REED HOME Gryphon icon
Alumni Profiles
reed magazine logoSummer 2008

Tracking Disease in DNA

Anaka Narananan

Christopher Amos ’80 with colleague Dr. Margaret Spitz

Scientists used to hunt genetic diseases by tracking sickness and death through a genealogical diagram known as the family tree.

Christopher Amos ’80 trails such mysteries through a much bigger family—thousands of people connected to an ancient human ancestor by shared race or medical problems. A self-described “statistical genetics geek,” Amos helps to convert the secrets hidden in genetic coding into information that may help diagnose or even treat diseases. His genetic hunts have uncovered factors behind some of society’s most crippling diseases.

In his most recent success, Amos led the team that identified a gene that plays a significant role in lung cancer. The gene encodes a protein that can bind with nicotine, producing what Amos calls a “doubly whammy” for carriers: they are more likely to get addicted to nicotine and more likely to develop lung cancer when they smoke. The finding has helped revive interest in nicotine as a carcinogen.

“For years, the tobacco companies have been saying, ‘Well, it’s just the tar, and we can eliminate that,’” says Amos, who worked at Howard University and the National Institutes of Health before moving to M.D. Anderson Cancer Center in Houston, where he is now a professor of epidemiology and biomathematics. “But it might not just be the tar.”

Amos grew up in tobacco country, though he’s never been a smoker. (“In France,” he says, “I tried one of those Gauloises and I thought, ‘This is horrible. Why would anyone do this?’”) His father, a professor at Duke University, wanted to be a yeoman farmer and bought some land near Durham, North Carolina.

“We did a lot of picking up hay and chasing cows,” Amos says. “I don’t do it any more…It’s too hard.”

At Reed, Amos majored in math, mixing in biology and physics courses. His senior thesis involved geodesics, what he now calls “some kind of obscure way of doing computations, which I haven’t used since.” After graduating, he spent three years working as a bartender, installing solar water heaters, and whitewater rafting before heading to Louisiana State University for a doctorate in biometry, the field where statistics and biology are integrated. For biologists who work at the genetic level, and who generate mountains of data too massive to sift without supercomputers and sophisticated statistics, it is an increasingly vital discipline.

Amos’ projects start in deep freezes, where doctors across the world store samples of blood and tumor and other tissue harvested from patients.

Scientists screen these samples with gene chips that can find 300 thousand to 2 million genetic forms in one test. Comparing these forms and assessing the likelihood someone with a certain genetic mutation has of getting a given disease can lead researchers to the small stretches of chromosomes where disease risks lie. Amos narrowed down the location of his “double whammy” lung cancer gene to a section of chromosome 15, about 88 thousand DNA bases long out of about 100 million bases on that chromosome.

“The odds of that occurring by chance are equivalent to something like the number of stars in the universe,” Amos says. “But in the field, it’s actually not so unusual these days.”

Much of Amos’ effort goes to building research teams that can get their hands on enough patient samples to do this work. These studies often need DNA from thousands of people to separate signals of harmful genes from the background static of random mutations. The lung cancer study, published in Nature Genetics in May, screened samples from nearly 2,300 Caucasians, united by some long-dead common ancestor.

“The people who have an increased propensity for lung cancer are a little bit more closely related” than are other Caucasians, and are more likely to share a cancer-promoting gene, says Amos. “But how much more closely related is pretty minimal. That’s why you need so much data.”

Amos applies the same statistical tools to other diseases, especially autoimmune disorders. Last year, he helped link several genes to rheumatoid arthritis. But he’s drawn more notice for the discovery about lung cancer, which kills more U.S. residents than any other cancer.

Amos’ research team is now running detailed tests on a larger number of cancer patients to confirm precisely which gene forms link to lung cancer. Identifying the specific gene will let biologists determine the protein it makes and study how that protein helps cause addiction or uncontrolled cell growth. And that would be a target for drugs that might help people break the smoking habit or, Amos hopes, battle back at lung tumors.

—Andy Dworkin

reed magazine logoSummer 2008