Skip to content

Available Apps

On this Page

  • Which portal to use, and why there are two
  • Every application you can launch, what it is for, and what it needs
  • The applications that need a licence before they will start
  • What to do when the software you need has no app

These are the applications you can launch from the Interactive Apps menu in Open OnDemand. Each runs as a Slurm job on a compute node: you fill in a short form for cores, memory, wall time and any options specific to that app, press Launch, then click Connect once the job starts.

Because it is a normal Slurm job it has to queue, and it holds the resources you asked for until you delete the session or the wall time runs out. Open OnDemand Overview covers how that works and the limits that apply to OnDemand jobs.

Which Portal?

There are two portals, and you log in to both the same way.

The two OnDemand portals
Portal Address What is on it
Current ondemand.otago.ac.nz Open OnDemand 4.1, and most applications. Start here.
Legacy ondemand-legacy.otago.ac.nz Applications that have not yet moved across.

The legacy instance exists only because a set of applications had not been ported when the portal was upgraded in April 2026. It is shrinking over time, so check the current portal first.

Apps on the Current Portal

Available from Interactive Apps at ondemand.otago.ac.nz
App What it is for Notes
Otago HPC Desktop A full Linux desktop on a compute node, for anything without its own app
RStudio Server R and RStudio in the browser
JupyterLab Notebooks for Python, R and Julia Several environments
WhisperX UI Speech-to-text transcription and translation Faster with a GPU
AFNI Analysis and display of functional MRI data
Blender 3D modelling, animation and rendering Needs a GPU
DeepLabCut Markerless pose estimation from video Faster with a GPU
ESA SNAP Earth observation imagery, including Sentinel data
Fiji Scientific image analysis — an ImageJ distribution with plugins bundled
FlexPDE Finite element solver for partial differential equations Licence
Kilosort Spike sorting for multi-channel electrophysiology recordings Faster with a GPU
MATLAB Numerical computing and simulation Licensed
NetLogo Agent-based modelling of natural and social phenomena
Phenix Macromolecular structure determination from X-ray and cryo-EM data
phy Manual curation of spike sorting results
VSCodium Code editor in the browser — a freely-licensed build of VS Code
XDSGUI Processing and phasing X-ray, neutron and electron diffraction data Also on legacy

Apps on the Legacy Portal

Available from Interactive Apps at ondemand-legacy.otago.ac.nz
App What it is for Notes
CCP4 Macromolecular structure determination by X-ray crystallography
ChimeraX Molecular structure visualisation Needs a GPU
CLC Genomics Workbench Sequence analysis for genomics, epigenomics and metagenomics Licence
Connectome Workbench Visualising Human Connectome Project data
EcoAssist Classifying camera trap images Faster with a GPU
EPI2ME Desktop Oxford Nanopore's GUI for Nextflow bioinformatics pipelines Setup notes
FSL Analysis of FMRI, MRI and diffusion brain imaging data
GLOBEClaritas 2D and 3D land and marine seismic data processing Licensed
GRASS GIS Geospatial analysis, modelling and visualisation
ilastik Interactive image classification and segmentation using machine learning
MELTS Thermodynamic modelling of phase equilibria in magmatic systems Model versions
Open WebUI – Ollama Browser front end for running large language models locally Faster with a GPU
QGIS Desktop geographic information system
RELION Cryo-electron microscopy structure determination Faster with a GPU
SAGA GIS Geospatial analysis, with a large library of terrain and raster tools
SaTScan Spatial, temporal and space-time cluster detection
Specify Biological collections management
Stata Statistics, data management and econometrics Licence
UGENE Visualising, aligning, assembling and annotating DNA and protein sequences Launch options
XDSGUI Processing and phasing X-ray, neutron and electron diffraction data Also on the current portal

Apps That Need More Than a Launch Form

Most apps need nothing beyond filling in the form. These are the exceptions.

JupyterLab

JupyterLab comes in a number of domain-focused and application-specific variants, selected on the launch form. Each is a container, so the software stack is immutable — which makes your environment consistent from one session to the next and easier to reproduce later.

The environments are based on the Jupyter Docker Stacks images maintained by the Jupyter team. If you need a customised environment, email the eResearch Support team at rtis.support@otago.ac.nz.

FlexPDE

FlexPDE Lite is the evaluation configuration. It is free to use, but limits the number of simultaneous equations and mesh cells.

FlexPDE Professional can be activated over the internet with an appropriate serial number. Activation is machine-based, so unless you select the same node each time you will need to deactivate and reactivate on subsequent runs.

CLC Genomics Workbench

QIAGEN CLC Genomics Workbench is a sequence analysis platform for genomics, epigenomics and metagenomics.

