Prerequisites (Non-NCSU Attendees)

Author

Affiliation

Mery Touceda-Suárez

Bioinformatics Research Center

This page outlines what you should set up and review before the Nextflow workshop begins if you do not have an NCSU Hazel HPC account. You’ll need access to your own HPC or compute environment with Nextflow installed and the workshop’s tools available, either as pre-installed software/modules or as containers.

1 Install Nextflow on Your HPC

• Nextflow requires Java 17–21. Check whether Java is already available:

$ java -version

If it isn’t, load or install it (e.g. module avail java on many HPC systems) before installing Nextflow.

• Check whether Nextflow is already available as an environment module on your HPC:

$ module avail nextflow

If it is, you can simply module load nextflow and skip the manual install below.

• Install Nextflow by following the official installation instructions. The simplest approach is:

$ curl -s https://get.nextflow.io | bash

This creates a nextflow executable in your current directory. Move it into a directory on your $PATH (e.g. ~/bin) so it can be run from anywhere:

$ mkdir -p ~/bin
$ mv nextflow ~/bin/
$ export PATH="$PATH:$HOME/bin"

Add the export PATH line to your ~/.bashrc (or equivalent) so it persists across logins.

• Confirm Nextflow runs:

$ nextflow -version

2 Access to the Workshop’s Tools

The workshop’s exercises run a small number of common bioinformatics tools (e.g. FastQC and Trimmomatic) inside Nextflow processes. On Hazel, these are provided by the BRC as pre-built modules/containers (see Loading BRC Modules) — outside NCSU, you’ll need to provide equivalent access on your own system, using one of the following options.

2.1 Option A: Pre-installed software or environment modules

• Check whether your HPC already provides these tools as environment modules or pre-installed software:

$ module avail fastqc
$ module avail trimmomatic

If they’re available, note the module load command (or install path) for each — you’ll substitute it for the BRC module directives used in the workshop’s example pipelines.

2.2 Option B: Containers via Singularity/Apptainer and BioContainers

If the tools aren’t available as modules, you can run them from containers instead. BioContainers publishes ready-made container images for most common bioinformatics tools, built directly from their Bioconda packages.

• Check that a container engine is available on your HPC:

$ singularity --version
# or
$ apptainer --version

(Apptainer is the actively maintained successor to Singularity; the commands are interchangeable for our purposes.) If neither is installed, ask your HPC administrators — most institutional HPC systems provide one or the other as a module.

• Find the tool you need on BioContainers. Search for it at biocontainers.pro or browse its images directly on quay.io/organization/biocontainers, and note the exact image tag (tool version + build hash), e.g. fastqc:0.12.1--hdfd78af_0.

• Pull the container image with singularity pull (or apptainer pull):

$ singularity pull fastqc.sif docker://quay.io/biocontainers/fastqc:0.12.1--hdfd78af_0
$ singularity pull trimmomatic.sif docker://quay.io/biocontainers/trimmomatic:0.39--hdfd78af_1

Note

The workshop’s example pipeline scripts request tools with a module directive, e.g. module 'fastqc/0.12.1', which points Nextflow at a Hazel/BRC module. Outside Hazel, replace this with a container directive pointing at your .sif file (or the docker:// image URI directly), and enable Singularity in your nextflow.config:

singularity.enabled = true

process {
    withName: fastqc {
        container = '/path/to/fastqc.sif'
    }
    withName: trim {
        container = '/path/to/trimmomatic.sif'
    }
}

Tip

You don’t have to pre-pull images by hand — if singularity.enabled = true and a process’s container directive points at a docker:// URI, Nextflow will pull and cache the image automatically the first time that process runs.