INTERVIEW: Getting up to Speed

The thing about bioinformatics, according to Prof Terry Speed, is that it tends to attract people from a variety of disciplines, such as physicists with programming skills not interested in a career in defence, or mathematicians with a practical bent.

Speed therefore believes that the way to build up bioinformatics skills in Australia is to target researchers in other disciplines with an interest in the developing field via grants and fellowships.

"It's my bandwagon," says the researcher, who splits his time 50-50 between the Walter and Eliza Hall Institute, in Melbourne, and the University of California, Berkeley. "There's a need for targeted research grants and fellowships in areas like bioinformatics, where people might not have years of experience in the area. But there is not enough interest from the funding bodies."

And undergraduate programs in bioinformatics are not necessarily the key either, he says, as there is too much to be learned for an undergraduate degree. Instead, he advocates bioinformatics should be taught as a specialised degree at the postgraduate level. In Australia, such programs are just getting off the ground.

Typically, Speed says, he will take on researchers with an interest in bioinformatics and some programming experience to work on specific tasks, such as writing user interfaces for the software his group develops.

"I let them work on that for a while and then work out what they are interested in," he says. "Self-education is required."

There is plenty of scope in Speed's group for researchers to choose from, including microarray data analysis, statistical genetics, and biological sequence analysis. His teams combine their own research with collaborations with other researchers both inside and outside the institute, and also perform quite a bit of consulting work ranging from advising scientists on genome analysis to straight biostatistics.

Part of the Genetics and Bioinformatics division, the group is closely associated with Simon Foote's genetics team, but also works with other WEHI divisions including the Infection and Immunity division, which performs a lot of genetic analysis on the malaria and Leishmaniasis pathogens. Among the projects they are involved in are several that are part of the Gene CRC, including the search for genes shared among sufferers of multiple sclerosis, which has led to the development of novel approaches to the analysis of genome data.

"From the point of view of the rest of the world we are pigeonholed as microarray people, but we do a lot of other things," says Speed.

One of the key attributes of the group is the commitment to developing publicly available bioinformatics tools, especially for microarray analysis and sequence analysis.

"We have NHMRC funding to make tools available to Australian researchers," Speed says. "Core algorithms, basic methods -- we're trying to make them freely available. But they are not necessarily user-friendly."

Hand-in-hand with Speed's belief that software should be publicly available is his aversion to commercialisation.

"We are not interested in the commercialisation side -- it's not our philosophy," says Speed. To this end, the software developed by the groups is made available through public licences, and other researchers are required to make improved versions freely available as well.

Speed and his team have no connections with the Victorian Bioinformatics Consortium, an initiative of the State government in Victoria primarily involving Monash University, CSIRO and Agriculture Victoria's Plant Biotechnology Centre. "We decided not to join, but would be willing to collaborate," he says. "But it hasn't happened at this point."