Setting IntendedFor for fmap/epi image

[dkp]

Under Series Mapping, I set IntendedFor on the fmap/epi image (this is the reverse phase-encode B0 image for correcting the DWI image).

When I got the Dataset review, IntendedFor was set to "no-series" and I had to reset it again.

[dlevitas]

@dkp are you able to share/email me the subject's data? I'm unable to reproduce this issue with your previously supplied sub-CAM003 data and one of my other datasets.

[dkp]

@dlevitas I was using my sub-188 dataset from OSF ezbids_dicom_188.zip.

BTW, thanks for fixing the event upload issue from Mac vs Windows! The whole interface seems cleaner and smarter than when I last tried it. My students love your interface. I was also excited to see you thinking about offering a containerized version.

[dlevitas]

Thanks @dkp,

This issue should now be fixed, there was a strange file indexing problem being caused by the func/events files. Let us know if you encounter any other issues.

[dkp]

@dlevitas Uh oh, so sorry to report this, but now the dataset review has gotten some weird set of values in the Intendedfor field, despite my having set those IntendedFor targets correctly in the Series Mapping ; (

[dlevitas]

@dkp,

Thanks for letting me know.

I see that there are two separate subject IDs (sub-188 and sub-sub188). I assume this is from the Subjects/Session page, where the subject ID was accidentally set to sub118 rather than 118. Does fixing that address your issue? Regardless, I'll take a look at the IntendedFor mapping.

[dkp]

Aha, let me recheck my dumbness ; )

[dlevitas]

I'm unable to reproduce the issue with the same subject and mismatched sub IDs.

If you're using the same URL from yesterday, you would need to restart a new one to reflect the changes I made. Apologies for not making that more clear if that was the case.

[dkp]

@dlevitas It looks like that was just a mistake on my part, but it looks like the problem with IntendedFor mac vs Windows files has resurfaced ; (
Yesterday it seemed like the specification of the events had been unified for Mac and Windows, but now that problem seems to have returned. Was I wrong to believe the Mac/Windows thing was resolved, or did I make some other mistake (I am using the Windows format, but uploading from a Mac):

bids-validator@1.11.0 1: [ERR] 'IntendedFor' field needs to point to an existing file. (code: 37 - INTENDED_FOR) ./sub-188/fmap/sub-188_dir-PA_epi.nii.gz Evidence: func/sub-188_task-nad1_run-3_events.nii.gz

Please visit https://neurostars.org/search?q=INTENDED_FOR for existing conversations about this issue.

I am starting with a fresh upload of the data every time I test.

[dlevitas]

It shouldn't have, the changes I made yesterday didn't touch the fix to #68.

I don't have access to a Windows; however, for testing purposes I used your Windows layout for the events files:

sub-188
├── task-nad1_run-01.tsv
├── task-nad1_run-02.tsv
├── task-nad1_run-03.tsv
└── task-nad1_run-04.tsv

Even with this structure I didn't encounter the issue. To give me a better sense of what might be going on, could you email me your ezBIDS URL?

[dkp]

@dlevitas I don't have access to Windows either! I am hoping for something that works for my students regardless of whether they are on Mac or Windows. I'm trying again, here: https://brainlife.io/ezbids/#6477ac2f55144c1c33fc9a53

Bids validator is complaining: bids-validator@1.11.0 1: [ERR] 'IntendedFor' field needs to point to an existing file. (code: 37 - INTENDED_FOR) ./sub-188/fmap/sub-188_dir-PA_epi.nii.gz Evidence: func/sub-188_task-nad1_run-3_events.nii.gz

[dlevitas]

The solution was to clear the browser cache. Before, newly pushed changes weren't being recognized, hence the validator issue.