📋 Setup and run group-level processing

At this point, you only have one subject, but you can run group-level MRIQC at any point, and you can run it over and over: It'll just update your results. Crucially, group-level processing generates different reports and tables than participant-level processing, but group-level processing depends on participant-level processing (if it does not find the participant-level processing it should run the participant and group levels, but that will take longer than the allotted time for this script!).

Group-level processing is very fast (as long as the participant-level processing has already been completed). Group-level processing does not need a subject number because it is processing all the participants it finds.

✅ Log on to HPC OOD

✅ Navigate to the mri_practice directory and note the path to this directory (Use pwd to display the path), as you'll need it for the export MRIS statement modification.

✅ Open your copy of runmriqc_group_dev.sh in the OOD editor. You are going to make the same modifications you did with runmriqc_dev.sh, so feel free to open that one and double-check yourself:

✅ Modify the account, e.g., change --account=dkp to your own account name, e.g., --account=akielar.

✅ Modify #SBATCH --mail-user=dkp@arizona.edu to use your email address instead of mine.

✅ Modify the export MRIS statement to point to your mri_practice directory.

✅ Type elgato to switch to that machine.

✅ Run the slurm job (it is really fast if it works):

sbatch --export sub=CAM003 runmriqc_group_dev.sh

My group-level processing took 24 seconds to run! It generates several files labeled group*. Obviously, if you have more than one participant, you should expect this to take longer to run. Note, however, that there is no participant label for the group-level processing.

Caution

Sometimes MRIQC (especially the group level) inexplicably fails to run. If this happens to you, remove any intermediate files it created and try again.

🟢 QA & MRIQC 4: Pr (5)