Duke Research: Creating Knowledge. Serving Society

Keep me posted on research news.
Learn more >>

After decades of prying biology apart in ever-smaller pieces – from cells to molecules, and from forests to sub-species – a growing number of Duke researchers has begun moving in the opposite direction, probing how thousands of pieces come together to form complex ecosystems.

Using powerful new computers that can spot patterns in mountains of data, they are finding surprising new connections within systems that vary from Costa Rican rainforests to the bacteria inside a premature baby.

“It’s a new way of thinking,” said Steve Haase, an assistant professor of biology in the Duke Institute of Genome Sciences & Policy (IGSP). “We’ve spent decades on a reductionist approach to science” in which researchers would painstakingly disturb one small part, a single gene for example, and then try to figure out what it had been doing. “That method has been phenomenally successful. But now, with genome technologies, we have the opportunity to look at the dynamics of all the genes at the same time.”

This new, systematic approach to biology is like the difference between trying to understand a Swiss watch as a pile of discombobulated parts or as a ticking timepiece.

Beyond just being able to spell out 100 million letters of the A-G-C-T of DNA code before dinnertime, biology’s new machines can measure which genes are actively producing proteins and which are silent; which stretches of the chromosome are packed for long-term storage and which are unpacked and ready to use. They can count the different versions of a single gene in a given organism, or sort out how many different species there might be in a given sample.

The technology is bringing scientists closer to seeing the ticking watch every day.

Duke biologists Francois Lutzoni and Daniele Armaleo are using a machine known simply as “the 454″ to simultaneously sequence the genomes of two organisms — a fungus and an alga — that work together to form what we call lichen. Already, the new work has yielded an exciting surprise: the first direct evidence that genes have been transferred between the fungus that gives lichen its structure, and the single-celled green algae that fuels it by photosynthesis.

The 454, or more formally the Roche GS-FLX, is what’s known as a “long-read” sequencing machine. It can read the sequence of 400,000 sections of a DNA sample that has been chopped into 435-letter chunks, and then stitch the results back together with software. An 8-hour run yields 174 million letters of DNA code.

Its companion, the Illumina GAII (aka “Solexa”), is a “short read” sequencer that parses 50 million chunks of 35-letter DNA in a two-day run, yielding 1.5 billion letters of code.

Lichen is a symbiotic community of algae and fungus working together, and controlling each other’s genes.

Because the 454 is really good for genome-diversity projects, microbiologist Patrick Seed of the Department of Pediatrics is using it to identify the bacteria that colonize the intestines of extremely preterm infants, born before 28 weeks gestation. The makeup of the gut “microbiome” — a complex ecosystem that sustains each of us — has important health consequences for human adults and animals, and may weigh heavily in the survival of pre-term infants.

Duke biologists Greg Wray and Tom Mitchell-Olds will use the 454 in an attempt to profile all of the active genes in the leaves and roots of some of the Costa Rican rainforest’s most dominant tree species. Their goal is to see how gene expression profiles adapt to conditions — during the wet and dry season and under different temperatures, for example. They may be better able to understand and predict how the forest will respond to conditions in a changing global climate.

Wray, who is director of the IGSP’s Center for Evolutionary Genomics, has spearheaded the effort to add the new capacity to the IGSP’s primary sequencing facility in the Biological Sciences Building. “People were coming to me and saying, ‘I’m tired of sending my samples out for sequencing, paying a lot of money, having no control over what happens to them, and then having to wait six months. Why don’t we just get some of these machines ourselves?’”

So Duke did, to the tune of about $1 million dollars. Depreciation on the valuable instruments, as well as the salaries, maintenance and supplies to keep them running, will come from grant funding, but Duke invests in the infrastructure to maintain its leadership, Wray said.

Incidentally, IGSP’s information technology people also had to set up 28 terabytes of storage — think of it as a half-million PCs worth of hard-drive — to catch all the data coming out of the new sequencing machines. A single graphic image showing the output of a run can be as large as a terabyte.

Wray calls this kind of science “drinking from a firehose.”

IGSP Director Hunt Willard says it’s more like surfing. “Riding a breaking wave is inherently messier and costlier than coasting along later on calmer seas. We may get lots of spray in our face, but I’m betting it’ll be well worth it. It wouldn’t mean much for us to want to be ‘almost’ cutting edge.”

Kendall Morgan is the Assistant Director for Communications in the Duke Institute for Genome Sciences & Policy.

R/V Cape Hatteras

Tina sails to the Gulf of Mexico aboard the Duke Marine Lab's research ship.

Blog: BNet Pharma