ob-genomics

Pipelines and ETL for aggregating public genomics data.

Installation

From the ob-genomics directory,

$ pip install -e .

Set up a config file

To configure, make a file config.yml from the template:

$ cp config-template.yml config.yml

Adjust paths to your temp folder, reference data folder, and database.

Usage

Create and populate the database

$ ob-genomics init  # DELETES and creates a new database, then loads gene, tissue, and sample metadata
$ ob-genomics build  # Luigi pipeline for loading large genomics datasets into database

Start up the Shiny app

Start up the Shiny app from the Dockerfile.shiny image, mounting this directory and binding port 80.

$ sh run_shiny.sh

Connect to the database directly via R/dplyr

> source('apps/genomics/database.r')
> kras <- tcga_expr_by_gene('KRAS')