Another great meeting

The mega data cluster’s fourth meeting was held in the impressive new building housing the Harry Perkins Institute for Medical Research.

Peter KlinkenWe were delighted to have Professor Peter Klinken, WA’s Chief Scientist, give a brief introduction supporting the formation of the mega data cluster. He noted that the technologies involved underpin local initiatives across diverse areas such as oil & gas, natural biodiversity, radio astronomy, bio-technology, and agriculture. He encouraged everyone to recognise we now have an opportunity to capitalise on the great projects we have – especially the Pawsey Centre, and the SKA –  as well as building on our proven expertise.


Tarun and slide - v2Looking at data on the broad scale, Professor Tarun Weeramanthri gave an overview of how epidemiologists combine data from many sources to test hypothesis about health trends and impacts. He was keen to note that epidemiologists do not just process data looking for correlation without causation, but rather form hypotheses which can be tested against the data. Epidemiologists are concerned about the quality and integrity of the data, looking for introduced bias at every step in the design, collection, processing, and analysis of the data. He made the point that “big data” is not always necessary: sometimes it is the smaller and complex data sets that yield best results, and having more data does not always improve the significance of the analysis.

One of Tarun’s key points was about the importance of place in epidemiological analysis. His unit is working with the CRC for Spatial Information to develop spatial and statistical tools, now being used by individual researchers to pose their own questions and test them against the data.

In closing, Tarun presented a vision of “Precision Public Health” – interaction of genes, environment and health underpinned by spatial tools, data linkage and bioinformatics. He then posed a couple of interesting challenges:

  • How do we get the ‘best of both worlds’: to get the most out of small, quality and complex data sets, as well as big data. That is, to combine epidemiological rigour with information analyst computing power.
  • Who in WA is working on the philosophy of big data?
  • Where are the ‘big stories’, such as the radio astronomy story, that go with the ‘big data’ ?


Nigel - 2Our second speaker, Winthrop Professor Nigel Laing, took us to the micro level describing the challenges of finding genetic causes of muscle diseases in newborn children. His laboratory was the first in the world to identify a specific genetic mutation for one sub-group of these disorders, affecting proteins important for muscle contraction.

He described the challenge of searching for a single faulty gene, given that the human genome consists of about 3 billion DNA base pairs and the natural variation in genetic material between individuals. Nigel provided a good analogy: a competent typist working 24 hours a day, every day, would take 34 years to type out the full genome. The search is like looking for one typographic error in the resulting work!

The increasing power of computing to process data and the speed with which DNA can be analysed has significantly empowered genetic research. However, the process of searching for causes of disease remains highly dependent on the skills and knowledge of researchers in knowing the right places to investigate. Collaboration remains an essential aspect. In recent years, the laboratory has increased the rate they are able to isolate specific mutations causing genetic disease.

Nigel commented whilst there might be benefits in collecting DNA samples from everyone in Australia at birth, which soon could possibly be technically feasible and within economic reach, the amount of data would amount to many petabytes of data per year.

Our thanks to Professor Peter Leedman and the Harry Perkins Institute for hosting the meeting, and especially to Meredith Eddington for looking after us so well.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s