Extract from micrographs and alignments3D

Particle Picking
1

Hi,

I was wondering if there is a way to use the extract from particles job without updating the 3D alignments. I noticed that if I disable the “Recenter using aligned shifts” option, it does not update the 2D alignments but still re-aligns the 3D ones. If you have any suggestion or alternative solution, please let me know.

Thank you in advance!
Katerina

2

Hi @katefil,

Welcome to the CryoSPARC forum! Thanks for the report. Could you let us know how you are determining if the 3D alignment shifts are being recentered when “Recentering” is disabled? I have been unable to reproduce this internally. Note that the job may still appear to create alignments{3,2}D as an output of the particle stack, but they are intended to be identical to the input alignments{3,2}D if recentering was disabled.

Best,
Michael

3

Hello Michael,

I have proceeded with an helical refinement job and downloaded the alignments3D, location and blob files. After merging them, I then proceeded to the “Import particles” job and the “Extract from micrographs” job. At the end of the last job, I downloaded the same files again, merged them and compared them with the previous ones. I noticed that _rlAngleRot, Tilt and Psi have been changed. I also tried ‘Homogenous Reconstruction’ after the export job and again the values had changed (compared to the imported and the values from the export job). Am I missing something…?

Thanks in advance for your feedback.
Katerina

4

Hi @katefil,

Thanks for providing more information. Could you clarify what you are using Import Particles for at this stage? If you have combined your original particles with imported new particles after performing helical refinement, and those particles came with alignments in their star file, perhaps this is why the orientations appear to change (because of the new particles)? Extract from Micrographs cannot change orientations of the particles that were input to the job.

If the comparisons are being done between two star files (using csparc2star?), we recommend instead using CryoSPARC Tools for comparison of particle datasets, because the particle unique identifiers (the uid field) are not dropped.

Best,
Michael

5

After the helical refinement, I downloaded the relevant files (output files), merged them and used them as input to a new “Import particles” job. The amount of particles was the same, there were no new particles. I hope this clarifies my workflow.

Indeed, I used the csparc2star command to create the star files. I will proceed with the cryosparc tools, as you suggested, and let you know.

Thank you for the suggestion!

6

Hi @katefil,

Sounds good! I am still unsure as to what csparc2star and Import Particles are being used for in this case, but do let me know if you experience anything unexpected when viewing particles’s alignments using CryoSPARC Tools.

Best,
Michael