Biologist does the math to get inside cells

David Baumler’s computer is full of bugs, but the UW–Madison Genome Center scientist isn’t the least bit worried about it. That’s because Baumler’s bugs aren’t the malicious, viral kind, but elegant mathematical representations of bacteria that are taking the study of cells out of the Petri dish and into the PC.

Engineers routinely use computer models to design and test everything from medical instruments to manufacturing processes before building them in the real world, but until recently no comparable models of living cells existed. Now, with the models Baumler is constructing, bioengineers can tinker with possible genetic improvements to virtual cells before going through the labor of adding and subtracting genes in actual microorganisms.

“In the past, bioengineering has been a tedious, trial-and-error process,” says Baumler, who collaborates on the work with genetics professor Nicole Perna and chemical and biological engineering professor Jennie Reed. “But now, this type of approach can quickly tell us which changes are going to be beneficial and which are dead ends, to keep us from going down those dead ends in the first place.”

In the models, each chemical reaction, including the reactants, products and enzyme that carries it out, is represented by a single mathematical equation. Thousands of these equations are then loaded into an algebraic matrix. It’s like rendering a giant chart of all the cell’s biochemical pathways in math, says Baumler, except that unlike a static chart, the matrix allows him to put the cell’s reactions into play. That is, by using it to solve all the equations simultaneously, he can see how the entire network works together to produce cell biomass, for example.

Baumler can also watch what happens when certain reactions are removed, just as biologists do when they create “knock out” mutants lacking specific genes and the enzymatic functions they encode. Supported by the UW BACTER Institute, a U.S. Department of Energy-funded training program in computational biology, Baumler is specifically hunting for changes that will boost production of a biofuel, ethanol, by bacteria.

“Part of the idea is that cells, unlike machines, tend to have wasteful pathways. From the standpoint of what we’re interested in, they don’t contribute,” he says. “So the question is, can some of those pathways be eliminated, so that we maximize our end?”

What’s more, by constructing models of several E. coli strains and their relatives within the family enterobacteria, Baumler is paving the way toward mixing and matching the best of these bugs. For example, the ability of E. coli, the world’s most well-studied bacterium, to ferment ethanol has researchers eyeing it as a possible industrial producer. However, the bacterium can only ferment glucose and other simple sugars, which means that corn stover and other cellulose-rich plant materials would still need expensive pretreatment to break them down into sugars before E. coli could make use of them.

Among its enterobacterial relatives, however, are a number of plant pathogens that can break down these tougher plant parts with ease. If the genes responsible for this capability could be added to an ethanol-producing E. coli strain, the result would be a single microbe that could carry out both steps of the process.

Ideas like this are nothing new; indeed, researchers have been swapping genes between bacteria and other organisms for decades. What is new, says Baumler, is the ability to test-drive genetic changes within the context of an organism’s entire metabolism. In other words, rather than reducing a cell to its parts, such as individual genes, pathways and so on, the new approach, called systems biology, permits study of the cell as a whole.

It’s not for the faint-hearted, however, or, shall we say, the math-phobic. In fact, Baumler — an experimental biologist by training — spends much of his spare time learning weighty subjects like linear algebra and computer programming. “When I started in this field, I basically went back to school,” he laughs.

But going back to school doesn’t mean he’s going backward. For Baumler, pursuing systems biology is the most natural progression of his career he can imagine.

“I’ve characterized single enzymes, I’ve characterized strings of enzymes in pathways,” he says. “And now, I’m looking at the entire network with math.”