Reassign particles to micrographs w/ missing items

Particle Picking
1

Right now Reassign Particles to Micrographs will fail if some of the particles have micrograph paths that aren’t in the exposure input. Would be nice to get “failed_x” outputs for those instead, and have usable outputs for the matching subset.

3 Likes
2

Hi @DanielAsarnow , we’ve noted this feature request.

1 Like
3

This is much needed! I was actually under the impression that this would not happen, so I was confused that my job was failing, until I found Daniel’s post.

Please make it available as soon as possible =)

@DanielAsarnow, what workaround did you employ meanwhile?

Cheers!
Andre’

4

Not tested, but if you run a curate micrographs job first, you should get a particle output just corresponding to the particles on the input micrographs - I believe this should work as a workaround

1 Like
5

I think I did something like @olibclarke’s suggestion here.

1 Like
6

Hi @kstachowski

When importing micrographs from relion and particles from those micrographs, and reassigning particles on CS processed micrographs, naming discrepencies occur like

Example source micrograph filename:

 108-1_100-102_Sep24_22.07.58_X-1Y-1-0

Example query micrograph filename:

 108-1_100-102_Sep24_22_07_58_X-1Y-1-0

leading to errors like “ assert qname in inv_index_source, “Could not find match for %s” % qname“

Relion adds _ near dates.

Is there a workaround this?

Thanks

7

Hi @atomicresolution3.5

The easiest way would be to use a bash command like sed -E 's/([0-9]{2}).([0-9]{2}).([0-9]{2})/\1_\2_\3/g' input.txt > output.txt where all occurrences with format 22.07.58 are converted to 22_07_58. For that command, the input file should be your particle.star file from RELION. If the timestamp format changes your command would need to change to accommodate the difference in format from what you posted.

Best,

Kye

2 Likes
8

Hi @kstachowski

Thanks, I know this would be the simplest way forward, but it would be great to have something of a process in cryosparc where for discrepancies like this, we could use a custom awk script and keep our upstream process still connected.

I used awk for now.

Thanks though for the swift reply.

Best

2 Likes