Welcome to fMRIPrep Workbench Documentation¶
This documentation covers the fMRIPrep Workbench template, a generalizable workflow for fMRI preprocessing that handles the full pipeline from scanner acquisition downloads to fMRIPrep execution.
Version: 0.3.0
Overview¶
The fMRIPrep Workbench transforms fMRI preprocessing scripts into a generalizable workflow that handles:
- Automated transfer of scanner acquisitions from FlyWheel to server
- DICOM to NIfTI conversion with dcm2niix/heudiconv
- Dummy scan removal and fieldmap-based susceptibility distortion correction setup
- Quality control verification (metadata and volume counts)
- fMRIPrep anatomical-only workflows (for manual FreeSurfer editing)
- FreeSurfer manual editing utilities (download/upload with automatic backup)
- fMRIPrep full workflows (anatomical + functional)
- FSL FEAT statistical analysis (Level 1, 2, 3 GLM)
- Data management utilities (tarball/untar for inode optimization)
- Interactive TUI launcher for all pipeline steps
Quick Start¶
Option 1: Using Docker (Recommended)¶
# Pull the image
docker pull shawnschwartz/fmriprep-workbench:latest
# Clone the repository
git clone https://github.com/shawntz/fmriprep-workbench.git
cd fmriprep-workbench
# Start the container
./fmriprep-workbench start
# Launch the interactive TUI
./fmriprep-workbench launch
Option 2: Using Singularity (HPC Clusters)¶
# Download or build the Singularity image
singularity build fmriprep-workbench.sif docker://shawnschwartz/fmriprep-workbench:latest
# Run the TUI
singularity exec --bind $(pwd):/workspace fmriprep-workbench.sif /opt/fmriprep-workbench/launch
Option 3: Manual Installation¶
1. Create Repository from Template¶
Click "Use this template" on the GitHub repository to create your own copy.
2. Clone Your Repository¶
git clone https://github.com/your-username/your-repo-name.git
cd your-repo-name
3. Configure Settings¶
Copy the configuration template and customize for your study:
cp config.template.yaml config.yaml
# Edit config.yaml with your study-specific parameters
4. Set Up Subject List¶
cp all-subjects.template.txt all-subjects.txt
# Add your subject IDs (one per line, just the number without "sub-" prefix)
5. Run the Pipeline¶
# Interactive mode (recommended)
./launch
# Or manual execution (14-step workflow)
./01-run.sbatch # Step 1: FlyWheel download
./02-run.sbatch # Step 2: DICOM conversion
./03-run.sbatch # Step 3: Prep for fMRIPrep
./04-run.sbatch # Step 4: QC metadata
./05-run.sbatch # Step 5: QC volumes
./06-run.sbatch # Step 6: fMRIPrep anat-only (optional)
./toolbox/download_freesurfer.sh # Step 7: Download FreeSurfer (optional)
./toolbox/upload_freesurfer.sh # Step 8: Upload FreeSurfer (optional)
./07-run.sbatch # Step 9: fMRIPrep full workflows
./10-fsl-glm/setup_glm.sh # Step 10: FSL GLM setup
./08-run.sbatch <model-name> # Step 11: FSL Level 1
./09-run.sbatch <model-name> # Step 12: FSL Level 2
./10-run.sbatch <model-name> # Step 13: FSL Level 3
./toolbox/tarball_sourcedata.sh # Step 14: Tarball utility
What's New in v0.3.0¶
Major Features:
- 🐳 Docker Containerization: Complete Docker support with automated CI/CD to Docker Hub
- 📊 FSL FEAT Analysis: Full statistical analysis pipeline (Level 1, 2, 3 GLM)
- 🧠 FreeSurfer Editing: Download/upload utilities with automatic backups
- 📚 Documentation Migration: Modern MkDocs system with dark mode and PDF export
- 🔢 14-Step Pipeline: Complete end-to-end workflow from acquisition to analysis
Breaking Changes:
- Directory renaming:
07-run-fmriprep→09-run-fmriprep,08-fsl-glm→10-fsl-glm - Pipeline expanded from 7 to 14 steps
- Documentation format changed from Sphinx to MkDocs
See Changelog for full details.
Features¶
Automated Transfer : Transfer scanner acquisitions from FlyWheel to server
DICOM Conversion : Convert DICOM to NIfTI using heudiconv/dcm2niix
Dummy Scan Removal : Remove initial dummy scans based on configuration
Distortion Correction : Fieldmap-based susceptibility distortion correction setup
Quality Control : Built-in verification of metadata and volume counts
fMRIPrep Integration : Run anatomical-only or full workflows with FreeSurfer manual editing support
FreeSurfer Editing : Download/upload utilities with automatic backup for manual surface editing
FSL FEAT Analysis : Complete statistical analysis pipeline (Level 1, 2, 3 GLM)
Data Management : Tarball utility to optimize inode usage on shared filesystems
Interactive TUI : User-friendly launcher for all 14 pipeline steps
YAML Configuration : Portable, cross-platform configuration system