The project attempts to explore metadata representation in Knowlege Graphs and attempts to build automated systems for optimal querying across multiple knowledge graphs to drive data-driven science. As a simple use case, the project attempts to use such systems to link knowledge graphs containing protein-protein interactions and protein-drug interactions to showcase viable drugs for repurposing needs.
Below are instructions to setup this project on your local machine using a Anaconda or miniconda environment. Please ensure that you have installed git on your machine beforehand. Further instructions can be found here.
Note: These instructions assume that the directory (dsc199) containing all the files will be located in the path C:\Users\<username>. Feel free to change where you store your files, but ensure that you are in the same directory for every step.
1. Clone this project to your local machine.
Fork the repository into your own GitHub account (a fork is a copy of the repository), and run the following on the Anaconda/miniconda prompt:
git clone https://github.com/<your-user-name>/dsc199.git
You may be asked to enter your GitHub credentials.
2. Create a conda environment.
Using a conda environment will help manage modules/dependencies and isolate working environments. The file requirements.txt specifies the Python version and required libraries.
cd dsc199
conda create --name dsc199 python=3.7
Once the environment is created, activate it in the Anaconda/miniconda console.
conda activate dsc199
3. Install modules.
Using the requirements.txt file located in the dsc199 directory, install Jupyter Notebook and the required libraries using pip or conda.
pip install -U jupyter
pip install -r requirements.txt
4. Launch the Jupyter Notebook.
Once the environment is fully set-up, launch Jupyter Notebook in the console.
jupyter notebook
This will open up a web-browser where you will access all of the files associated with dsc199.
For Windows users: Alternatively, you can write a batch script to start-up the notebook instantly.
On Notepad, write the following script:
echo off
CALL C:\Users\<username>\miniconda3\Scripts\activate.bat C:\Users\<username>\miniconda3\envs\dsc199
CD /D C:\Users\<username>\dsc199
jupyter notebook
echo on
Save the file as dsc199.bat. This script will automatically open a browser with Jupyter Notebook access.
Content will be updated as weeks roll by!