Hello good people!
We are using the default preprocessing pipeline "default preprocessing pipeline for volume-based analyses with indirect normalization to MNI-space when fieldmaps are available" to analyze resting state data. We have phase and magnitude images that we have put in a secondary dataset.
It seems like the vdm files are created, but then there seems to be an issue with CONN identifying/finding the vdm files it creates for the fieldmaps.
This is the error it gives:

Failed 'Realign & Unwarp'

Error using spm_vol>spm_vol_hdr
File "/gpfs/group/HHD/default/Sandlab/AbSANT fMRI R21/Anna's Qualifying/PWA fMRI raw data/preprocessing attempt5/psuma_r21_12/functional/vdm_f202202071130-0004-00001-000001-01.nii" does not exist.

The file name it is looking for has the same name as the functional (resting state) file for each participant with a "vdm_" prefix added. However, the vdm file it created in the previous step has the same name as the raw phase/magnitude fieldmap files.
Any suggestions?
Thanks!

Hi Chaleece,

That is strange, the VDM-creation step should have created a new secondary dataset labeled 'vdm' pointing to the correct vdm files (only when that does not exist CONN will try to find vdm_*.nii files named like the functional data). Could you please double-check by typing:

conn_datasetlabel('vdm')

and seeing whether that returns a number (if the 'vdm' dataset exists) or an empty array (if it does not)?

Best
Alfonso
Originally posted by Chaleece Sandberg:

Thanks so much, Alfonso! Yes, that was the problem. The array is empty. It's weird because there appear to be vdm files in the folder. Now we need to find out why it didn't create the vdm dataset :)
Originally posted by Alfonso Nieto-Castanon: