Little Lab on the Prairie
The good people of Lethbridge probably won’t like this, but it feels a little odd coming to a small university in what can only be described as a rural city to learn about one of Canada’s newest and fastest computers
by Jeff Gailus
Don’t get me wrong. Lethbridge has its virtues – climate, for one, and a reputation for friendliness, not to mention an almost infinite source of carbon-free power blowing in the wind – but the cutting-edge zeitgeist of a technologically obsessed 21st century? All I saw driving into town were wheat fields.
But when I finally arrive at the university and meet Stacey Wetmore, I know I’m in the right place. Wetmore, a computational chemist whose research recently earned her a prestigious Canada Research Chair at the University of Lethbridge, has spent more than 10 years using computers to figure out how everything from radiation to industrial pollutants trigger chemical reactions that damage our genetic material, DNA. These are not academic questions: disrupted DNA can lead to serious diseases like cancer, and the opportunity to figure out how that damage is caused – and how it is repaired – will help open the door to finding a cure.
“If we understand what [chemical reactions] actually cause the disease, then we’re in a better position to figure out how to stop it,” Wetmore explains while I watch, out her window, as mule deer parade across southern Alberta’s drought-dry prairie. “This puts us in a better position to develop cures, rather than just drugs that [try to] take care of the end result,” such as chemotherapy treatments for cancer.
To do this Wetmore needed a computer capable of crunching enormous amounts of data at incredible speeds. Using some of the $625,000 she received with her Canada Research Chair, the University of Lethbridge partnered with computer maker Dell Inc. to build Uracil (named for a molecule of damaged DNA that Wetmore and her team study). Uracil is a supercomputer of sorts that doubles as one of Canada’s most powerful virtual chemistry labs – 86 servers networked together into a cluster, Alberta’s biggest, with the power of 700 personal computers.
When Wetmore opens the door to the large concrete storage room in which Uracil lives and breathes, air rushes past us with a whoosh. In its attempt to quell the very palpable heat of 688 voracious microprocessors crunching data faster than the speed of thought, the air conditioner has created a vacuum.
“Five years ago, it would have been impossible to run the calculations we are currently running in our lab,” Wetmore says. “With the installation of this cluster, we will be able to advance our research even further by using larger models and more accurate theories.” Each Dell server, or node, is a supercharged version of a laptop computer, about the size of a home-entertainment CD player. They are stacked one upon the other, all 86 of them, in three black metal racks. Bright blue cables (the same kind most of us use to bring high-speed internet into our homes) link them together into a single computing unit. When I walk behind the servers, heat washes over me, the room humming with the energy of a thousand bees.
Computational chemistry is a complicated business, but the computers, and the software that drives them, have made it increasingly possible to figure out what’s going on in our bodies
at the molecular level. Experimental chemists – the men and women in white lab coats, stirring solutions and gazing wistfully at Petri dishes – have been able to uncover many of the biochemical mysteries that determine how our bodies function. But it is the marriage of mathematical models and increasingly sophisticated, and faster, computers that allow virtual chemists like Wetmore and her team of computationally inclined graduate students to explore not only what damages DNA, but how and why the damage occurs in the first place.
“It might be easier if I show you,” says Wetmore, turning to her computer and opening up the GaussView software that makes computational chemistry look as easy as word processing. She cites as an example the fact that substituted phenols (found in pesticides and the byproducts of industrial processes, such as turning trees into wood pulp) have been linked to various kinds of cancer and leukemia. With a few deft movements of her mouse, she pulls up a three-dimensional version of a single molecule of DNA, which she can view from any direction. It is guanine, one of four nucleobases that combine in innumerable patterns to make up every strand of DNA in every living thing, including me and her and every one of the billions of blades of prairie grass that gleam like the sun itself outside the window. She points out the sugar molecule that forms part of the double helix’s backbone, where the guanine attaches, and then the phenoxyl radical that, in the real world, would attach itself to the guanine – in the process, creating a whole new molecule. Which is to say, damage the DNA and, if it’s not repaired, cause the organism beholden to it significant grief.
If this were a real analysis, Wetmore would expand the model to include more and more of the DNA, and then ask Uracil to figure out how one damaged guanine molecule might lead to more damage to
the DNA helix, and perhaps, over time, more disease. Even more important, Wetmore’s team can use Uracil to figure out how enzymes repair damaged DNA. Every day, in each and every one of our bodies, DNA is damaged and repaired well over 10,000 times, in a constant cycle reminiscent of London during the Blitz. “If we can understand how nature repairs the damage,” Wetmore says, “that will give us clues about how to fix it,” an obvious boon given the damage can lead to potentially fatal maladies like cancer.