Hi,
I am not sure what is wrong with my star file, but something is leading to the below error. I took my consensus refinement from cryosparc and polish in relion. I then edit a star file made by subset selection from previous 3d classification to point to the shiny particles within Polish/job/asdf folder.
I get the below error. I will post the star file header and first line in a comment below.
[CPU: 249.2 MB] Loaded 8816 particles.
[CPU: 249.2 MB] Common fields:
[CPU: 249.2 MB] ctf/accel_kv : {300.0}
[CPU: 249.2 MB] ctf/cs_mm : {2.7}
[CPU: 249.2 MB] ctf/amp_contrast : {0.1}
[CPU: 249.2 MB] ctf/exp_group_id : {1}
[CPU: 249.2 MB] blob/psize_A : {0.83}
[CPU: 249.2 MB] blob/shape : [400 400]
[CPU: 249.2 MB] --------------------------------------------------------------
[CPU: 249.2 MB] Making plots...
[CPU: 286.5 MB] Traceback (most recent call last):
File "cryosparc_master/cryosparc_compute/run.py", line 84, in cryosparc_compute.run.main
File "/home/lab/software/cryosparc2/cryosparc2_master/cryosparc_compute/jobs/imports/run.py", line 445, in run_import_particles
fig = plotutil.plot_images_simple(dispparts, rows=4, cols=8, radwn=6, figscale=1.5)
File "/home/lab/software/cryosparc2/cryosparc2_master/cryosparc_compute/plotutil.py", line 495, in plot_images_simple
rD = img.get_original_real_data()
File "/home/lab/software/cryosparc2/cryosparc2_master/cryosparc_compute/particles.py", line 113, in get_original_real_data
return self.blob.view().copy()
File "/home/lab/software/cryosparc2/cryosparc2_master/cryosparc_compute/blobio/mrc.py", line 124, in view
return self.get()
File "/home/lab/software/cryosparc2/cryosparc2_master/cryosparc_compute/blobio/mrc.py", line 121, in get
data = n.fromfile(file_obj, dtype=self.dtype, count= n.prod(self.shape)).reshape(self.shape)
**ValueError: cannot reshape array of size 0 into shape (400,400)**
Here is the star file. Importing the subset selection star file pointing at the original extract folder worked fine, but now doesn’t work when pointing to the shiny folder. Thanks for any help.
# version 30001
data_optics
loop_
_rlnVoltage #1
_rlnSphericalAberration #2
_rlnAmplitudeContrast #3
_rlnOpticsGroup #4
_rlnImageSize #5
_rlnImagePixelSize #6
_rlnImageDimensionality #7
300.000000 2.664582 0.100000 1 400 0.830000 2
# version 30001
data_particles
loop_
_rlnImageName #1
_rlnMicrographName #2
_rlnCoordinateX #3
_rlnCoordinateY #4
_rlnAngleRot #5
_rlnAngleTilt #6
_rlnAnglePsi #7
_rlnOriginXAngst #8
_rlnOriginYAngst #9
_rlnDefocusU #10
_rlnDefocusV #11
_rlnDefocusAngle #12
_rlnPhaseShift #13
_rlnCtfBfactor #14
_rlnOpticsGroup #15
_rlnRandomSubset #16
_rlnClassNumber #17
_rlnGroupNumber #18
_rlnNormCorrection #19
_rlnLogLikeliContribution #20
_rlnMaxValueProbDistribution #21
_rlnNrOfSignificantSamples #22
000003@Polish/job029/rawdata/name_00008Gr_00001Sq_v02_00002Hln_00002Enn_frames_shiny.mrcs J4052/imported/name_00008Gr_00001Sq_v02_00002Hln_00002Enn_frames.mrc 4869.000000 3077.000000 -177.85141 96.841782 12.038957 -1.91938 0.155625 16516.466797 15820.322266 -26.81530 -2.35405 0.000000 1 1 8 1 0.376435 2.588725e+05 0.651087 2
Running relion_reconstruct on the subset of particles I am trying to import gives some clue and it seems multiple particle stacks have the issue. Not sure what can fix it. The consensus shiny.star file imported fine, but not the subset selection.
“ERROR: readMRC: Image number ### exceeds stack size ### of image”
Hi!
Has somebody found a fix for this problem?
I’ve encountered the same issue and I also haven’t been able to fix it.
Best,
Armando
Hi Armando,
I used subset select in relion to make a new star file based on rlnClassNumber, which I had to grab from a previous star file. I lost a bit under 10% of particles which seem to have been automatically removed with subset select, but I am not really sure.
Hope this helps.
Is there any update on this? Is it fixed in a recent patch? Or maybe I am doing something wrong or my star file is wrong.
If I run two extractions in csparc, one binned and the other not binned, can I run 3Dclass on the binned one and then point that star file to the unbinned extraction (ie particle names are exactly the same between two extract jobs with identical particle input, only difference is binning)? If I polish the consensus from the unbinned extraction, I should be able to point the 3dclass subset star file to these particles, but it throws the “ValueError: cannot reshape array of size 0 into shape (400, 400)”
Seems like maybe there are some rogue/corrupt/empty particles, have you tried the “Check for corrupt particles” job, maybe that might give some useful info?
I think that gives the following error: “Warning: Could not find file asdf” and “AssertionError: Files referenced by the input star file are missing.” Which I bypass by deleting the line(s) referenced (sometimes multiple steps) until it does not throw that error.
How do you check for corrupt particles?
there is a new “Check for corrupt particles” job type - I have not used it, but maybe in this case it could give some helpful info?
That is good to know! Just as a check, the consensus shiny.star imports fine, but pointing a subset to it does not import. The subset was generated by duplicating the parent extract job in cryosparc with binning, refine, 3D classify in relion, subset select, point the subset select to shiny.star. Maybe it would be best to use grep to generate a new star file grabbing lines from shiny.star based on the particle names from the subset star file.