Connectome Special

Probabilistic tractography and graph measures (connectome) from dMRI

See repository

Connectome Special runs probabilistic tractography (MRtrix iFOD2) on a white matter FOD, anatomically constrained with a 5-tissue-type mask from the associated structural image, and maps the resulting streamlines to a connectome. The connectome dimensions are set by the SLANT atlas (labels in /SUPPLEMENTAL/slant_origlabels.txt). Outputs include global and nodal graph measures (computed with the Brain Connectivity Toolbox), connectomes weighted by number of streamlines, average streamline length (mm), and average FA, plus a QA document.

Usage

The pipeline is wrapped in a Singularity container (the recipe is in the repository). Input directories must contain PreQual-preprocessed diffusion data and SLANT segmentations; bind them as follows:

singularity run \
    --bind ${workingpath}/PreQual/:/DIFFUSION/,${workingpath}/Slant/:/SLANT/,${workingpath}/Output/:/OUTPUTS/ \
    ${singularity_path}

By default the script generates 10 million streamlines (see Newlin et al. 2023, cited below, on the robustness of streamline count to graph measures). The main.sh wrapper in the repository handles directory binding and job-array execution.

Nancy R. Newlin, Francois Rheault, Kurt G. Schilling, and Bennett A. Landman. “Characterizing streamline count invariant graph measures of structural connectomes” Journal of Magnetic Resonance Imaging, 2023.

Jean-Christophe Tournier, et al. “MRtrix3: A fast, flexible and open software framework for medical image processing and visualisation” NeuroImage, 2019.