Filesystem Organization

This page will tell you how our files and folders are organized.

• /projects: This folder is backed up nightly and is accessible from all our computers (and the SCC). Every user in the lab has their own folder with the same name as their account. Put your important code/data there. Our back up space is limited though, so whenever possible try to use scratch instead.
• /scratch: This folder is NOT backed up but is also accessible from all our computers (and the SCC). Like with projects, every user has their own folder here. This is a good place for working files that can easily be regenerated if they get deleted by accident. Once you know your work is good enough to backup, move it to /projects
• /home/${USERNAME}: Your home folder, which is unique on each machine. While your /projects or /scratch are shared between machines the /home folders on each machine are local to that machine. This means anything in your home folder on machine X will not be available on machine Y. Your code, data, writing, and figures should therefore be kept in /projects or /scratch. This is particularly important for the queue because when your job is sent to another node of the cluster it won't be able to see any of the files you can see in your home folder and your job will fail!
• /archive/data: Contains data from past and current studies that the lab manages. This is another one of our directories that is available from the SCC. See Data Organization and Data Naming for information on how the data in our archive is organized (in great detail!) if you would like to know more.
• /archive/code: We store copies of our lab created code (e.g. datman, ciftify and the dashboard) here for use on the archive data.
• /archive/code/config: This is where we store the bash scripts + config scripts for our archive. The <study>_management.sh scripts describe exactly what we're running on each study every night. The <study>_settings.yml files detail which types of scans are collected per site for each study that we're managing. You can look at these to get an overview of what data we should have in each study folder.
• /external: Contains data from publicly-available data sets. The sub-folder admindocs holds all of the access documents for our public data sets. Some of our public data sets have been moved to Scinet. If this is the case there will be a README file left behind in their folder indicating how to find them.
• /archive/resources: This contains some atlases we have available, some of our clinical data and a guide to using our MRS analysis pipeline.