mamana/gwaslab

gwaslab Container

GWAS data analysis and visualization toolkit for processing and visualizing genome-wide association study results.

Tools Included

• gwaslab - Python package for GWAS summary statistics QC, harmonization, and visualization
• pandas - Data manipulation and analysis
• numpy - Numerical computing
• scipy - Scientific computing
• matplotlib - Data visualization

Purpose

This container provides a complete environment for:

• GWAS summary statistics quality control
• Data harmonization and formatting
• Manhattan plots and QQ plots
• Regional association plots
• Lead variant annotation

Usage

Basic Usage

bash
# Run Python interactively
docker run --rm -it -v $(pwd):/data mamana/gwaslab:3.6.13 python3

# Run a Python script
docker run --rm -v $(pwd):/data mamana/gwaslab:3.6.13 python3 /data/analysis.py

Example: Load GWAS Summary Statistics

bash
docker run --rm -v $(pwd):/data mamana/gwaslab:3.6.13 python3 -c "
import gwaslab as gl
import pandas as pd

# Load summary statistics
mysumstats = gl.Sumstats('/data/sumstats.txt',
    snpid='SNP',
    chrom='CHR',
    pos='BP',
    ea='A1',
    nea='A2',
    beta='BETA',
    se='SE',
    p='P')

# Basic QC
mysumstats.basic_check()

# Save harmonized data
mysumstats.data.to_csv('/data/harmonized_sumstats.csv', index=False)
"

Example: Create Manhattan Plot

bash
docker run --rm -v $(pwd):/data mamana/gwaslab:3.6.13 python3 -c "
import gwaslab as gl

mysumstats = gl.Sumstats('/data/sumstats.txt',
    snpid='SNP', chrom='CHR', pos='BP',
    ea='A1', nea='A2', beta='BETA', se='SE', p='P')

# Generate Manhattan plot
mysumstats.plot_mqq(save='/data/manhattan_qq.png')
"

Directory Structure

Container Specifications

• Base Image: python:3.10-slim
• Architecture: linux/amd64
• Working Directory: /data

Building Locally

bash
# Build
docker build -t test-gwaslab:local ./containers/gwaslab

# Test
docker run --rm test-gwaslab:local python -c "import gwaslab; print(gwaslab.__version__)"

# Interactive shell
docker run --rm -it test-gwaslab:local /bin/bash