The Pawsey Supercomputing Centre


Supercomputers and global food security

Assistant Professor Laura Boykin and her colleagues at The University of Western Australia’s Plant Energy Biology ARC Centre of Excellence are using the Pawsey Supercomputing Centre’s resources and expertise to assist farmers in poverty-stricken Eastern Africa by building more accurate identification systems for highly invasive pest species, including Bemisia tabaci, the silverleaf whitefly. These systems could potentially bring great benefits to researchers and farmers all over the world, including Australia.

The Challenge

Whiteflies are one of the most pervasive pests on Earth, being found on every continent except Antarctica. While some species of whitefly are harmless native populations, many are highly invasive pests. These invasive species feed on valuable crops and spread viruses, causing more plants to die.

A/Prof Boykin’s project in particular looks at the silverleaf whitefly, Bemisia tabaci. This species is spread throughout the world, including populations in the United States, Australia and Eastern Africa.

In Eastern Africa in particular, these pests cause widespread devastation to the cassava plant, an important food staple. Destruction of these crops can cause famine and economic hardship in the region.

A key problem for researchers and farmers is the visual similarity between various species of whiteflies. This makes differentiating between harmless and invasive species almost impossible, as well as hindering management strategies. Different species respond to different strategies, such as pesticides and biological controls.

“It’s a pest which is found all around the world affecting agriculture wherever they go, and the techniques that we’re developing with the East African whiteflies can be applied with researchers and farmers all around the world,” says A/Prof Boykin.

The Solution

A/Prof Boykin is using phylogenetic techniques to better understand the relationships between whiteflies around the world. Running a program on ‘Magnus’ using the Markov chain Monte Carlo method, the project is able to genetically distinguish between Bemisia tabaci and other harmless species that look nearly identical.

“This project is all about ‘knowing the enemy’. To a farmer, these species all look the same. This project could help develop diagnostic tests that will tell farmers if they have a harmless species or one they have to get rid of ASAP, which is invaluable,” says A/Prof Boykin.

The cassava plant

The cassava plant

The Outcome

The genetic datasets involved in this project were so large, they would be almost impossible to calculate using a desktop computer. In addition, the sophisticated techniques used in this project are also highly computationally intensive. This requires considerable supercomputing power to perform efficiently.

Using ‘Magnus’ and the Pawsey Supercomputing Centre’s resources and expertise, these calculations can be performed in a practical timeframe.

“The Pawsey Supercomputing Centre has really helped,” says A/Prof Boykin.

“The beauty of this particular project is it really is a true engagement. It’s not just me using the resources, I’m interacting with the scientists and staff there as well. It’s made a huge difference in pushing this research forward.”

Download PDF version here.

Comments are closed.