Hoffman2:Scripts:setup subject 3
Setup_subject script usage
Setup_subject_3 (also named as setup_subject_bids) is built on top of setup_subject_2. The major goal is to apply BIDs format on the data structure. Besides of using BIDs structure, it added other useful features as well.
Setup_subject_3 has all the functions which exist in setup_subject_2 and works the same way with the same options as setup_subject_2.
- Change participant.tsv to participants.tsv
- Change default of mprage to be skipped
- Add TaskName field to Json files of functional data
- Add default dataset_description.json file and --datadesc option
- Add --dcm2niix option
- Add --intendedfor flag for adding "intendedfor" field to FieldMap
setup_subject_3 [option] basedir subjectID
basedir is where you want the processed data to be stored in your Hoffman account,
subjectID is the subject folder name under /Dicom/data/YOURGROUP/ in Dicom server
Options for both BIDs mode and non-BIDs mode
download data from Dicom server to you hoffman account It requires 3 argument, separated by comma(,) (If you have downloaded the data already, this option is not needed)
1. hostname - (ccn for ccn Dicom server, bmap for BMC Dicom server). 2. userid: Your Dicom server user ID 3. path: path to the subject data in dicom server (/Dicom/data/YOURGROUP)
-d (or --do) : use this option to specify the processing you want to add
After NIFTI convert, these options can be used to run FSL mcflirt, bet and melodic also diag, and mprage.
mcflirt : run mcflirt bet : run bet melodic : run melodic diags : run diagnostics mprage : process anatomical scans sge: submit the job in batch mode
(Using option diags requires mcflirt, bet and melodic to be processed first)
-s (or --skip)
this option to specify the processing you want to skip
mcflirt : run mcflirt bet : run bet melodic : run melodic diags : run diagnostics mprage : process anatomical scans sge: not using batch mode, which means you need to start interactive mode first before running setup_subject_3 script
Set SPM compatibility
Options for BIDs mode only
BIDs format: specify using BIDs structure to group result files. However this option often being used with --config option to choose configuration file. Without configuration file, the name conversion might not be correct under BIDs format rule.
Specifying the BIDs naming configuration file path and name. Users should generate this file based on their data and BIDSs formatting rule. Here's more about the BIDs configuration file.
Specify session directory name. When the subject has multiple sessions, by default the processed results will all go under sub-xxxx/[anat | func | dwi | fmap] without separation based on sessions. If you want to have all the NIfTi converted files to be located in one specific session directory, you can use this option and all files will go under sub-xxxx/ses-NN/[anat | func | dwi | fmap ] (NN represents session number, e.g. 02)
No compression of dicom/ folder. Normally raw dicom data will be compressed into dicom*.tar.gz at the end of NiFTI convert. With this option, it will not compress. This option is good when you know you want to run the setup_subject_3 script again and want to save time on compression and decompression of the data. You should run without this option on the last run so it will save the disk usage.
Force the data's group to be set as the GROUPNAME specified. The GROUPNAME need to be an existing valid group name in Hoffman.
Specify pipeline name used for analysis ( fsl is the default value)
Specify DCM2NIIX module version. The version has to match the available mricron module version in Hoffman2. For example, to use mricron/20200331,
To find the version list to choose from, use
module av mricron
Specify dataset_description.json file location. If there's no dataset_description.json file in your base bids directory and this option is not used, it will copy a default simple template to the bids folder. If your base directory has dataset_description.json already, it won't be over written. This option should have the full path to your dataset_description followed.
Add "intendedfor" field to fieldmap's json file with paths to the functional data. This option is a flag and optional for Bids format.
When this flag is on, the script will look for files under dwi/ and func/ folder and add the file paths to *_epi.json files in fmap/ folder for "intendedfor" field.
folder subjectID will be created under basedir the user specified. Then Dicom Raw data will be under dicom/ folder during NiFTI convert then compressed into dicom.tar.gz Result of NiFTI convert will be under raw/ folder
ls basedir/subjectID analysis behav dicom notes raw
basedir/ |_ sub-[subjectID]/ |_ [ses-SESSIONID]/ |_ anat/ dwi/ fmap/ func/ |_ sourcedata/ |_ sub-[subjectID]/ |_ [ses-SESSIONID]/ |_ dicom/ setup_log.xxxxxxx |_ derivatives/ |_ PipelineName/ |_ sub-[subjectID]/ |_ [ses-SESSIONID]/ |_ anat/ dwi/ fmap/ func/
Load the script module first:
module load ccnscripts
Download data and run NIFTI convert
setup_subject_3 --download ccn,dicomuserid,/dicom/data/YOURGROUP/ -s mprage basedir subjectID
( /dicom/data/YOURGROUP/subjectID is the path to the raw data in Dicom server. We skip mprage here since it will be run by default)
Download data and run NIFTI convert and other FSL processing:
setup_subject_3 --download ccn,userid,/dicom/data/YOURGROUP/ -d mcflirt,bet,melodic,diags,mprage --spm basedir subjectID
Process data only without download
setup_subject_3 -d mcflirt,bet,melodic,diags basedir subjectID setup_subject_3 -d mcflirt,bet,melodic,diags,mprage basedir subjectID setup_subject_3 -d mprage basedir subjectID
setup_subject_3 -s mcflirt,bet,melodic,diags basedir subjectID
Download data and run debabel with Bids format, skip mprage.
setup_subject_3 --download ccn,userid,/dicom/data/YOURGROUP/ --bids --config ~/bidsname-sub123.conf -s mprage basedir subjectID
Log File: provides useful details about what commands were run, their output and the script's status. It's located at the following location for bids mode and non-bids mode