Unbin symmetry expanded particles

After running a job with binned particles that have been symmetry expanded, like 3D classification on a single protomer, is it possible to unbin for reconstruction? Usually I would use the original unbinned refinement output particle blob in the reconstruction particle input, but this loses the symmetry expansion.

Pathway is:
Refinement with unbinned particles (C4) → downsample → homogeneous reconstruct (C4) → symmetry expansion → 3D classification → heterogeneous reconstruct (C1) with unbinned refinement particle blob has ~1/4 of the # particles.

I thought maybe also running symmetry expansion on the refinement with unbinned particles and using that blob as input might work, but it gives the same 1/4 particle number.

Do I have to symmetric refine, symmetry expand, then downsample the symmetry expanded particles? I saw a post about this making n larger particle stack, so seems like a waste of disk space if that’s still the method.

Hi @cryofun – we’ve recorded your feature request for a simpler way to handle metadata of particles which have been symmetry expanded. For now, this is best done with cryosparc tools. There’s a script which should do something similar to what you are looking for here.

1 Like