negative-control-genomicc

Motivation

This is a repository used to generate a biological negative control for the GenOMICC cohort. It takes as input BGEN files for a given cohort, an input list of binding QTLs (bQTLs) and trans-actors (ex: expression QTLs; eQTLs), as well as regions of the genome to exclude, and pick a randomly-sampled MAF-matched SNPs for the input SNPs you specify (bQTLs + eQTLs).

The idea behind this is to test the biological hypothesis that epistasis through altered binding of a transcription factor (TF) and a trans-acting QTL which regulates binding at a different level (for example, by expression of that TF) drives complex trait variation in a genomic cohort.

Requirements

You must have nextflow and singularity installed. This can be achieved through conda and the env.yaml file provided in this repository.

Conda

You can install conda by the following the directions here: https://conda.io/projects/conda/en/latest/user-guide/install/index.html

Example:

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh

Nextflow & Singularity

Once installed, you can create the nextflow environment by:

conda env create --file env.yaml

Setup

You must provide the following information in a run-specific configuration file (example: genomicc.config)

OUTDIR : Path to output directory

BQTLS : Path to CSV containing the bQTLs you would like to find negative control SNPs for

EQTLS : Path to CSV containing the eQTLs you would like to find negative control SNPs for

ASSEMBLY : Assembly name (either hg19, grch37, hg38 or grch38)

EXCLUSION_REGIONS : Path to CSV file containing the LD blocks for lead SNPs, which you would like not to sample from

CONFIG_FILE : Config file for targene test run

ESTIMANDS_FILE : Estimands file for targene test run

BGEN_FILES : /path/to/BGEN/cohort_chr{1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}.{bgen,bgen.bgi,sample}"

Run

To run the pipeline, start a screen session, activate your nextflow environment and then run with the profile for your platform (in my case, Eddie):

screen
conda activate nextflow
nextflow run main.nf -c your_config.config -profile eddie -resume

Output

Results will be stores in the OUTDIR specified in your configuration above.