the refinement package panel, and then use another program to combine the mrc stacks and the par / star files.
So if I create a refinenment package, it asks for a acceleration voltage. One of my datasets is 200kv and the other is 300kv. In many of the other software, this is not used after 2d class averages. I am assuming its the same within freealign, do I assume correctly? If it is the case, than this number is somewhat arbitrary in the refinement package when moving forward?
The CTF is used for basically all stages of the processing. In some older packages you could phase flip your data and then kind of forget about the CTF, but this is not the way it is done in cisTEM or Relion.
You cannot mix voltages in the same dataset in cisTEM currently. I'm not sure you can in other software either? The voltage is most definitely used in the 3D refinement in all packages.
This can certainly be done in sphire and in simple3. I am just somewhat limited in the computational resources that I currently have so wanted to try it out in cisTEM. Would a better approach be to do 2 separate refinements and then sum the maps?
Hi,
There is not an easy way to do this from within the GUI. One way you can do it is to export the two refinement packages you want to combine, then manually combine the stacks and parameter / star files, then reimport.
Cheers,
Tim
Can anyone tell me if this approach worked? I seem to be missing something.
I have 1 dataset processed to 2d classes with one refinement package, and a second dataset processed to 2d classes with a different refinement package.
Once I have picked the appropriate classes, I want to combine the particles associated and start an ab ignition job.
Can someone tell me which files I am combining?
Hello Tim,
Can I used paste to append one file in another one or you have some other better way to combine them?
thanks
Gaya
Hello Tim,
I exported refinement packages from two cisTEM datasets in Frealign format and would like to combine the mrc and par files from the two sets to reimport. I was thinking of using 'cat' to combine them, but wondered if I need to get rid of the header on one or both of the mrc files first in order to do so. If so, is there anything similar that would need to be removed from the par files before combining?
thanks for your help,
Michelle
Hi,
if you the classifications are in the same project, then you can combine them by selecting more than one class selection in the new refinement package wizard.
If they are not, then you will have to export them by clicking on the export button in the refinement package panel, and then use another program to combine the mrc stacks and the par / star files.
Cheers,
Tim
So if I create a refinenment package, it asks for a acceleration voltage. One of my datasets is 200kv and the other is 300kv. In many of the other software, this is not used after 2d class averages. I am assuming its the same within freealign, do I assume correctly? If it is the case, than this number is somewhat arbitrary in the refinement package when moving forward?
I say this because ctf correction often happens in the 2d classification process.
The CTF is used for basically all stages of the processing. In some older packages you could phase flip your data and then kind of forget about the CTF, but this is not the way it is done in cisTEM or Relion.
Hi,
You cannot mix voltages in the same dataset in cisTEM currently. I'm not sure you can in other software either? The voltage is most definitely used in the 3D refinement in all packages.
Cheers,
Tim
This can certainly be done in sphire and in simple3. I am just somewhat limited in the computational resources that I currently have so wanted to try it out in cisTEM. Would a better approach be to do 2 separate refinements and then sum the maps?
That should work to point, but won't give the optimal weighting / filtering.
I assume this would be pretty significant if I was pushing down to like 2Angstrom, but not too much somewhere up at 4Ang?
That's probably true, you may have to sharpen the maps a little more also.
Tim
No worries, ill have a go and see what I get.
Cheers.