Methylhead pipeline for methyl-seq analysis

Snakemake conversion of the methylhead Nextflow pipeline for targeted methyl-seq analysis.

Overview

This pipeline processes targeted methyl-seq data from raw FASTQs through to:

• Quality control and trimming
• DNA methylation-aware alignment (BWA-meth and BSMAP)
• Methylation calling
• Cell type deconvolution
• DNA methylation scores
• Association testing

Requirements

• Mamba
• Snakemake ≥ 7.0
• Python ≥ 3.8 with packages pandas and yaml
• Apptainer ≥ 1.1.0

Installation

Mamba will be used to install all other requirements, so it must be installed first (install guide)

The remainder can be installed as follows:

mamba config --add channels conda-forge
mamba config --add channels bioconda
mamba config --set channel_priority strict
mamba create -y -n methylhead -c bioconda -c conda-forge snakemake pandas 
mamba activate methylhead
mamba install -c conda-forge -c apptainer

Finally, clone this repository.

git clone <repository-url>

You can verify correct installation by analysing a small example dataset.

Quick Start

1. Prepare inputs

Pipeline inputs include samplesheet.csv, phenotypes.csv, models.csv, and panel.csv. See our guide for preparing these inputs and examples.

2. Configure the pipeline

Create a pipeline configuration file config.yaml by modifying config-example.yml.

The main modifications are in the 'inputs' and 'paths' sections.

3. Run the Pipeline

params=(
	--snakefile methylhead/Snakefile  ## path to pipeline Snakefile
	--configfile config.yml           ## path to pipeline configuration file
	--cores 32                        ## number of processors to use
	--printshellcmds                  ## print rule shell commands 
	--rerun-incomplete                ## rerun rules with incomplete outputs
	--rerun-triggers mtime            ## rerun rules with outputs older than inputs
	--show-failed-logs                ## print log outputs of failed rules
	--verbose
)
snakemake "${params[@]}" 

License

Same as original methylhead pipeline.