Track proximal sources of functional GIFTIs

[tsalo]

Related to #3261.

Changes proposed in this pull request

• Add an outputnode to init_ds_volumes_wf with the saved bold path.
• Add sources field to init_bold_surf_wf's inputnode.
• Infer Sources for functional GIFTI outputs.
  • If the user has requested T1w-space outputs, then this should be bold from the associated ds_volumes_wf and fsnative2t1w_xfm
  • If the user didn't want T1w-space volumetric outputs, then this should be the raw BOLD file, motion_xfm, boldref2anat_xfm, fsnative2t1w_xfm.

Documentation that should be reviewed

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 73.86%. Comparing base (ecf8580) to head (79eb79e).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3263      +/-   ##
==========================================
+ Coverage   73.82%   73.86%   +0.04%     
==========================================
  Files          56       56              
  Lines        4279     4286       +7     
==========================================
+ Hits         3159     3166       +7     
  Misses       1120     1120              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[tsalo]

The GIFTI metadata from ds005 look good:

{
  "RepetitionTime": 2.0,
  "SliceTimingCorrected": false,
  "Sources": [
    "bids:raw:sub-01/func/sub-01_task-mixedgamblestask_run-01_bold.nii.gz",
    "bids::sub-01/func/sub-01_task-mixedgamblestask_run-01_from-orig_to-boldref_mode-image_desc-hmc_xfm.txt",
    "bids::sub-01/func/sub-01_task-mixedgamblestask_run-01_from-boldref_to-T1w_mode-image_desc-coreg_xfm.txt",
    "bids:deriv-0:sub-01/anat/sub-01_from-fsnative_to-T1w_mode-image_xfm.txt"
  ],
  "TaskName": "mixed-gambles task"
}

This is how they should look if the user doesn't want T1w-space outputs. If they do, then I think the Sources should look like:

{
  "Sources": [
    "bids::sub-01/func/sub-01_task-mixedgamblestask_run-01_space-T1w_desc-preproc_bold.nii.gz",
    "bids:deriv-0:sub-01/anat/sub-01_from-fsnative_to-T1w_mode-image_xfm.txt"
  ],
}

[effigies]

I would include the white-matter and pial surfaces for each hemisphere, since those determine the coordinates at which the BOLD is sampled. If that's difficult to do at the moment, I'm fine with postponing it to another PR.

[tsalo]

The hemisphere loop happens within init_bold_surf_wf, so I need to add those sources there, but I can't pin down where exactly those two files are grabbed in the workflow. I think it'll have to wait for another PR.