Csparc2star-missing magnification and pixel size

Import and Export
1

Hello,
I am having difficulty transferring particles from cryosparc to relion after using the csparc2star.py command. I started my processing (motion correction, ctf refinement, particle extraction, etc.) in relion (v3.0) and then continued classifying and refining the particles in cryosparc (v2.12.4). I would now like to bring the refined particles back to relion to run a bayesian polishing job and see if that improves my structure at all. I generated the star file using the most recent version of the pyem package, with the command below:
csparc2star.py --copy-micrograph-coordinates /data/uma/processing/relion3/Extract/job020/particles.star particles_selected.cs passthrough_particles_selected.cs /data/uma/processing/relion3/cryosparc/J803_converted_particles.star --loglevel debug

Everything seems to run okay, but when I try to run a particle extraction job in relion, I get the following error:
ERROR: Preprocessing::initialise ERROR: input _data.star should contain rlnMagnification and rlnDetectorPixelSize.
The output of my “csparc2star.py” command claims to have copied over the Magnification and DetectorPixelSize features, but when I manually open the .star file, I don’t see either of these features in the header of the file. Has anyone on the list seen this problem before? If so, is there a problem with my csparc2star.py command that can be easily addressed? Thanks in advance for your help/input.
Uma

2

you can follow these steps to re-create a proper star file: Feature request - preserve MicrographName column from star file.
Most often all you need from the cs file is the list of selected particles. Use this list to filter the original relion extract star file, add back the header. It works with relion 3.0 and 3.1. TBH, this is my default option…

Peter

3

Just a Relion2/3.0 thing (they got help on the mailing lists).

csparc2star.py --relion2

1 Like
4

Maybe you can try to firstly export the job containing your target particles, which would make the .cs file contain all the parameters, then use csparc2star.py to transform the exported .cs file.