3D variability analysis with mask and particles created in relion (scipion)

Hello,
I am new to cryosparc and wanted to test 3D variability analysis for one of the refinement job I completed in relion. I created export particles in relion, and moved these particles stacks and the mask to a server where we have cryosparc installed. When I run 3D variability it gives me following error

[CPU: 95.4 MB] Traceback (most recent call last):
File “cryosparc_worker/cryosparc_compute/run.py”, line 52, in cryosparc_compute.run.main
File “/sparkki/cryosparc/software/cryosparc/cryosparc2_worker/cryosparc_compute/jobs/runcommon.py”, line 847, in check_default_inputs
assert False, 'Non-optional inputs from the following input groups and their slots are not connected: ’ + missing_inputs + ‘. Please connect all required inputs.’
AssertionError: Non-optional inputs from the following input groups and their slots are not connected: particles.alignments3D. Please connect all required inputs.

Does anyone know if I am doing something wrong?

Hi @srehan,

3D Variability needs to have aligned particles (i.e. particles that have been processed through a consensus refinement) and it is complaining that the imported particles that you have do not have alignment information (poses and shifts). You can check the “Outputs” tab of your import particles job to see exactly what was imported. Also, in the streamlog of your import particles job, you will see notes about what fields were found in the .star file and what is being imported.
Did you use a .star file that includes pixel size, pose, and shift information? Probably this would be a .star file created during/after refinement in Relion.

If import is not workable, you can also just run a refinement job in cryoSPARC using the imported particles and imported initial volume. This will align all the particles and you should be able to then run 3DVA from the output of refinement.

Thanks. I created particle stacks from consensus refinement in relion and as you can see alignment parameters are included. Screen Shot 2021-01-18 at 21.03.51. It is .star file. and stacks are mrcs.

This is very big data set and I don’t want to repeat the whole workflow again.

Issue fixed.
Apparently there was issue of random subset while creating stacks in relion. I checked “Ignore half-set split if missing” while particle import in cryosparc and everything run smoothly. Lets see how 3DVA ends.

best
SR

1 Like