buske / variant-subscription-service

Subscribe to important updates on genomic variants of interest, such as changes in ClinVar classification

Variant Subscription Service (VSS)

Subscribe to important updates on genomic variants of interest, such as changes in ClinVar classification

Setup

Prerequisites

• Python 3.3+
• mongodb

Installation

git clone https://github.com/buske/variant-subscription-service.git
cd variant-subscription-service

virtualenv -p python3 .virtualenv
source .virtualenv/bin/activate
pip install -r requirements.txt

Run server

Start mongodb

mongod --dbpath /path/to/db

Run in development mode

FLASK_APP=wsgi.py flask run --reload

Run in production mode

Instructions for Ubuntu 16.04

• Install nginx
• Set up nginx

sudo cp deploy/vss.nginx /etc/nginx/sites-available/vss
sudo ln -s /etc/nginx/sites-available/vss /etc/nginx/sites-enabled/
systemctl restart nginx

• Set up uWSGI service

sudo cp deploy/vss.service /etc/systemd/system/vss.service
sudo systemctl start vss
sudo systemctl enable vss

• Override necessary environment variables in production.cfg (exclude from VCS)

cp config.py production.cfg
nano production.cfg

• Define environment variable VSS_SETTINGS with the FULL path to this cfg file, e.g. VSS_SETTINGS=/path/to/production.cfg

Import ClinVar data

Fetch source data

Soon, you will be able to fetch the latest source data with vss-dat-import

In the meantime, you must parse the ClinVar data yourself using the Macarthur Lab parser. As of 15 Sept 2017, you must use the unmerged pr/33. The clinvar_alleles.single.b37.tsv.gz file is what is required by the importer.

Import parsed ClinVar data

You can then run the importer periodically (by hand or crontab) to notify users of changes to ClinVar classifications:

VSS_SETTINGS=/path/to/production.cfg python -m vss.scripts.import /path/to/clinvar_alleles.single.b37.tsv.gz