Snakemake Local Example

This repository provides a very small example of running a Snakemake workflow locally that calls Python and R code to generate a simple visualization.

The workflow:

• Python: generate a mock gene expression count table (TSV).
• R + ggpubr: visualize the mock counts as a boxplot with a p-value.

Table of Contents

• Project Background
• Install & Setup
• Usage
• Directory Structure

Project Background

This example is intended for teaching and local experimentation. It is kept intentionally simple so that you can focus on the Snakemake concepts:

• how rules declare input and output files,
• how Snakemake tracks dependencies, and
• how to integrate small Python and R scripts.

Install & Setup

You need:

• Python (3.9+ recommended),
• R (any modern version),
• Snakemake installed in some environment, and
• the R package ggpubr (plus its dependencies such as ggplot2) for the visualization rule.

Create the mamba/conda environment

This repo includes an environment.yml that installs Snakemake, base R, and r-ggpubr for visualization:

mamba env create -f environment.yml
mamba activate snakemake-example

If you use conda instead of mamba, replace mamba with conda.

If you do not already have mamba installed, see the Mamba installation guide, then return to the Install & Setup section here.

Usage

From the top level of this repository:

snakemake -j 1

Snakemake will:

1. Use Python to generate a mock counts table in results/mock_counts.tsv.
2. Use R and ggpubr to generate a mock counts boxplot in results/mock_counts_boxplot.png.

You can re-run the command and Snakemake will only recompute steps whose inputs have changed.

Directory Structure

$ tree -a snakemake-example/
snakemake-example/
├── README.md
├── Snakefile
├── environment.yml
├── requirements.txt
├── results
│   ├── mock_counts.tsv            # created by Snakemake (Python step)
│   └── mock_counts_boxplot.png    # created by Snakemake (R + ggpubr step)
└── scripts
    ├── generate_mock_counts.py
    └── visualize_counts.R

The files under results/ are created by Snakemake and can be deleted and regenerated at any time.