Skip to content

4_modules_and_installing

Mark Ravinet edited this page Sep 12, 2025 · 3 revisions

Modules, programs and installations

Introduction

Now that you know how to work on the cluster and manage your data, you will need to use different software during your analyses. There are two ways to currently do this. The first is to use a set of modules installed on the cluster. Alternatively, you can install programs yourself. Finally, you can use templates or pre-made environments to install a suite of tools that a pipeline might need to run.

Modules on the cluster

Preinstalled programs on Saga can be easily accessed using the module command. This is very similar to loading packages in R - the main difference here however is that you cannot install them yourself.

Once you have logged in, you can check the modules you have loaded like so:

module list

If you want to see all the modules available to you, you can use the following:

module avail

You can also use this for specific information on certain packages - i.e:

module avail bcftools
# a similar option
module spider bcftools

One thing to note with this example is that there are multiple options available. Often there are multiple versions of a software installed. It is important to make sure you use the one you intend to!

Let’s say we want to load bcftools, we would use the following:

module load BCFtools/1.17-GCC-12.2.0

You will need to do this everytime you log in, as that automatically purges all loaded modules. To unload a specific module:

module unload BCFtools/1.17-GCC-12.2.0

If you want to get rid of everything loaded:

module purge

For more information on modules, see here

Installing things yourself

Often you will need to install things yourself. The easiest way to do this is to use miniconda - i.e. a stripped down version of the full conda package manager.

As of August 2024, we are moving towards a model of using a shared conda installation, managed via the modules pre-installed on the Saga infrastructure. For more details on this, see this guide.

If you would like to use a custom conda installation, please ask me first unless I have discussed it with you already.

Clone this wiki locally