25 - 26 May 2017
Plant & Food Research - Lincoln,
New Zealand

Preliminary Programme

The final programme is now available, click here to download it.

Day 1 : Presentations (start 8.50am)

The first day will comprise chaired presentations with a concurrent poster session. There will be a session of “lightning poster presentations" with an award given for the best poster.

We have invited a number of keynote speakers who are renowned experts in these areas. They will not only present their work but also facilitate practical workshops. Keynote speakers include:

Session 1 Exploration:

Shannon Schlueter (The University of North Carolina at Charlotte, USA; Leader of the UNCC Data Science Initiative; “Universal data science challenges: Perspective from disparate data domains”):

Data science is largely a domain agnostic discipline. Best practices and novel solutions to once intractable challenges are routinely gleaned from seemingly disparate data domains. Methodologies once rarely identified outside of the realm of bioinformatics and computational biology are now commonly used to address challenges in fields such as image recognition and social sentiment analysis. But is this synergy mutually beneficial? Can the life sciences and bioinformatics claim benefit from practices found in other data rich domains?

Session 2 Application:

Charles Unsworth (Auckland University; “Machine learning for neural applications”)

In the first part of the seminar, I will discuss how we have applied ANNs for the detection of volatile organic compounds (VOCs) from the in vivo firings of insect olfactory neurons. In the second part of the seminar, I will discuss how we have applied Machine Learning approaches to identify Sharp wave transients in the EEG in order to aid the early detection of Hypoxic Ischemia in Newborn Infants.

Richard Newcomb (Chief Scientist, Plant & Food Research; “Data to do with Smell”).

Smell is one of the most complex of our senses. Thousands of odours, hundreds of receptor genes, a few sensory perceptions and all with just one nose. In this talk I outline how we have used analysis and visualisation of data from genomes and odour space to unravel how odours are perceived in insects and humans

Session 3  Big Data in the Science Industry

Tatiana Lomasko (CSST; “Big Space Data: What you do with it?”)

The focus of this seminar is presenting the opportunities rasising from the establshment of the NZ Centre of Space Science Technology (CSST). New Zealand needs to capitalise on the rapidly developing global opportunity provided by satellites, both for Earth observation and for data telemetry. This opportunity is relevant for many sectors of New Zealand's economy and society including primary production, energy and water management, national security, urban planning and hazard management. I will present the CSST’s plans to profit from extracting the full value available from existing space-based measurements that are often made freely available by international space agencies as well as from developing a national capability to own and operate our own satellites to enhance our country's wealth and well-being.

There is exponential growth in the volume of satellite data, primarily imagery, being made available. This "Earth Observation" market was estimated at NZ$2.7 billion in 2015 (by North-ern Sky Research) and is predicted to double by 2024. The market segment comprises data, data processing, and information products, with value-added information products undergoing the most rapid growth and having the greatest future potential.

Finlay Thompson (DragonFly; "Data science practices in the real world”)

While doing research in a reproducible way is recognised as desirable goal, this ambition can sometimes be neglected when confronted with the heat of tight deadlines and busy research programmes. At Dragonfly we have developed/discovered a set of practices that ensures our work is always reproducible. I will talk about what those principles are, and how committing to reproducible research has increased the reliability of our work, and the productivity of our team.

Bruno Lago (Catalyst): Cloud Computing for Data Scientists

Cloud Computing is a disruptive technology, but only when done right. In this presentation Bruno will explain how data scientists and researchers can benefit from cloud computing, describe options to store and process massive amounts of data, and what software testing methods can be used to ensure the integrity of the code, process and results.

Day 2 Workshops

Day two will comprise two to four practical workshops/training sessions after a session of demonstrations of visualisation tools as well as a live demonstration of modelling incidental common dolphin captures in New Zealand's commercial trawl fisheries. 

Topics include:

  • A live demonstration of modelling incidental common dolphin captures in New Zealand's commercial trawl fisheries (Finlay Thompson, Dragonfly)
  • Demonstrating the Catalyst Cloud Computing Platform (Bruno Lago, Catalyst)
  • Machine Learning for the Non-Computer Scientist: Where's the Easy Button (Shannon Schlueter, UNCC)
  • Managing Large Scale Data in a High-Performance Computing environment using the NeSI computational infrastructure (NeSI)
  • Data Visualisation (Linley Jesson, Marcus Davy, Plant & Food Research)
  • Synergy from Heterogeneous Data (Shannon Schlueter, UNCC)