Use unbinned particles after classification using binned particles in cryoSPARC

Troubleshooting
1

Hi,
I have the same doubt as this, but I am not able to get the unbinned particles at all.
My pipeline -imported unbinned particle stack from RELION, and went with ab-inito reconstruction and heterogenous refinement (5 classes) after binning the particles to 2x in cryoSPARC 2.9.0.
I have tried 2 methods as described here - Use fullsize images after refinement of rescaled particles

  1. I can’t re-eactract the binned particles to its original size. Tried downsampling option, which doesn’t work, as I have to input my particles from heterogenous refinement, which are already binned to begin with. But, the output shows that the particles are still binned (I checked the metadata).

  2. I have also tried the ‘Extract from Micrographs option’ - after importing the micrographs (Motioncorr done in RELION) and inputting the binned-particles from heterogenous classification. I get this error:

**Traceback (most recent call last):**
**  File "cryosparc2_worker/cryosparc2_compute/run.py", line 78, in cryosparc2_compute.run.main (/home/installtest/deps_manage/cryosparc2_package/deploy/stage/cryosparc2_worker/cryosparc2_compute/run.c:3954)**
**  File "cryosparc2_compute/jobs/extract/run.py", line 100, in run_extract_micrographs_multi**
**    particles_dset =   rc.load_input_group(input_group_name='particles')**
**  File "cryosparc2_compute/jobs/runcommon.py", line 241, in load_input_group**
**    assert is_connected, "Slot %s.%s must be connected!" % (input_group_name, slot['name'])**
**AssertionError: Slot particles.location must be connected!**

None of them are working - any suggestion?
Quereshi

2

Dear @KutaQureshi,
Thanks for reporting.

If you have done the following:

  1. imported unbinned (full-size) particles
  2. ran downsample particles job in cryoSPARC 2.9
  3. processed binned particles

And now you wish to continue processing with the subset of binned particles you have identified, but wish to use the unbinned data, you can do this simply by using the “low level results” UI:

  1. Create a new refinement job (or any other processing you want to do)
  2. Find the subset of binned particles you wish to process, and connect these to the input of the new job
  3. In the job builder of the new job, expand, using the dropdown, the particles input group
  4. Find the original import particles job that contains the unbinned particles and go to the “outputs” tab
  5. From the output of the import job, find the blob output. Drag this and drop it on top of the blob input in the job builder, inside of the particles input group. This will replace the previously connected blob input.
  6. Run the new job.

This will cause cryoSPARC to use the original imported data (blobs) but only keep the subset of particles defined by the output of your processing jobs that used binned particles.
See a screenshot below.

38%20AM
38%20AM2733×917 343 KB

1 Like
3

Hi,
Can we change the box size of selected particles in similar way if we have imported two particle stack with different box size?

Thanks,

Regards,
Priyanka