This page covers information on how to use the 'Pawsey Bio - Ubuntu 22.04 - 2023-XX' image for Nimbus. Instructions on how to choose this image when creating your instance can be found here. This Bio-image is created to cater to bioinformatics users who prefer to have their instances pre-installed with software, tools and datasets commonly used in the bioinformatics domain, including over 8000 Biocontainer tools.
Commonly used software
|Ansible||An automation platform that Pawsey uses to automate a number of software deployment|
|CernVM-FS||A read-only file system for accessing files on shared repositories||See 1. Biocontainers and Reference Genome data|
|A popular container engine|
|Google Chrome||A web browser||Make sure to SSH in to your instance with X11 forwarding, i.e. |
|Jupyter Notebook||For using an interactive Jupyter Notebook||See Run Jupyter Notebook Interactively|
|Lmod||A modules environment that we use at Pawsey for loading sotware|
|Nextflow||A popular workflow manager - "Nextflow enables scalable and reproducible scientific workflows using software containers. It allows the adaptation of pipelines written in the most common scripting languages."||Nextflow can leverage the containers from /cvmfs/singularity.galaxyproject.org or container modules - See 4. Using Nextflow|
|Pip||A Python package installer|
|Python3||A popular programming language used in many Bioinformatics software|
|For using R interactively||See Run RStudio Interactively|
|Singularity||A popular container engine that can be used on HPC|
|Singularity-HPC||A container modules installer||See 2. Using Biocontainers|
|Spack||A package management tool|
A virtual desktop application
|X2go server has been pre-installed on the image. To use it, you will need to install X2go client on your local machine - see Installing X2go Client|
On this page, we will only cover instructions for how to use access and use Biocontainers and reference genome data sets. For instructions for the software listed above, please see the software's original documentation page.
1. Biocontainers and Reference Genome data
CernVM-FS is a read-only file system that was developed by another supercomputing centre (Cern). It allows files such as container tools, reference datasets and other shared resources that are commonly used by many researchers to be accessed, added to, and updated in the one location. At Pawsey, we currently cache the Biocontainer tools and reference genome datasets that are on Galaxy Project's repositories. The list of Biocontainer tools available can be searched on https://biocontainers.pro/registry.
To use the Biocontainer tools, you can skip this step and proceed to the next section, 2. Using the Biocontainer tools.
To view these repositories, you can do the following:
List the Biocontainer tools repository:Note: It may take a minute or two to load the folders. When you have done it once, it will not take as long to show again.The
/cvmfs/singularity.galaxyproject.org/allsubdirectory is where the entire list of 8000+ Biocontainers can be found, with the alphabetical subdirectories being symlinks to them.
List the reference genome sets and other data files:
Please note that the data sets may not be comprehensive, and this service is not meant to replace your current methods for accessing public datasets.
To use these data files for your analyses, copy the absolute file path in your workflow/pipeline. For example, with the reference genome
Hg38, the file can be found in the following location, and specifically under the
So the full absolute path for the
Hg38sequence file would be:
If you run into any errors with accessing the file system, run the following to re-install it:
If it is still causing errors, you may need to reboot your instance.
2. Using the Biocontainer tools
Singularity-HPC (SHPC) is a software for container modules. In this Pawsey Bio - Ubuntu 22.04 - 2023-XX image, we have integrated the use of SHPC seamlessly with CernVM-FS. This means that you can easily access and use over 8000 Biocontainers (and up to the latest versions) without needing to understand container syntax.
If you are using the now deprecated 'Pawsey Bio - Ubuntu 20.04 - 2021-11' image, you will not be able to seamlessly use Biocontainer tools without first installing them using SHPC in the next section, 3. Adding a local SHPC registry. To avoid that, we recommend that you recreate a new instance with the 'Pawsey Bio - Ubuntu 22.04 - 2023-XX' image.
