While it would be too difficult (and maybe impossible) to run heudiconv with the --datalad option and then merge the resulting git/git-annex, it would be relatively easy to run datalad save -m 'Add sub-X/ses-Y' after the merge-datasets step.
We can just add a --datalad argument and reference the datalad Singularity image I have installed.
Based on the simplification proposal, we're actually using heudiconv's datalad option, since we'll be writing out to the final directory instead of scratch.
While it would be too difficult (and maybe impossible) to run heudiconv with the
--datalad
option and then merge the resulting git/git-annex, it would be relatively easy to rundatalad save -m 'Add sub-X/ses-Y'
after the merge-datasets step.We can just add a
--datalad
argument and reference the datalad Singularity image I have installed.