Repository with code to calculate and analyse structural connectome data based on FA
Jupyter Notebook Python
Latest commit 4dd0b69 Feb 14, 2017 joebathelt updated
Permalink
Failed to load latest commit information.
FA_Connectome.py updated Feb 14, 2017
README.md
Structural_connectome_analysis.ipynb updated Feb 14, 2017
get_ICV.py
overview.png first commit Sep 20, 2016

README.md

Association between structural connectome organisation and literacy and numeracy in children

This repository contains the scripts to generate FA connectome files and analyse these with graph theory. The main command line script executes the connectome construction from diffusion-weighted images. The analysis of the resulting connectivity matrices is carried out in the Jupyter Notebook file.

To generate FA connectome files:

python FA_Connectome.py --base_directory --out_directory --subject_list --ROI_file

input options:

--base_directory: directory containing raw data. The data needs to be organized in Brain Imaging Data Structure format (http://bids.neuroimaging.io/), i.e. {base_directory}/{subject}/dwi/{subject}_dwi.nii.gz', {base_directory}/{subject}/dwi/{subject}_dwi.bvec, {base_directory}/{subject}/dwi/{subject}_dwi.bval

--out_directory: path of the folder where the output will be directed

--subject_list: subject IDs separated by commata, e.g. CBU16001,CBU16002,CBU16003,etc.

--ROI_file: file containing a parcellation of the brain in MNI space, e.g. AAL atlas.

Included scripts

  • FA_Connectome: Python command line script that generates FA connectome matrices. In this matrix, the connection between two ROIs is expressed as the FA associated with streamlines that intersect with both ROIs.

  • Structural_connectome_analysis.ipynb: Jupyter Notebook with analyses of structural connectome data using graph theoretical methods. I recommend using an online notebook viewer, e.g. http://nbviewer.jupyter.org/

Dependencies

The scripts need a few neuroimaging packages to work:

These python modules are also necessary (These can generally by installed using pip, e.g. pip install nipype):

Overview of the workflow to create brain morphometry maps:

alt tag