Linear GWAS Package

This package provides a simple tool for performing Genome-Wide Association Studies (GWAS) on continuous phenotypes using linear regression.

Installation

To install the package, access via GitHub repository link.

git clone https://github.com/chloekeggen/gwas-project.git

It is recommended to download Anaconda and run package within Anaconda terminal. Also, in order to ensure dependencies and paths are maintained correctly, it is recommended to create a new Anaconda environment. Then, continue with the following steps.

cd gwas-project
pip install -r requirements.txt
python setup.py install

Usage

After installing the package, you can use gwas-tools-cli.py to perform GWAS on your data.

gwas-tools-cli --vcf <path_to_vcf_file> --pheno <path_to_phenotype_file> --out <output_file_prefix>

Replace <path_to_vcf_file> with the path to your VCF file containing genotype data, <path_to_phenotype_file> with the path to your phenotype file, and <output_file_prefix> with the desired name for the output files.

Optional arguments

--maf <maf_threshold>
--h OR --help

Adjust MAF threshold for filtering SNPs as needed; the default is 0.05. Use --help for a list of valid arguments that can be used

Output

• <output_file_prefix>_results.csv: CSV file containing the results of the linear regression analysis.
• <output_file_prefix>_manhattan_plot.png: Manhattan plot visualizing the results of the GWAS analysis.
• <output_file_prefix>_QQ_plot.png: QQ plot visualizing the results of the GWAS analysis.

Example using given smaller phenotype and genotype files

gwas-tools-cli --vcf subset_lab3_gwas_CHR_18_19_20.vcf.gz --pheno subset_lab3_gwas_CHR_18_19_20.phen --out gwas_results

This command will perform GWAS on the subsections of genotype and phenotype data files from Lab 3 (ie: data from chromosomes 18, 19, 20), and save the results into 3 gwas_results files.

Dependencies

• pandas
• numpy
• pyvcf3
• statsmodels
• matplotlib