BOO 2023

In this short guide, we will introduce the Noteable environment and show how to install the QSPRpred package.

Noteable

You can access the Noteable environment from the Brightspace module of the course. Open the link and select "Standard (Collaborative session)". This will launch a JupyterLab server instance and redirect you to it. You can look at it as your personal virtual computer running the Linux operating system. JupyterLab is a development environment that enables you to write and execute code on this machine. This is facilitated either with simple scripts or with interactive notebooks (Jupyter Notebooks). The latter are documents that contain both code and text, and allow you to run the code and see the results in the same document. This is what we will use in this course most often.

Using the Terminal

We will use the shell (terminal) to install the QSPRpred package, and it is also the best way to execute programs and manage our files. You can open a new terminal by clicking on the + sign in the left sidebar and selecting "Terminal". You can also open a terminal by clicking on the "File" menu and selecting "New" -> "Terminal". Or you can use the launch menu which you should see when you open the JupyterLab server. Click the black icon and a terminal will open. You can type commands in the terminal. Try these:

ls # list files in the current directory
ls -la # list all files in the current directory, including hidden files, and show more information
pwd # print the current working directory

The output of the pwd command should be /home/jovyan. This is your home directory. You can create new directories and files here, and you can also delete them. However, you can also do so in the GUI via the file view on the left side of the screen. Try to create a directory to see it appear in the file view.

mkdir test # create a new directory called "test"

You can open and create files in the GUI, but you can also do so in the terminal. Try to create a new file called test.txt in the test directory.

cd test # change directory to "test"
touch test.txt # create a new file called "test.txt"

If you need to go back to the previous directory, you can use the cd .. command. Try to go back to your home directory.

cd .. # go back to the previous directory

You should now be back in your home directory:

pwd # print the current working directory

Your home directory can also be abbreviated as ~. List the home directory to see the test directory you created earlier:

ls ~ # list the home directory

The command line is a very powerful tool, and you can access basically anything on the system from it. For example, try to list hardware information about your computer with the following commands:

egrep "proc|vendor|model|MHz" /proc/cpuinfo
grep "Mem" /proc/meminfo

It is useful to know how to use the command line, see this cheatsheet for more examples, but let's move on to installing software and eventually the QSPRpred package.

Installing Software with Anaconda

We like to use the Anaconda Python distribution to install software. It is a free and open-source distribution of software packages that comes with a package manager called conda. It is already available on our machine:

conda --version # check the version of conda

Conda operates with so called 'environments'. In fact, we are already operating within the base environment. You can see this in the terminal prompt. You can list currently available environments with the following command:

conda env list # list available environments

It is good practice to create a new environment for each project. This way, you can install different versions of the same software in different environments, and you can also easily share your environment with others. Let's create a new environment called qsprpred. Because notable is a bit special and only persists your home directory across virtual machines, we need to specify the path where we want to create the environment:

conda create python=3.10 --prefix ~/software/envs/qsprpred

This will create a new environment called qsprpred in the ~/software/envs directory. You can activate the environment with the following command:

conda activate ~/software/envs/qsprpred

You should see the environment name change in the terminal prompt. We only installed the Python interpreter in this environment, but we can install many new programs this way. For example, try to install the htop utility:

conda install htop

You can now run the htop program to see a list of running processes and usage of resources on your machine:

htop

Pretty neat, right? You can exit the program with the q key. Let's install the QSPRpred package now:

pip install git+https://github.com/CDDLeiden/[email protected] # installs a version of the code by specifying the tag (v2.0.0.dev1)

This will use the pip installer to fetch a version of the QSPRpred package from GitHub and install it in your environment. You can check by importing the package in Python:

python -c "import qsprpred; print(qsprpred.__version__)" # execute Python code from the command line

Running the QSPRpred Tutorial

You can browse the source code of QSPRpred on GitHub and the tutorial code is there as well. You can browse the notebooks in this directory, but you cannot run or edit them on GitHub. That is why we need to create our own copy of the code. We can do this by cloning the repository:

git clone -b v2.0.0.dev1 https://github.com/CDDLeiden/QSPRPred.git ~/software/QSPRPred

However, before we navigate to the tutorial directory, we need to link the qsprpred environment to JupyterLab. We can do this by installing the ipykernel from this environment to our home directory:

python -m ipykernel install --prefix=$HOME/.local/ --name 'Python-QSPRpred'

WARNING: Make sure to execute this with the python in the qsprpred environment.

This will enable a new kernel in JupyterLab that we can use to run notebooks. You should be able to see the new option after clicking the large + sign in the left sidebar. If it does not appear, refresh your browser. You will now have this kernel option every time you open a Noteable session.

Now we can navigate to the tutorial directory with the file browser and run and edit the notebooks with QSPRpred examples.