This repository holds the Jupyter notebooks for assessing subcortico-subcortical connectivity identified
via tractography. Within each notebook, there are related functions to load the files, process the data as
necessary, create figures and view tractography in 3D via DIPY. All notebooks are linted with Black
prior to saving. A full list of imported libraries can be found at the end of each notebook.
Note: The embedded table of contents does not work on Github or jupyter lab (which
has its own table of contents module). It does however work in jupyter notebook
Every time the notebooks are updated, this repository will also be updated! If the notebook cannot be loaded on Git, you may need to try again in a bit.
The minimally preprocessed data is made available from the Human Connectome Project. The processed data will be made available through the Federated Research Data Repository (FRDR) - link coming soon!
The processing workflow is currently being developed into a Snakebids workflow for ease of use. Example scripts used for processing the original data can be found in the following repository: https://github.com/kaitj/dbsc. Transformations can also be found in this repository.
A summary of the processing performed for the present study can be found here.
If you are accessing the notebooks on a local copy or on Graham you will need to set up the
virtual environment to be able to run the code cells. The easiest and recommended way to do this
is via poetry (v1.2.0a2). You set up the environment with the following command:
poetry install --with analysis
poetry run python -m ipykernel install --user --name=subcortical_py3
After installing the required libraries, fire up a Jupyter instance with the following command
poetry run jupyter lab. Make sure you select the subcortical_py3 kernel installed!
If you prefer to set up a virtual Python environment (preferably Python 3.7). You can use the following block of code to create the necessary virtual environment.
# Load Python module if on Graham / CBS
# This is not necessary if running on a personal computer
module load python/3.7
# Replace <venv_dir> with the path to set up the environment
python -m venv <venv_dir>
source <venv_dir>/bin/activate
# This will install all the necessary libraries into the environment.
pip install -r <requirements.txt>
# Install jupyter and jupyter lab
pip install jupyter jupyterlab
python -m ipykernel install --user --name=subcortical_py3
If you are trying to create a multipanel figure to perform QC across subjects,
matplotlib will need to be upgraded from 3.3.4 to 3.4.3 (this will break some functionality in
analysis notebooks).
If you are using Poetry, you can edit version listed in pyproject.toml from ~3.3.4 to ~3.4.3.
After updating, run poetry update. You can then run the JupyterLab as before.
If you are using a virtual Python environment, it is easiest to create a new environment. Follow the
instructions above, replacing pip install -r requirements.txt with
pip install -r requirements_multipanel.txt