3D Classification require particles.alignments3D

3D Classification
1

Hi everyone,

I am trying to use the 3D Classification job for my extracted particles and using a volume extracted from ChimeraX as an mrc file. The import of the map to CryoSparc didn’t have any problems, but when I tried running the 3D classification job with 2 inputs, (1) exported particles from 2D class and (2) the imported volume as initial volume, I encoutered the following error:

Traceback (most recent call last):
  File "cryosparc_master/cryosparc_compute/run.py", line 60, in cryosparc_compute.run.main
  File "/gpfs/nisbfs/home/cryosparcuser_dgx/cryosparc/cryosparc_worker/cryosparc_compute/jobs/runcommon.py", line 1010, 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.

I thought that the job would not require any alignments to begin? If it does, then should I run an ab-initio job which has the alignment, then merge that together with the imported map? If yes, then how can I do this?

Thank you very much for your help!

2

It does require alignments - it is classification without alignments (meaning your particles already need to be aligned to some kind of consensus refinement). I would recommend running a homogeneous or non uniform refinement first.

3

hi @olibclarke, Thank you for your suggestion! I did find that the homogenous refinement does do alignments.3D for the particles and export the particles with that.
And using it the 3D classification is running!

4

Hi! sorry for intrusion. But how do you do alignments3D? In align 3D maps it kinda does it, but needs couple of volumes?

I have unbanned particles and the resolution dropped significantly when I used them for reconstruction of the volume I got with binned particles. I think due to noise and misalignment…

5

Hi there,

I don’t think you can use unbinned and binned particles together? MIght be a better idea to re-extract binned particles to unbinned size and try again. Because there will be differences in size and resolution, the reconstruction won’t be good.

I am not sure how to do alignment only, but all Refinement jobs I think have 3D alignments compounded into them.

All the best!