Link Search Menu Expand Document
Reward Project Documentation

CogTrain

Table of contents
  1. Data Processing Flow & Important Links:
  2. Plan for the Data
  3. Data Acquisition
  4. Download and Storage
  5. Curation Process
    1. BIDS Validation:
    2. BIDS Optimization:
  6. Preprocessing Pipelines
    1. fMRIPrep (version 20.2.3)
      1. Exemplar Testing:
      2. Production:
    2. XCP-ABCD (version 0.0.8)
      1. Production:

Data Processing Flow & Important Links:

  • Flow diagram that describes the lifecycle of this dataset curation and preprocessing:

flow

Plan for the Data

  • Why does PennLINC need this data?

    Acquired at UPenn

  • Goal:

    Curate and preprocess an amalgam of datasets for a harmonized PennLINC resource

Data Acquisition

  • Data acquired by Joe Kable
  • Describe the data:
    • number of subjects = 166
    • types of images = bold (4 runs task-RISK, 4 runs task-ITC, rest), T1w, fieldmaps, DWI

Download and Storage

  • Original data available from Kable Lab
  • Source data (NIfTI) on CUBIC in /cbica/projects/wolf_satterthwaite_reward/original_data/bidsdatasets/CogTrain.
  • Data was copied to /cbica/projects/wolf_satterthwaite_reward/Curation/bidsdatasets/CogTrain and checked in to datalad after removing PHI (below)
  • JSON’s within origial_data were updated using cubids-add-nifti-info.

Curation Process

  • Data curation by Tinashe Tapera on the CUBIC project user wolfsatterthwaitereward
  • Link to final CuBIDS csvs: NA

BIDS Validation:

  • BIDS validation output at /cbica/projects/wolf_satterthwaite_reward/Curation/code/validate_outputs/CogTrain/CogTrain_validation_2.csv:

    • INCONSISTENT_PARAMETERS ( Not all subjects/sessions/runs have the same scanning parameters. ) : 244 counts

    • PARTICIPANT_ID_PATTERN (Participant_id column labels must consist of the pattern “sub-".) : 166

Data at this stage were approved for preprocessing.

BIDS Optimization:

  • All cubids optimization results are available in /cbica/projects/wolf_satterthwaite_reward/Curation/code/iterations/iter<ITERATION_NUMBER>/CogTrain/

  • Final optimization resulted in /cbica/projects/wolf_satterthwaite_reward/Curation/code/iterations/iter1/CogTrain/CogTrain_summary.csv

Preprocessing Pipelines

fMRIPrep (version 20.2.3)

  • Tinashe Tapera was responsible for running preprocessing pipelines/audits on CUBIC

Exemplar Testing:

  • Used cubids to create exemplar dataset: cubids-copy-exemplars /cbica/projects/wolf_satterthwaite_reward/Curation/bidsdatasets/CogTrain/BIDS /cbica/projects/wolf_satterthwaite_reward/Testing/CogTrain/exemplars_dir /cbica/projects/wolf_satterthwaite_reward/Curation/code/iterations/iter1/CogTrain/CogTrain_AcqGrouping.csv
  • Path to exemplar dataset (annexed to datalad): /cbica/projects/wolf_satterthwaite_reward/Testing/CogTrain/exemplars_dir
  • Path to fmriprep container: Original in ~/dropbox, datalad in /cbica/projects/wolf_satterthwaite_reward/Testing/exemplars_test/fmriprep-container

Testing directory was deleted to save space on CUBIC on 12/2/21, once production completed

Production:

  • 291/293 sessions completed fMRIPrep successfully
  • Path to production inputs: /cbica/projects/wolf_satterthwaite_reward/Curation/bidsdatasets/CogTrain/BIDS
  • Path to fmriprep run command: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/fmriprep-multises/analysis/code/fmriprep_zip.sh
  • Path to production outputs: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/fmriprep-multises/output_ria
  • Path to fmriprep production audit: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/fmriprep-multises-audit/FMRIPREP_AUDIT.csv
  • Path to freesurfer production audit: NA

XCP-ABCD (version 0.0.8)

Production:

  • 289/293 sessions completed XCP successfully
  • Path to production inputs: /cbica/projects/wolf_satterthwaite_reward/Curation/bidsdatasets/CogTrain/fmriprep-multises/merge_ds
  • Path to xcp run command: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/xcp-multises/analysis/code/xcp_zip.sh
  • Path to production outputs: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/xcp-multises/output_ria
  • Path to xcp production audit: NA
  • Path to xcp derivatives: /cbica/projects/wolf_satterthwaite_reward/Production/CogTrain/xcp-derivatives/XCP
  • Path to xcp derivatives (concatenated): NA