Snakemake workflow for GWAS in aspen

The goal of this workflow is to provide a flexible interface to performing GWAS in Populus tremula.

Requirements

• snakemake
• conda (miniconda)
• cookiecutter (optional)

The recommended approach would be to create a snakemake conda environment that is used for running the workflow.

conda create -n snakemake snakemake
conda activate snakemake

Usage

Configuration

Configure the workflow by editing the file config.yaml and the traits.tsv file.

Execution

Test the configuration by performing a dry-run

snakemake --use-conda --dry-run

You can execute the workflow locally using $N cores with

snakemake --use-conda --cores $N

Cluster environments

It is also possible to execute the workflow in a cluster environment, and the easiest approach is probably by using an existing snakemake profile. Profiles can be generated through snakemake by modifying the cluster entry in config.yaml and running

snakemake --use-conda cluster_config

This will generate a profile with the desired name in $HOME/.config/snakemake. The profile can be utilised by running snakemake with the --profile flag, specifing the generated profile

snakemake --profile <profile_name>

For more information on how to configure the cluster parameters, take a look at the snakemake documentation.

In the cluster/slurm.yaml you can find a suggested job configuration for SLURM. This together with a profile means you can run the workflow like

snakemake --profile <profile_name> --cluster-config cluster/slurm.yaml

Test data

There is a small test dataset included under test/data. The traits are in this case just random numbers sampled from a standard normal distribution. Running the workflow without modifying config.yaml will use this data.