Local Installation & Configuration🔗

Create Conda Environment🔗

There currently isn’t a lab-wide datalad environment set up on MSI with the correct dependencies for the steps below. Please create your own conda environment (also refer to instructions here if needed - also note that you may need to first load conda module):

Check for available env names and make sure you choose something unique (e.g. for bobsrepository, my conda env name is datalad_BR):

module load conda
conda info –-envs
conda create --name <my_datalad_env>

Install DataLad and git annex🔗

source activate my_datalad_env
conda install -c conda-forge datalad
conda install -c conda-forge git-annex=*=alldep*

Note that the git-annex install of the standard distribution will require that you include the signature=v4 flag in git annex initremote commands in order to work correctly. Do NOT use the MSI modules for datalad or git-annex because the versions are too old to work for this process (git-annex version 8.20201 or higher required).

Install OpenNeuro CLI🔗

• Install deno: conda install conda-forge::deno
• Run deno run -A jsr:@openneuro/cli --help - should see usage if it’s working

Configure Git Credentials🔗

Make sure your git credentials are configured. This will be required when creating the sibling GitHub repository:

cd ~
git config --global --add user.name "Bob McBobFace"
git config --global --add user.email bob@example.com

Also make sure to run the following command in order to automatically use “main” instead of “master” for the main branch of your new GitHub repository that will be created:

git config --global init.defaultBranch main

Store AWS Credentials & GitHub Token (Optional)🔗

Whenever you work with your DataLad repository, you need to first activate the conda environment and then set your AWS access and secret keys as environmental variables in order to be able to push changes to AWS:

module load conda
source activate datalad_BR
export AWS_ACCESS_KEY_ID="<access_key_id>"
export AWS_SECRET_ACCESS_KEY="<secret_access_key>"

Alternatively, you may store the secret keys in .aws/fcpindi.sh and simply run the following in place of the 2 export lines above:

source ~/.aws/fcpindi.sh

This file should include the following variables, used by DataLad to interact with AWS and GitHub:

• AWS_ACCESS_KEY_ID
• AWS_SECRET_ACCESS_KEY
• DATALAD_CREDENTIAL_GH_TOKEN