Neurodesktop access

How to use Neurodesktop on the server

Author

Published

September 27, 2024

Neurodesktop is a Linux environment specifically developed for open and reproducible neuroimaging. For details, see https://www.neurodesk.org

To use the neurodesktop on our lab’s server, please follow the steps below.

Neurodesktop is accessible from the office, via VPN (for employees only, or for those who’s student account has been explicitly activated for server access by the IT), and in the student computer room.

Each neurodesktop user is limited to 256GB RAM out of 504GB and to 20 out of 32 CPU cores.

Getting started

In your browser, type https://pslg-nv1.uni-graz.at and sign up for a account
After you sign up, You account needs to be authorized by the technical staff. This may take a few days. If there is no progress, please check back with us. @Lab members: please use the same account name as your local server user account. That way you will automatically have access permissions for all your data via the serverpath /storage/neurodesk/users/{username}. For example, Natalia should use natalia_nv as username.
Once your account is authorized, you can sing in with your credentials and you will see a JupyterLab session. You can start the neurodesktop by clicking on the icon. The session can be suspended and resumed later, with all your applications remaining as you left them.

You should see the desktop with all the apps:
1. You can use https://pslg-nv1.uni-graz.at/hub/home to control your container (start/stop) or change your password.
  
  There are different Neurodesktop versions available on the JupyterHub. If you need another Jupyter Notebook or another version of the Neurodesk, please let us know. Additionally, if anything is not working as expected, do not hesitate to report it.
Enjoy!

General info

Storing data

If working via neurodesk, you SHOULD store data in your home directory. This data is kept after server restart and is backed up

Sharing data

All neurodesk users have read and write permissions to /shared neurodesktop folder, which should be used to store things used by everyone (like atlases, software, shared data)

Data transfer

There are currently two alternatives
1. If you don’t have a local server account, you can transfer the data via nextcloud client, which is included in neurodsktop
2. “file upload” button at the top of the jupyter lab menu (up-pointing arrow) or download by right-clicking on the relevant files
Using FreeSurfer

To force freesurfer/freeview to see the /storage folder, source freesurfer from the terminal as follows
```
ml freesurfer/7.4.1
cd /storage
freeview
```
To set up custom subject directory in a container, use the following (instead of export SUBJECTS_DIR=)
```
export SINGULARITYENV_SUBJECTS_DIR=[yourpath] export APPTAINERENV_SUBJECTS_DIR=[yourpath]
```
Using fitlins

Using FitLins (not a part of neurodesktop basic set of tools) with Singularity:

singularity run /shared/poldracklab_fitlins_latest-2022-10-17-32670dc66bf7.simg [your fitlins commands]