Connecting plant scientists with their data: Using Zegami, a Software-as-a-Service solution for the visualisation of time-course image data collected at the Australian Plant Phenomics Facility.

Mr George Sainsbury1

1Australian Plant Phenomics Facility, Adelaide, Australia

At the Australian Plant Phenomics Facility (APPF), we capture tens of thousands of images of individual plants a day. Analysis of these images is automated, and the general wisdom is that “if you take 10,000 photos a day, they’re not for looking at.” However, there are many reasons why researchers and operational staff might want to look at them; from quality control to determination of growth stages.

We partnered with Zegami; a company formed by Oxford University Innovation after humble beginnings in South Australia. The software-as-a-service (SaaS) they offer was originally inspired by Microsoft PivotViewer, but has expanded significantly, with major user experience improvements and Zegami is now integrating deep learning to help discover more about the images in the collections.

We developed an Export-Transform-Load (ETL) tool from the LemnaTec system at The Plant Accelerator (TPA) to Zegami’s service, hosted on Microsoft Azure. Data ingest and processing is fully automated, running in Docker on Pawsey’s Nimbus cloud. Images and data are updated daily. Our collections are some of the largest on Zegami and as such we have collaborated on service and API improvements.

Zegami allows researchers conducting experiments at TPA to view and share their experiments from the comfort of their own workspaces. Many experiments are publicly available and Zegami is one piece of a larger puzzle to make data at APPF more FAIR.


Biography:

George Sainsbury is the Data Architect & Software Engineer at The Plant Accelerator (TPA), the lead node of the Australian Plant Phenomics Facility (APPF). George is responsible for “all things IT” at TPA and his role encompasses software development, research data management, systems administration and everything in between. 2020 will be George’s fifth year at eResearch Australasia.

Jupyter on NeSI HPC – a product management approach to deliver interactive computing

Mr Thomas Berger1,2, Mr Chris Scott1,2, Mr Yuriy Halytskyy1,2, Mr Blair Bethwaite1,2

1New Zealand eScience Infrastructure – NeSI, Auckland, New Zealand
2University of Auckland, Auckland, New Zealand

NeSI is using product management techniques to guide delivery of a new interactive computing and analysis service leveraging JupyterHub.

Previously researchers were enabled to set up Slurm interactive sessions and a combination of port forwarding and SSH tunnelling via CLI, though this is complex for users and difficult to support, so we had seen very little uptake. A hub interface allowing the users to authenticate and request computing resources via the web, dealing with proxying and port forwarding in the background, is taking away complexity.

Product management gives us a framework to discover and prioritise user needs orthogonal to the technology concerns that often come to the fore (to the detriment of user experience) in eResearch and HPC solutioneering. In this talk we’ll discuss the context, describe our initial architecture and requirements, and share how product management has helped us validate and prioritise.


Biography:

Thomas Berger is Product Manager with New Zealand eScience Infrastructure (NeSI) hosted by the University of Auckland.

https://orcid.org/0000-0003-0747-0366

Unlocking access to the National Virtual Core Library with nvcl_kit

Mr Vincent Fazio1

1C.S.I.R.O., Clayton, Australia

Australia’s National Virtual Core Library (NVCL) is an online library containing mineralogy data and imagery from than one million metres of drill core rock samples from all around Australia.

Up until now this data was only available for downloading from government geologic survey websites and the AuScope Portal.

‘nvcl_kit’ is a  Python library which makes this huge dataset available to geoscience practitioners via Python’s more convenient and powerful interface.  Tools such as Jupyter notebooks can be employed to their maximum advantage to develop, demonstrate and share useful and hopefully insightful code documents.


Biography:

Vincent Fazio is a Senior Software Engineer with the Informatics Platforms team within the Minerals Resources business unit at CSIRO.

Recent Comments

    Categories

    About the conference

    eResearch Australasia provides opportunities for delegates to engage, connect, and share their ideas and exemplars concerning new information centric research capabilities, and how information and communication technologies help researchers to collaborate, collect, manage, share, process, analyse, store, find, understand and re-use information.

    Conference Managers

    Please contact the team at Conference Design with any questions regarding the conference.

    © 2018 - 2020 Conference Design Pty Ltd