edvincandon / ffpe-kidney-transplant-classifier

🧬 RT-MLPA assay as a simple molecular tool for diagnosis and classification of rejection in FFPE kidney transplant biopsies.

🧬 SVC Classifier for diagnosis of kidney transplant rejection

The RT-MLPA assay as a simple molecular tool for diagnosis and classification of rejection in FFPE kidney transplant biopsies.

SVC Classifier

The fitted classifiers & standard scalers were pickled and are available in the models/ folder. Two models are available : one with DSA input & one without.

Setup ⚙️

Installation

Create a fresh python venv and install dependencies

python -m venv env
source ./env/bin/activate
pip install -r requirements.txt

Server

Start the Flask server with hot-reload

# run the flask server on http://127.0.0.1:5000
flask run

Web app

The front-end react app is bundled via parcel. Files will be built in the parent /api/static/js folder

cd web
yarn run watch # hot reload
yarn run build # build files

Deployment ⚙️

Heroku

# run the app locally
heroku local web
# deploy to kidney-transplant-classifier.herokuapp.com/
git push heroku master