Saturday, October 1, 2011

Building Core Power

Performing a single trial run of his software takes nearly two days on a typical computer for Zaid Abdo, assistant professor of mathematics, statistics and bioinformatics at the University of Idaho. This wouldn’t be so bad if Abdo only had to do one or two trials, but his research requires 5,600.

“If I tried to do my work on a single computer I would not be alive by the time it finished,” he says. Luckily, he has access to the bioinformatics core at the University of Idaho. The array of cluster computers cuts the time he requires from more than 30 years to only a few months.

The bioinformatics core is supported by the Initiative for Bioinformatics and Evolutionary Studies (IBEST) - a collaboration of scientists who apply their diverse talents to health questions too grand for any one scientific field to solve. The group is constantly expanding the core using small pieces of an extensive five-year Centers of Biomedical Research Excellence (COBRE) grant, which totals nearly $9.5 million. In fact, the facility recently doubled its processing power with the installation of a few new server towers. This allows computationally intensive projects – like Abdo’s – to thrive.

Abdo is creating software that categorizes the countless number of different microbes crucial to human development, physiology, immunity and nutrition. Though these microbes outnumber human cells in a person’s body 10-to-1, little is known about them. So scientists are creating a “map” to understand their relationship to each other and human health. Abdo’s software helps put new microbes into the correct categories; a task made difficult since even closely related microbes can vary wildly within a single sample.

Besides processing power, the bioinformatics core is also home to more than 60 terabytes – 20,000 gigabytes – of data storage capacity. This allows the core to house and provide services such as all of the major bioinformatics databases from around the world, which are updated at least once a week.

“I’ve got hundreds of thousands of DNA sequences that I have to compare to all known genomes to figure out what sort of organisms they came from, ,” says James Foster, director of the Bioinformatics Core, who is studying organisms left behind by retreating glaciers. “Instead of making hundreds of thousands of queries to an outside source, it’s actually faster to download the whole database, which involves moving terabytes of data.”

Another scientist making use of the core is Larry Forney, director of IBEST, who believes the core is the best facility of its kind between Denver and Seattle.

He also believes it could be much better.

Forney is an advocate for collaboration among all universities interested in biomedical research in the western states to upgrade the Core’s resources. Such a project would transform the University of Idaho from being strong in bioinformatics to being one of the country’s premier institutions in the field. Additionally, it would allow scientists interested in the field access to the technology required to perform the most advanced research. If successful, the expanded Core would open the door to countless exciting projects already being developed in the minds of University of Idaho faculty.

“Sequencing the genome of endangered species, measuring the genetic diversity of a threatened species, finding why a particular crop is resistant to cold, insects or pesticides – these are only a few of the possibilities,” says Forney. “We can even use a process called reverse transcription to discover which how gene expression is affected by different behaviors. But in order to do that, we have to be able to handle the massive amounts of data.”

Read More

Return to News