When using Biocontainer tools, you will be required to export the paths for your data directory(ies) to Singularity, so that they can be readable by the container. For example, if your data directory is /data, then you would run the following to add it to the Singularity bind path:
To search for versions and information on a particular tool, e.g. for cuttlefish, use
To check for availability and to load the tool, use the
module loadcommands:Note that on first use, the tool might take a 30 seconds or so to run the command as the container is being accessed from the filesystem for the first time
To check for the list of modules loaded:This list will be cleared whenever you log out of your instance. After logging back in, you will need to reload the module for it to be on the list for use.
To install another version not available as a module:
To use another version of the tool available from the above
shpc showlist, use shpc to install the module, ensuring to use and keep the cvmfs path of the container:
Update the modules cache, then, run the same module commands to load the tool in this version:
You will notice that the previous version of the tool is now swapped out for the version you just loaded
If you prefer to use the biocontainers without SHPC, you can do so by using the absolute path for each of the biocontainers. Note that you would require knowledge on how to use Singularity to do so. The version of Singularity installed on the Nimbus Bio image is 3.8.7 and instructions can be found here: Singularity exec.
For example, to use
cuttlefish version 2.2.0–hf1761c0_0:
When using Singularity, if you run into an issue with
no loop devices found, please use the solution provided here: Using Containers#Commonissues
3. Adding a local SHPC registry
If there are versions (usually older ones) of a Biocontainer tool that is present in the cvmfs repository but not on the
shpc show list (i.e. the default recipe), you can create a local SHPC registry and add/update a recipe file for the Biocontainer tool.
Clone the remote SHPC-registry and add it as a local registry:
Look up for available versions of the tool, e.g. cuttlefish:
To add a different version to the recipe file for the tool, e.g.
To install the tool:
If you are using the now deprecated 'Pawsey Bio - Ubuntu 20.04 - 2021-11' image, run the following steps first:
By setting your
module_baseto this new location, all new container modules will be installed to this path.
Now when you do a module avail, the newly installed
2.0.0--hf1761c0_1version will be available:
Since you have created your own local registry, shpc will default to your local registry whenever you do a look up with
shpc show. To look up the full list of Biocontainer tools with the latest versions, you will need to add a flag to point to the remote (Github) shpc-registry in your search:The shpc-registry is kept up-to-date with the latest versions of all Biocontainers on a nightly update.
4. Using Nextflow
Nextflow makes use of containers to run your workflows sequentially. Each step of your workflow is called a process. For each process, Nextflow pulls the appropriate container required for that step to run it. You can prevent Nextflow from pulling the container and using what is present on your instance to save time and space.
To do so, you would create an additional config file to point Nextflow to either 1) the paths of the containers on /cvmfs/singularity.galaxyproject.org, or 2) the module paths on your instance. Nextflow prioritises this custom config file above the default
nextflow.config file(s), if present, in other directories for your workflow.
Nextflow pipelines for Bioinformatics
Nextflow has a repository of pipelines that are available through https://nf-co.re. These are becoming increasingly popular, as more peer-reviewed pipelines are added by the community. A couple of popular ones include:
Configuring Nextflow to use Biocontainers
Please note this is only an example of how you can configure your Nextflow workflow to use the containers available from your instance.
Suppose you are using the
nfcore/rnaseqpipeline. Note that in the
FASTQCprocess, there are a few parameters for the tool. These will be over-written by the config file that you will create in the next step.
Nextflow prioritises a custom config file over any other config files or values defined in the workflow files. To ensure that Nextflow uses the existing container for
fastqc, you would create and use a custom config file, choosing either of the two ways:
cvmfs_path.configfile with the following:
Then run the workflow with this config:
modules_path.configfile with the following:
Then run the workflow with this config:
*_path.configfile to ensure that every process has the path for the existing container to run that part of the workflow from. More info on config files for processes can be found here: https://www.nextflow.io/docs/latest/config.html#scope-process
- No labels