Global CTF refinement

CTF Refinement
1

Hi,
I have a couple of sessions where groupID was not defined when the micrographs were imported into the session. Consequently, I am writing scripts to set groupID for particles, according to their originating micrographs, in mid session. If I perform Global CTF refinement with options Fit Tilt and Fit Trefoil, are ctf/shif_A, ctf/tilt_A, and ctf/trefoil_A the only cs (file) fields that are updated?

Here is a section from the output

[CPU: 2.30 GB] Previous group params:
[CPU: 2.30 GB] Shift (A): [0. 0.]
[CPU: 2.30 GB] Tilt (A): [0. 0.]
[CPU: 2.30 GB] Tilt (mrad): [0. 0.]
[CPU: 2.30 GB] Trefoil (A): [0. 0.]
[CPU: 2.30 GB] New group params:
[CPU: 2.30 GB] Shift (A): [0.09158283 0.09160191]
[CPU: 2.30 GB] Tilt (A): [ 9452.94957951 12304.3968832 ]
[CPU: 2.30 GB] Tilt (mrad): [0.35010924 0.4557184 ]
[CPU: 2.30 GB] Trefoil (A): [3119.76531163 -319.86023387]
[CPU: 2.30 GB] Updating tilt values for group.
[CPU: 2.30 GB] Updating trefoil values for group.
[CPU: 2.30 GB] ------- Group 2 -------

2

Did you consider comparing the values inside the relevant columns of the cs files in question? There is a tutorial explaining the cs file format and providing motivating examples.

3

Hi,
Yes there is a tutorial and I have written the scripts to make the changes. However, the tutorial does not indicate which fields are modified during group CTF refinement. I was concerned that my script does not extract information from all fields that are updated during CTF group refinement. Are the above mentioned fields the only fields that are updated?

4

Hi @rkhayat,

Currently, the cs files don’t store information about which fields have been modified within a result group, so in general one can only tell whether the groups as a whole are modified (i.e. CTF refinement will modify ctf but not alignments3D for example). Wolfram’s suggestion of comparing columns of the cs file before and after a job is, in general, the only way one would determine which exact fields are modified within a result group.

For CTF Refinement in particular though, you are correct in stating that ctf/shift_A, ctf/tilt_A and ctf/trefoil_A are the only CTF parameters modified if the “Fit tilt” and “Fit trefoil” options are activated.

Best,
Michael

1 Like
5

Greeting there,

I am working the Global CTF refinement now. I have question as below that want to be confirmed before I start it:

I am using version 3.3.1. I have a couple of sessions collected at different date. I imported them separately. After the particle picking on each import, I merged them for processing. If I had right understanding, the grouping of imports are retained on my particles. I can do global CTF refinement directly. Am I correct?

Thanks in advance for your clarification.

6

Dear @yanhezhao,

If you imported movies/micrographs/particles through separate jobs, each imported set should receive a unique exposure group ID by default, which is used to group micrographs/particles together when doing Global CTF Refinement.

You can always run an “Exposure Group Utilities” job on your particles in the “Info only” mode to see how many groups you have, as well as how many particles/micrographs of each group there are.

Best,
Michael

7

Thank you so much for the clarification, @mmclean.