Open butellyn opened 5 years ago
HI @butellyn
Miccai atlas is subject specific, it can only be used directly with structural processing if jlf
is ran. Other option is to use standalone tool: https://xcpengine.readthedocs.io/utils/roiquants.html?highlight=quantifyatlas
Would it be possible to use the labels without jlf? For instance, could I feed an MNI brain with the labels and then use the transform between each of my brains and MNI to get the segmentation?
Yes, you can use labels without jlf. To use quantifyAtlas
, the label (atlas) and the reho/alff must be in the same space(MNI or other) and the same resolution.
I am working on using quantifyAtlas, but I seem to be placing the --entrypoint flag incorrectly. Here is my call:
`SNGL=/share/apps/singularity/2.5.1/bin/singularity SIMG=/data/joy/BBL/applications/bids_apps/xcpEngine.simg OUTPUT=/data/jux/BBL/studies/nasa_antartica/processedData/rest_containerized/xcpEngine
${SNGL} run --cleanenv -B /data:/home/ebutler/data --entrypoint /xcpEngine/utils/quantifyAtlas /data/joy/BBL/applications/bidsapps/xcpEngine.simg -v /home/ebutler${OUTPUT}/${id0}/${id1}/reho/${id0}${id1}_reho.nii.gz -s mean -a /home/ebutler/data/jux/BBL/projects/nasa_antartica/xcpdocker/MNIMiccai/mniJLFLabels.nii.gz -p ${id0},${id1} -r /home/ebutler/data/jux/BBL/projects/nasaantartica/miccaiNodeNames.txt -o /home/ebutler${OUTPUT}/${id0}/${id1}/roiquant/miccai/${id0}${id1}_miccai_reho.csv`
The docs give an example call to docker, not to singularity, so I'm not sure what has to change. Can you see what's wrong?
@butellyn
you have to use exec
instead of run
$SINGL exec -B /data:/home/home/ebutler/data \ /data/joy/BBL/applications/bids_apps/xcpEngine.simg \ /xcpEngine/utils/quantifyAtlas \ -v /home/ebutler${OUTPUT}/${id0}/${id1}/reho/${id0}_${id1}reho.nii.gz \ -s mean -p ${id0},${id1} \ -a /home/ebutler/data/jux/BBL/projects/nasa_antartica/xcpdocker/MNIMiccai/mniJLFLabels.nii.gz \ -r /home/ebutler/data/jux/BBL/projects/nasa_antartica/miccaiNodeNames.txt \ -o /home/ebutler${OUTPUT}/${id0}/${id1}/roiquant/miccai/${id0}${id1}_miccai_reho.csv
I am trying to get regional values, according to the Miccai atlas, for reho and alff. I ran my data through fMRIPrep and then XCPEngine. XCPEngine ran without any errors, but I didn't get out reho or alff according to the Miccai atlas.
Cohort file
Design File
Error message None
Runtime Information I used singularity