singularity_playground

This repo maintains the definition files used for building singularity images remotely. Each branch maintains a different environment but they usually have miniconda with python2 or python3 as base. The idea is to have images that are ready to be used and implemented into workflows e.g. Snakemake or Nextflow. To do this all images circumvent activation of environments via activation scripts or command line functions e.g. conda activate my_env as those types of solutions are not well supported for images that are activated through workflows. Instead all necessary tool dependencies are installed into the base environment of the singularity image.

Building

Before the doing anything make sure singularity is installed and that you have privileges to run singularity as root. To build the image type the following.

git clone --branch branch_name https://github.com/AroArz/singularity_playground 
cd singularity_playground
sudo singularity build branch.sif Singularity.branch.def

Pulling

You can also pull images from singularity hub by entering the following into the command line.

singularity pull shub://AroArz/singularity_playground:branch

Testing

You can test the image after pulling by shelling or running a command. The following shows an example for the Graphlan branch.

Using shell:

singularity shell graphlan.sif
graphlan.py --help

Using run:

singularity run graphlan.sif graphlan.py --help