Getting started with jupyter lab

Setting up your personal conda environment

The server you are accessing has a package management system called conda. Learn more about conda here. Conda works by creating 'environments' which contain software that is only accessible from within that environment. This means that making a change in an environment doesn't affect the rest of the server. This is also useful because you can completely control the software within your environment -- no admin privileges are required to install new programs. In this tutorial, you will initialize conda, create a personal conda environment, install a software program into it, and set up your jupyter notebook to recognize that environment automatically.

These steps must be completed in a 'terminal' session:

1. Initialize conda for your user account:

conda init

2. Close and reopen your terminal session. You should now see (base) infront of the command line.
3. Create your personal environment (replace <username> with your username). When prompted, select y and hit 'enter':

conda create -n <username> ipykernel

3. Activate your environment:

conda activate <username>

4. Install a software program:

conda install -c bioconda salmon

5. Verify that salmon was correctly installed (will show "Salmon" in ascii art if it worked):

salmon swim

6. Link your environment with your jupyter profile:

.conda/envs/<username>/bin/python -m ipykernel install --user --name '<username>' --display-name "<username>'s environment"

7. Return to the Launcher (+ symbol) or press ctrl + shift + L (windows) you should see an option to launch a jupyter notebook with <username>'s environment

NOTE: Every time your open a new terminal, you will be in the (base) environment -- however, you do not have permissions for downloading software into the (base) environment (it will produce an error if you try). Before you can download a program, you will need to run conda activate <username>. Doing this should change the (base) environment to one which displays (<username>).

Setting up git

The first task is to set up a github account, create a new repository, and push your first commit

1. Visit github.com and create an account if you don't already have one
2. Go to your repositories page and create a new one
3. Open jupyter lab and create a new 'terminal' page
4. Use the terminal to create a directory for your workshop project (rename this with the actual name of your project): mkdir myproject
5. Navigate to the project directory: cd myproject
6. Configure that directory as a git repository: git init
7. Add a new file: echo "# myproject" > README.md
8. Add the file (and any others in this folder) to your git repository: git add .
9. Configure your repository author information:

git config user.email "yourName@livemail.uthscsa.edu"
git config user.name "Your Name"

10. Add your first commit (don't forget the message -m "my message"): git commit -m "My first commit!"
11. Link your local repository to your github.com repository: git remote add origin https://github.com/yourName/github_repo_name.git
12. Push your local repository to github.com: git push -u origin master
13. Check your github.com repository -- it should that you successfully added a new file "README.md" which says "myproject".