How Do I Snakemake?¶
A guide to getting up and running with Snakemake.
Snakemake is a powerful tool for building complex workflows in Python. It is similar to GNU make, from whence it derives its name.
However, Snakemake is more powerful than make, and there are some important
differences (that are also reasons for make users to migrate to Snakemake):
-
Snakemake is implemented in Python, so Snakemake is able to bootstrap on the capabilities of Python. Snakefiles are also Python programs, so Python modules can be used in workflow definitions.
-
Snakemake integrates seamlessly with conda, and a conda environment file can be specified for a given rule.
-
Snakemake can be used with Singularity (containerization technology for running Docker containers on high performance computing (HPC) clusters) or with Kubernetes (technology for running multiple Docker containers across multiple machines).
-
Snakemake integrates with HPC batch and queue systems, and can be used to run workflows using HPC or batch systems.
Links¶
How Do I Snakemake on pages.charlesreid1.com: https://pages.charlesreid1.com/how-do-i-snakemake
Source code for tutorial on git.charlesreid1.com: https://git.charlesreid1.com/charlesreid1/how-do-i-snakemake
Installing¶
Installing Snakemake - installation instructions for Snakemake using pip or conda.
Terminology¶
Snakemake Terminology - a glossary of Snakemake terms used throughout this documentation and what they mean.
Converting Workflows to Snakemake¶
Converting Workflows to Snakemake - strategies for converting shell script workflows into Snakemake workflows.
Example Overview: Filtering Sequence Reads
Stage 1: Shell Script + Snakefile
Stage 2: Replace Script with Snakefile (Hard-Coded)
Stage 3: Replace Script with Snakefile (Wildcards)
Useful Resources¶
Following is a list of useful Snakemake resources: