Bacteria need sugars to survive. So they grab sugars where they can – either by making them or by taking them up from the environment – and mold them into a form that can be used nutritionally (to make energy) or structurally (to build a cell wall, for example). In turn, a bacterial cell’s sugar give-and-take can influence its environment, whether that’s water, soil or the human gut. With the long-term goal of developing ways to manipulate bacteria for a desired outcome, like new antibiotics or producing alternative energy, scientists are piecing together the complicated machinery that bacteria use to modulate sugars. In doing so, they face the major challenge of figuring out which genes are involved and what roles they play in sugar processing.

Sanford-Burnham’s Dr. Andrei Osterman addressed this problem in a talk he gave last week at the San Diego Consortium for Systems Biology’s 5th Annual Systems to Synthesis symposium, held at the Salk Institute for Biological Studies. Two types of bacteria that Dr. Osterman uses to study sugar processing pathways, Thermotoga maritime and Shewanella oneidensis, may have potential industrial applications to produce biohydrogen or clean up nuclear waste.

Early in his talk, Dr. Osterman summed up his group’s method for pinpointing what a gene does. “Coming from Russia, I think of it as a very American approach,” he joked. “We try to figure out what’s going on by taking a look around the neighborhood.”

He means the genomic neighborhood, of course.

In bacteria, proteins that work together are often coded by genes that reside near one another in the genome. Dr. Osterman’s group uses computer models to reconstruct and predict sugar pathways in T. maritime, S. oneidensis, and other bacteria, taking hints from a gene’s structure and location. Since there are so many different species of bacteria and so many genes to look at, Dr. Osterman’s computer programs help speed things up. This way he can compare thousands of genes across many species.

Then, once he has a hunch that a certain gene or group of genes is responsible for a particular function, the work moves into the lab, where scientists do experiments to test the computer’s theories. While computer modeling is quickly developing thousands of new leads for gene function, the next step is to speed up the lab work.

“We are now trying to find a more high-throughput way to test predictions,” Osterman explained. “Because it’s becoming unbearable doing things one by one.”

As Dr. Osterman expands and confirms his sugar machinery maps using model bacteria species, he hopes to build a “genomic encyclopedia” that can be used to confidently recognize similar genes and pathways in more complex species.