FIX: --cifti-output missing required space

[mgxd]

Addressing #211 - first step will be to add --cifti-output to circle testing

[mgxd]

@madisoth could you verify this patch fixes your issue? I pushed a docker image with this patch to nipreps/nibabies:fix-cifti-output.

[madisoth]

Hi @mgxd, I got a test subject processed and the cifti output looks good.

But I did hit a separate error; think it's just this fMRIPlot call needs to be updated to work with niworkflows after the bump to 1.5.3

File: /out/sub-<SUBJECT>/log/20220429-144654_9d9649e2-9e55-4899-9115-d4936fd67b2d/crash-20220429-203608-tmadison-conf_plot-8d2bbe66-b1cb-4484-9fc2-3c581aedf9e0.txt
Working Directory: /work/nibabies_22_0_wf/single_subject_<SUBJECT>_wf/func_preproc_ses_<SESSION>_task_auditory_run_01_wf/carpetplot_wf/conf_plot
Inputs: 
confounds_file:
confounds_list: [('global_signal', None, 'GS'), ('csf', None, 'GSCSF'), ('white_matter', None, 'GSWM'), ('std_dvars', None, 'DVARS'), ('framewise_displacement', 'mm', 'FD')]
in_func:
in_mask:
in_segm:
str_or_tuple:
tr: 1.2
Traceback (most recent call last):
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/plugins/multiproc.py", line 67, in run_node
    result["result"] = node.run(updatehash=updatehash)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 516, in run
    result = self._run_interface(execute=True)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 635, in _run_interface
    return self._run_command(execute)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 741, in _run_command
    result = self._interface.run(cwd=outdir)
  File "/opt/conda/lib/python3.8/site-packages/nipype/interfaces/base/core.py", line 428, in run
    runtime = self._run_interface(runtime)
  File "/opt/conda/lib/python3.8/site-packages/nibabies/interfaces/confounds.py", line 413, in _run_interface
    fig = fMRIPlot(
TypeError: __init__() got an unexpected keyword argument 'mask_file'

[mgxd]

oof, yes niworkflow 1.5.x series included some API breaking changes. I'm going to try to get 22.0.x in a stable state, and then start on 22.1.x to alleviate this.

I've backported an older version of fMRIPlot directly into the code - I've pushed it to this branch https://github.com/nipreps/nibabies/tree/docker/fix-cifti-output

You can either:

1. Wait until the Docker patch builds and build a new image
2. Clone that repo and directly patch it in to the image you used above

Sorry for this hassle, and thanks for helping debug.

[pep8speaks]

Hello @mgxd! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2022-05-11 17:30:34 UTC

[mgxd]

following up - did the latest patch end up fixing the error you ran into @madisoth?

[madisoth]

Getting this error now-- just need to check whether sort_rows is str I think? (see above)

220504-22:40:25,633 nipype.workflow WARNING:
	 Storing result file without outputs
220504-22:40:25,642 nipype.workflow WARNING:
	 [Node] Error on "nibabies_22_0_wf.single_subject_<SUBJECT>_wf.func_preproc_ses_<SESSION>_task_auditory_run_01_wf.carpetplot_wf.conf_plot" (/work/nibabies_22_0_wf/single_subject_<SUBJECT>_wf/func_preproc_ses_<SESSION>_task_auditory_run_01_wf/carpetplot_wf/conf_plot)
220504-22:40:25,976 nipype.workflow ERROR:
	 Node conf_plot failed to run on host acn76.
220504-22:40:25,979 nipype.workflow ERROR:
	 Saving crash info to /out/sub-<SUBJECT>/log/20220504-180819_e143d479-2102-4a3d-8f2c-a591b11b5baa/crash-20220504-224025-tmadison-conf_plot-475d8554-7def-49df-96b8-7e71310a2e51.txt
Traceback (most recent call last):
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/plugins/multiproc.py", line 67, in run_node
    result["result"] = node.run(updatehash=updatehash)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 516, in run
    result = self._run_interface(execute=True)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 635, in _run_interface
    return self._run_command(execute)
  File "/opt/conda/lib/python3.8/site-packages/nipype/pipeline/engine/nodes.py", line 741, in _run_command
    result = self._interface.run(cwd=outdir)
  File "/opt/conda/lib/python3.8/site-packages/nipype/interfaces/base/core.py", line 428, in run
    runtime = self._run_interface(runtime)
  File "/opt/conda/lib/python3.8/site-packages/nibabies/interfaces/confounds.py", line 415, in _run_interface
    fig = fMRIPlot(
  File "/opt/conda/lib/python3.8/site-packages/nibabies/utils/viz.py", line 106, in plot
    plot_carpet(
  File "/opt/conda/lib/python3.8/site-packages/nibabies/utils/viz.py", line 249, in plot_carpet
    return _carpet(
  File "/opt/conda/lib/python3.8/site-packages/nibabies/utils/viz.py", line 305, in _carpet
    if sort_rows.lower() == "linkage":
AttributeError: 'bool' object has no attribute 'lower'

[madisoth]

@mgxd the plots issue looks good now, but now I see the medial walls aren't masked out in the 91k dtseries outputs. If I run wb_command -file-information on any of the "91k" func outputs I get this:

CIFTI Dim[0]:                   335
CIFTI Dim[1]:                   96854
ALONG_ROW map type:             SERIES
    Start:                      0.000
    Step:                       1.200
    Units:                      Seconds
ALONG_COLUMN map type:          BRAIN_MODELS
    Has Volume Data:            true
    Volume Dims:                91,109,91
    Volume Space:               2,0,0,-90;0,2,0,-126;0,0,2,-72
    CortexLeft:                 32492 out of 32492 vertices
    CortexRight:                32492 out of 32492 vertices

but I would expect this:

CIFTI Dim[0]:                   335
CIFTI Dim[1]:                   91282
ALONG_ROW map type:             SERIES
    Start:                      0.000
    Step:                       1.200
    Units:                      Seconds
ALONG_COLUMN map type:          BRAIN_MODELS
    Has Volume Data:            true
    Volume Dims:                91,109,91
    Volume Space:               -2,0,0,90;0,2,0,-126;0,0,2,-72
    CortexLeft:                 29696 out of 32492 vertices
    CortexRight:                29716 out of 32492 vertices

In those BOLD dtseries.nii files the values of the medial wall grayordinates are all zeroes though.

[mgxd]

Ok, thank you for the diligence in getting this in shape! This just goes to show how much a regression test on the CIFTI outputs is needed.

Now I am not sure if we ever were producing HCP compatible CIFTIs (well, at least since porting over the dcan-infant-pipeline alignment)...

I've pushed a new patch, I've tested it locally and everything looks good, but will hold off on merging this until you confirm.

[mgxd]

BTW, @madisoth do you want to add yourself to the zenodo - if so, feel free to open a PR