Environment: Python 2.7

Dependencies: python : numpy pandas matplotlib json argparse bioread dcm2bids cmd : dcm2niix edf2asc

Using the command line

1. conda create --name [name] --file [/Path/to/requirements.txt]

   • where 'name' is the name you would like to give your environment
   • where '/Path/to/requirements.txt' is the path to requirements.txt, downloaded from github

2. conda activate [name]

   • where 'name' is the name you gave your environment in step 1

   • if command line throws an error during this step, try instead using source activate [name]

3. pip install bioread dcm2bids argparse

4. conda install -c conda-forge dcm2niix

Additional

5. Move the file edf2asc (provided via email) to the following directory: [/Path/to/Anaconda/Folder]/envs/[name]/bin
   • where 'Path/to/Anaconda/Folder is the path to the anaconda source folder
   • where 'name' is the name you gave your environment in step 1
6. Download DTIPrep from the following link: www.nitrc.org/projects/dtiprep. Select the version appropriate for the system.
7. Move contents of downloaded DTIPrep files (unzipped) to the following directory: [/Path/to/Anaconda/Folder]/envs/[name]/bin
   • where 'Path/to/Anaconda/Folder' is the path to the anaconda source folder
   • where 'name' is the name you gave your environment in step 1

1. COINS-images2bids/COINS_BIDS_setup.py : pulls relevant information from COINS runsheet and prepares for dcm2bids

• arguments
  • --runsheet /Path/to/COINS/runsheet
  • --keysheet /Path/to/COINS/keysheet (COINS-images2bids/COINS_run_sheet_key.csv)
  • --temp_json /Path/to/config.json (COINS-images2bids/config.json)
  • --source_dir /Path/to/subject/source/directory
  • --subject_list /Path/to/list/of/subjects
• needed for input:
  • COINS runsheet
  • keysheet (downloaded from github)
  • model json file (downloaded from github)
  • a folder containing diicom data
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output:
  • source_dir/selected_scans.txt - contains file names of good scan runs for each subject, based on COINS sheet
  • source_dir/selected_physio.txt - contains file names of good physio data for each subject, based on COINS sheet
  • source_dir/selected_track.txt - contains file names of good tracking data for each subject, based on COINS sheet
  • source_dir/[subID]/[subID].json - contains file names of good scan runs for the subject, based on COINS sheet
  • source_dir/error_log.txt - a text file listing any subjects from the COINS sheet that did not compile correctly

2. COINS-images2bids/batch_dcm2bids.py : converts scan diicoms to nifti, outputs in BIDS format

• arguments
  • --source_dir /Path/to/subject/source/directory
  • --bids_dir /Path/to/bids/folder
  • --COINS_BIDS /Path/to/selected_scans.csv (step 1 output)
  • --subject_list /Path/to/list/of/subjects
• needed for input
  • selected_scans.csv (step one output)
  • a folder containing diicom data
  • a destination folder for the sorted bids data (if you give a path in --destination that does not already exist, it will be created)
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • destination - contains scan data organized in bids format (note: at this stage, participants with multiple sessions, e.g. sub-80084 and sub-80084b, are treated as separate subjects. They will be combined in a later step.)
  • source/error_log.txt - a text file listing any subjects from the COINS sheet that did not compile correctly
    • most common is 'subject not in source folder' - just indicates that your sub_dir did not contain diicom data for a particular subject, does not indicate malfunctioning code, does not cause a problem with the output produced from other subjects
    • this information is appended to the error log from step 1 - if a participant threw an error in step 1, they will appear a second time in this step

3. eprime-codes/batch_ce.py : generates csv files from eprime output

• arguments
  • --source_dir /Path/to/source/directory
  • --subject_list /Path/to/list/of/subjects
• needed for input
  • a folder containing all subjects' dicom data
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • /source_dir/[subid]/eprime_csvfiles/ - folder containing csv versions of the eprime data

4. eprime-codes/csv2tsv.py : converts csvs from step 7 to tsvs, puts in BIDS format

• arguments
  • --bids_dir /Path/to/bids/directory
  • --source_dir /Path/to/subject/source/folder
  • --subject_list /Path/to/list/of/subjects
• needed for input
  • a folder containing all subjects' dicom data
  • a folder containing bids organized data
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • /bids_dir/[subid]/func/ - tsv version of eprime data

5. eprime-codes/feat_excel.py : generates feat onset vectors

• arguments
  • --bids_dir /Path/to/bids/directory
  • --souce_dir /Path/to/subject/source/directory
  • --subject_list /Path/to/list/of/subjects
• needed for input
  • a folder containing bids organized data
  • a folder containing all subjects' dicom data
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • /bids_dir/subj_acc.txt - file containing subject accuracy information
  • /bids_dir/[subid]/feat/ - folder containing task event vectors based on eprime csv files (all stimuli; faces only; shapes only; accurate faces; inaccurate faces; accurate shapes; inaccurate shapes)

