Skip to content

Singularity images

Singularity is a container platform designed for high-performance computing environments. It enables users to create and run containers that encapsulate software and its dependencies in a portable, reproducible, and secure manner.

At the CIR server, we maintain several pre-built Singularity images to support the execution of various neuroimaging software tools, ensuring consistency across analyses and simplifying software management.

Location of Singularity images

Singularity images ara available at the following location: /scratch/singularityContainers

Available images

fmriPrep

fMRIPrep is a robust and user-friendly fMRI preprocessing pipeline that handles variations in scan acquisition protocols and that requires minimal user input, while providing easily interpretable and comprehensive error and output reporting.

Location: /scratch/singularityContainers/fmriprep.simg

Example to run fmriPrep: 

singularity run --bind $(pwd):$(pwd) --cleanenv ./scratch/singularityContainers/singularity/fmriprep.simg \
        path/to/data/dir path/to/output/dir \
        participant \
        --nthreads 20 \
        --omp-nthreads 10 \
        --fs-license-file path/to/freesurfer_licence \
        --output-spaces MNI152NLin2009cAsym \
        --fs-no-reconall \
        --work-dir  path/to/work/dir

heudiconv

heudiconv is a flexible DICOM converter for organizing brain imaging data into structured directory layouts, such as in BIDS.

Location: /scratch/singularityContainers/heudiconv.sif

halfpipe

ENIGMA Halfpipe is an intuitive tool designed to streamline reproducible fMRI analysis, covering preprocessing, individual, and group-level analyses. It leverages fMRIPrep for advanced preprocessing without requiring BIDS data conversion, and allows for real-time computation of common resting-state and task-based fMRI metrics.

Location: /scratch/singularityContainers/halfpipe.sif

mriqc

mriqc is an open-source tool that automatically extracts image quality metrics (IQMs) from structural, functional, and diffusion MRI data, without needing a reference image. It is designed for minimal preprocessing and is built using modular Nipype workflows that integrate tools like ANTs and AFNI. MRIQC adheres to BIDS and BIDS-App standards for compatibility and follows strong software engineering practices to ensure robustness and reliability, with continuous testing on diverse datasets.

Location: /scratch/singularityContainers/mriqc.sif

Running a Singularity Image

If the data to be preprocessed is already on the server, you are ready to run a Singularity image. E: 

$ singularity run --cleanenv fmriprep.simg \
    path/to/data/dir path/to/output/dir \
    participant \
    --participant-label label

Further documentation

For more information on how to use Singularity check here.