This repository contains simple scripts for applying the CBIG pre-processing pipeline (https://github.com/ThomasYeoLab/Standalone_CBIG_fMRI_Preproc2016) to a BIDS-formatted directory
# Whole BIDS directory:
BIDS_parallel.sh <num_of_threads_per_machine> BIDS_CBIG_fMRI_preprocess.sh <bids_dir> <config_file>
/fileserver/caladan_ssd/repos/BIDS_to_CBIG_fMRI_Preproc2016/BIDS_parallel.sh \
64 \
/fileserver/caladan_ssd/repos/BIDS_to_CBIG_fMRI_Preproc2016/BIDS_CBIG_fMRI_preprocess.sh \
/fileserver/corrin/collections/GSP \
/fileserver/caladan_ssd/repos/BIDS_to_CBIG_fMRI_Preproc2016/configs/Legacy_GSP.config
# Single Subject:
BIDS_CBIG_fMRI_preprocess.sh <bids_dir> <participant_id> <config_file>
/fileserver/caladan_ssd/repos/BIDS_to_CBIG_fMRI_Preproc2016/BIDS_CBIG_fMRI_preprocess.sh \
/fileserver/corrin/collections/GSP sub-0001 \
/fileserver/caladan_ssd/repos/BIDS_to_CBIG_fMRI_Preproc2016/configs/Legacy_GSP.config
- We have our own fork of the original CBIG repository that is confirmed to work with this code, and was used to create the GSP1000 normative connectome, available here: (https://github.com/bchcohenlab/CBIG)
- Minor modifications were made to allow for a separate smoothing mask for subcortical/cerebellum vs the rest of the brain, as was done in Buckner et al, 2011 but was retired from the CBIG pipeline since.
- The Legacy_GSP.config is the exact processing steps used to generate the GSP1000 connectome, and closely approximates the original Yeo1000 connectome generated in 2010 and viewable in neurosynth.org
If using the original CBIG pipeline
CBIG_preproc_fslmcflirt_outliers.csh
appears to have a bug:- Line 170: set numof_tps =
fslnvols $boldfile"_mc_tmp"
- Should be: set numof_tps =
fslnvols $boldfile
- Line 170: set numof_tps =
- Requires git 2.x (due to use of -C option)
- For Centos 7:
sudo yum install -y https://repo.ius.io/ius-release-el7.rpm https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm sudo yum install -y yum-plugin-replace sudo yum replace -y git --replace-with git224
- For Centos 7:
- Pipeline config files need all options for each step on a single line, not separated with slashes
- Comments in pipeline config files need # marks as the first character of the line (no preceeding spaces)
- Can use a
CBIG_config.sh
file in the<bids_dir>/code
directory (this can be omitted if you set paramaters at log-in) - Requires
jq
be installed to parse the BIDS json files - Requires a
task-rest_bold.json
either in the study directory or the<bids_dir>/<participant_id>/func
directory - Requires freesurfer already available for each subject and in standard directory:
<bids_dir>/derivatives/<freesurfer>/<participant_id>
- Requires that config files MUST use an extension of
.config
- Output is stored in
<bids_dir>/derivatives/CBIG_fMRI_preprocess_<config_name>/<participant_id>