6. COINS2physio/COINS_physio.py : generates jsons for physio processing

• arguments
  • --runsheet /Path/to/selectedphysio.csv (step 1 output)
  • --source_dir /Path/to/source/directory
  • --temp_json /Path/to/physio-template.json (COINS2physio/physio-template.json)
  • --subject _list /Path/to/list/of/subjects
• needed for input
  • selectedphysio.csv - produced in step 1
  • folder containing diicom data
  • COINS2physio/physio-template.json - downloaded from github
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • input_dir/physio-templates - a folder containing a json that details the file names for the good physio runs for each participant
  • COINS2physio/errorlog.txt - a list of subjects whose physio data did not compile/QC correctly.
    • This is distinct from the error_log.txt in steps 1 and 2. Information does not append to the original error log, and this error log is located in the code source folder.

7. proc-biopac-COINS/proc_biopac_coins.py : process and QC physio data

• arguments
  • --source_dir /Path/to/source/directory
  • --subject_list /Path/to/list/of/subjects/to/process
  • --template_directory /Path/to/template/directory (step 3 output)
  • --bids_dir /Path/to/BIDS/directory
• needed for input
  • folder with diicom data (esp. 01+physio)
  • a list of subjects to process (.txt file, formatted 'sub-80084', one subject per line)
  • folder with physio jsons for each subject (output from step 3)
  • a destination folder for bids output (should be the same as given in step 2)
• output
  • project_directory/[sub_ID]/QC/physio - a directory for each subject containing physio QC information and images
  • bids_dir/[sub_ID]/func/[].tsv - a .tsv file with the time course of the physio data, in bids format
  • project_directory/physio_values.csv - a list of relevant physio values extracted from the data (including heart rate, respiration rate, etc)

8. COINS2tracking/COINS_tracking.py : generates jsons for tracking data

• arguments
  • --runsheet /Path/to/selected_track.csv (step 1 output)
  • --source_dir /Path/to/source/directory
  • --temp_json /Path/to/tracking-template.json (COINS2tracking/tracking-template.json)
  • --subject_list /Path/to/list/of/subjects
• needed for input
  • selected_track.csv (output from step 1)
  • folder with diicom data
  • COINS2tracking/tracking-template.json - downloaded from github
  • a .txt file containing a list of subjects formatted 'sub-[subid]', one line per subject
• output
  • input_dir/tracking-templates - folder with json containing file names for good eye tracking runs for each subject
  • COINS2tracking/error_log.txt - list of subjects whose eye tracking data fails to compile

9. proc-tracking-COINS/proc-tracking-COINS.py : processes and QCs eye tracking

• arguments
  • --source_dir /Path/to/source/directory
  • --subject_list /Path/to/list/of/subjects/to/run
  • --template_directory /Path/to/template/directory (step 5 output)
  • --bids_dir /Path/to/BIDS/directory
• needed for input
  • folder with diicom data
  • list of subjects to process (.txt file, formatted 'sub-80084', one subject per line)
  • folder with jsons containing file names for good eye tracking runs (step 8 output)
  • bids formatted destination folder (should be same as given in step 4 and step 2)
  • eprime tsvs in each subjects bids directory (run steps 3, 4, and 5)
• output
  • project_directory/[subid]/QC/eyetracking - folder containing QC info and images for eye tracking data
  • bids_dir/[subid]/func/[].tsv - .tsv file with time series for eye tracking data

10. proc-DWI-COINS/proc-DWI-COINS.py : processes and QCs DTI scans

• arguments
  • --source_dir /Path/to/source/directory
  • --subject_list /Path/to/list/of/subjects/to/run
  • --protocol_template /Path/to/DTI-template.xml (proc-DWI-COINS/DTI-template.xml)
  • --COINS_BIDS /Path/to/selected_scans.csv (step 1 output)
  • --bids_dir /Path/to/BIDS/directory
• needed for input
  • folder with dicom data
  • list of subjects to process (.txt file, formatted 'sub-80084', one subject per line)
  • protocol template for DTIPrep - downloaded from github
  • selected_scans.csv (step 1 output)
  • a nifti version of the dti data in each subject's bids folder
  • bids formatted destination folder (should be same as given in steps 2, 4, and 9)

10. final-bids/finalize_bids.py : combines runs from same subject into single file --bids_dir /Path/to/bids/directory