postgwas-tools

postgwas-tools is a Python package that provides a collection of tools for post-GWAS (Genome-Wide Association Study) analysis, annotation, and visualization.
It simplifies steps such as plotting, SNP annotation, and result parsing for large-scale GWAS projects.

---

Installation

Option 1 — From source

git clone https://github.com/Dufouranto0/postgwas-tools.git
cd postgwas-tools
python3 -m venv venv
. venv/bin/activate
pip install -e .

Option 2 — From PyPI (planned)

pip install postgwas-tools

---

Package Structure

src/postgwas_tools/
├── annot/
│   ├── fuma/                 # FUMA-related post-GWAS visualization
│   ├── ldsc/                 # LD Score Regression utilities
│   ├── magma/                # MAGMA annotation helpers
│   ├── replication/          # Replication and cross-cohort checks
│   └── utils.py              # Common helper functions
└── mostest/                  # MOSTest-specific result processors

---

Quick Start Example

You can generate a multi-model Manhattan plot directly from the command line:

manhattan_plot -p /path/to/gwas1.sumstats /path/to/gwas2.sumstats \
               -t manhattan \
               -o ./results

Or using a file pattern:

manhattan_plot -p "/path/to/*.sumstats"

This will create a high-resolution manhattan_plot.png in the output directory.

If you just want a light manhattan plot, you can do:

light_manhattan_plot -p /path/to/gwas_input.txt

---

Documentation & Workflows

For detailed usage examples, including QQ plots, locus-based analysis, and replication workflows, see: WORKFLOW.md

---

Dependencies

• Python ≥ 3.9
• pandas, numpy, matplotlib, seaborn, scipy, qmplot, h5py

All dependencies are automatically installed when you run pip install -e ..

---

Citation

If you use postgwas-tools in your work, please cite:

Dufournet A. postgwas-tools: a Python suite for streamlined post-GWAS analysis. CEA NeuroSpin, 2025.

---

Links

• LICENSE
• Homepage
• Issue Tracker