Licensed software

The cluster has a small number of floating licences available for trial purposes, facilitated by Dr Sunali Mehta (sunali.mehta@otago.ac.nz) in the Pathology department. Please make sure you are authorised before consuming a licence.

Tick the 3D hardware-accelerated rendering option and request a GPU if you are using the 3D viewers.

MELTS

MELTS models thermodynamic phase equilibria in magmatic systems (melts.ofm-research.org). The launcher lets you pick the version and model:

  • rhyolite-MELTS 1.0.2 — original version with corrections. Old H2O model, no mixed fluids.
  • rhyolite-MELTS 1.1.0 — mixed fluid version that preserves the ternary minimum. Old H2O model.
  • rhyolite-MELTS 1.2.0 — mixed fluid version, best for mafic and alkalic melts. New H2O model.
  • pMELTS 5.6.1 — original version with corrections. Old H2O model, no mixed fluids.

UGENE

UGENE integrates dozens of well-known biological tools and algorithms for genomics, evolutionary biology and virology. Two launch options are worth knowing about:

  • Hardware-accelerated 3D improves the 3D viewer.
  • OpenCL improves the performance of a few algorithms, including Smith-Waterman and the UGENE Genome Aligner.

EPI2ME Desktop

Oxford Nanopore's EPI2ME Desktop gives you a graphical interface for running bioinformatics pipelines.

Importing other workflows

As well as the ONT workflows prepopulated in the Available Workflows tab, EPI2ME Desktop can import generic Nextflow workflows, including the 100+ curated pipelines of nf-core. Under Workflows, click Import workflow and paste the workflow's git repository URL — for example https://github.com/nf-core/<wf>.

How it behaves on the cluster:

  • Individual pipeline tasks are sent to Slurm and scheduled as separate jobs, with their own resources — taken from the workflow defaults, or from what you set in the Nextflow configuration tab.
  • For pipelines needing GPU compute there is no need to run EPI2ME itself on a GPU. The launch form does not offer the option; tasks that need a GPU are scheduled onto GPU-capable nodes automatically.
  • There is no need to change the Profile setting under Nextflow configuration. This instance defaults to the singularity profile, which uses Apptainer.

Launching the app for the first time creates a global Nextflow configuration at ~/.nextflow/config to make that work. If the file already exists, you may need to add the following yourself:

Terminal

process {
  executor = 'slurm'
  time = 6.h
  withLabel: 'gpu' {
    queue = 'aoraki_gpu'
  }
}

Warning

EPI2ME Desktop is not designed with HPC clusters in mind, so integrating it with the scheduler has taken a number of workarounds. Testing has been promising, but the added complexity — and the application's limited Nextflow configurability — may produce problems that are hard to troubleshoot.

Consider running the workflows from the command line with Nextflow instead.

Stata

Stata will not start until a valid licence file is installed for your account. That needs authorisation details from the University's software procurement office.

Requesting a licence

  1. Go to https://www.otago.ac.nz/its/services/software/stata.
  2. Click the Software Order Form link.
  3. Enter Stata in the Product field.
  4. ITS Software Procurement will email you a PDF containing your serial number, code and authorisation.
  5. Forward that PDF to the eResearch Support team at rtis.support@otago.ac.nz.

We then generate the licence file, install it for your account, and confirm when Stata is ready to use.

If you already hold a University Stata licence you can reuse it for the cluster — just forward the PDF. Licences are managed centrally to comply with the University's licensing terms.

Licensed Applications

Some applications need licensing sorted out before they will run. If you are planning work around one of these, start the licence process early:

Applications with licensing requirements
App What is needed
Stata A licence file generated for your account — see Stata above
CLC Genomics Workbench Authorisation to use one of a small number of floating licences
FlexPDE Professional A serial number, activated per node. FlexPDE Lite needs nothing, but is limited
MATLAB Covered by the University licence
GLOBEClaritas Covered by the University licence

If you are unsure whether your licence covers cluster use, ask the eResearch Support team at rtis.support@otago.ac.nz before you start.

If the App You Need Is Not Here

Three things to try, in order:

  1. Run it on the HPC Desktop. Plenty of graphical software runs happily in a desktop session without needing an app of its own.
  2. Load it as a module. Some of these applications are also available on the command line, including AFNI, FSL, Connectome Workbench, GLOBEClaritas, MATLAB, EPI2ME and R. Run module spider <name> to check, or see Software Applications for the full list.
  3. Ask us to install it. See How Do I Ask for Software to Be Installed? for what makes that request easy to act on.

What's next?