bhklab / predictio-webapp

An web application that allows users to explore gene signatures in immunotherapy studies of ICB-treated patients, and to obtain gene signature predictions using molecular data of their own studies of ICB-treated patients.

PredictIO Webapp

PredictIO is an web application that allows users to explore gene signatures in 19 immunotherapy studies of ICB-treated patients, and to obtain gene signature predictions using molecular data of their own studies of ICB-treated patients.

The latest version can be found at https://predictio.ca/

Setup Instructions

• Clone the repo

git clone https://github.com/bhklab/PredictIO-webapp.git
cd IOdb

• In the project directory, install all the server dependencies using `pip install -r requirements.txt
• In the project directory, install all client dependencies npm i
• Install Redis in your local environment (brew install redis for Homebrew installation)
• Start Redis server by running redis-server
• Start Flask backend by running flask run
• Navigate to client directory (cd client) and start the client by running npm start (development mode)
• Open http://localhost:3000 to view it in the browser.

Dependencies

• React
• React-Route
• Body-parser

Dev Dependenices

• Eslint

Database Generation

• Download raw_data directory from https://codeocean.com/capsule/6711882/tree capsule
• Run data generation script which would create CSV files for the database tables

python scripts/generate_seed_files.py

• To create tables and seed the database, run the following command:

flask seed-database

This command will run db/seed_database.py with app context. Modify the file if necessary.

Build Instructions

cd client && npm build

Builds the app for production to the build folder. It correctly bundles React in production mode and optimizes the build for the best performance.

Server

• To be determined.

predictio-webapp's Issues

Individual Dataset Page

Implement individual dataset page.

Update database data and RData file.

• Update the database.
• Verify that Biomarker Evaluation and Explore Pre-computed Signature features work with the new data.

Biomarker Evaluation: restrict the number of selected studies to three or more