Neuroimaging software containers

The initial release of the BIRC User Container for Research (BURC) is now available, providing a pre-configured, reproducible environment for a variety of neuroimaging analyses. Containers have several use cases:

  • Preserving the software environment used for an analysis, so that the analysis can be reproduced later
  • Running software on a high performance computing (HPC) or other system where you may not have the privileges necessary to configure software
  • Rapidly setting up a software environment on a new computer

The BURC provides a large (40GB+) smorgasbord of useful packages in an Ubuntu Linux environment that can be run on your own computer using Docker or on a HPC system with Singularity. More information and documentation is available on GitHub. Additional support is available via the BIRC_HELP-L listserv. To subscribe, send an email to listserv@listserv.uconn.edu with the body:

SUB BIRC_HELP-L Firstname Lastname

The BURC is available as:

  • A prebuilt Singularity container at /scratch/birc_ro/containers/burc.img on the Storrs HPC system
  • A prebuilt Singularity container for download (internal link)
  • An exported Docker image for download (internal link)
  • A Docker build file on GitHub.
  • As a Docker image on the iMacs in the data processing room at BIRC

Installed software includes:

DICOM converters

  • dcm2niix
  • dicm2nii (requires Matlab)
  • pydicom, nibabel, dcmstack, bidskit, heudiconv

Neuroimaging Analysis

  • AFNI
  • FSL with patched eddy_cuda
  • Freesurfer 6.0
  • ANTs
  • DTIPrep

Spectroscopy

  • Tarquin
  • Gannet 3.0 master (requires Matlab)
  • VESPA (in the vespa conda-2.7 environment)

M/EEG

  • Fieldtrip (requires Matlab)

Neuroimaging Pipelines

  • fmriprep
  • mriqc
  • C-PAC
  • nipype

Statistics

  • R
  • pystan and Rstan
  • MCMCglmm

Python

There are several versions of python installed:

  • A system python 2.7 at /usr/bin/python
  • Anaconda python 2.7 (/usr/bin/env python)
  • A python 3.6 Anaconda environment (python3)

Python environments

There are several Anaconda-based environments. Switch between them using source activate name

  • python3 Python 3.6 with nibabel, nipype and pystan
  • cpac (Python 2.7)
  • poldrack (Python 3.6) with fmriprep and mriqc
  • vespa (Python 2.7) with vespa