Hi,
I am currently working with the RBC FreeSurfer derivatives and am trying to understand the structural processing pipeline well enough to apply a comparable workflow to another dataset.
From the RBC paper, I found only a high-level description of the structural processing, but I was not able to locate the exact workflow or command used to generate the segmentation outputs.
I also looked through the public ReproBrainChart GitHub organization, but mainly found repositories for accessing existing datasets and derivatives rather than the original structural preprocessing pipeline. Is there a public repository, container definition, provenance record, or other documentation describing how these FreeSurfer segmentations were generated?
Please let me know if there is a resource I may have overlooked.
Hi @Casper_Cappelen,
Once you get the data following the instruction here, for example:
datalad clone \
https://github.com/ReproBrainChart/PNC_FreeSurfer.git \
-b complete-pass-0.1
All the code used to create this output is in /path/to/PNC_FreeSurfer/code, specifically in extract_freesurfer.sh, you’ll see
bash ./freesurfer_tabulate/collect_stats_to_tsv.sh \
${subid} \
${PWD}/inputs/data/freesurfer \
${PWD}/fstabulate-containers/.datalad/environments/fmriprep-20-2-3/image \
${PWD}/fstabulate-containers/.datalad/environments/neuromaps-main/image \
${PWD}/freesurfer/${subid}
So, recon-all was run with fmriprep-20-2-3, the output of which served as input to fstabulate. You can get the full pipeline, including the containers, with:
cd /path/to/PNC_FreeSurfer
datalad get -r freesurfer_tabulate
datalad get -r fstabulate-containers
If you have trouble with datalad get, you can run the following:
cd /path/to/PNC_FreeSurfer
git config -f .gitmodules submodule.fstabulate-containers.url https://github.com/ReproBrainChart/fstabulate-containers.git
git config -f .gitmodules submodule.fstabulate-containers.datalad-url https://github.com/ReproBrainChart/fstabulate-containers.git
git submodule sync --recursive