Genetic Testing Gets Personal

Perhaps most denigrated by experts are those that purport to identify people's nutritional needs from their DNA and then sell them dietary supplements at a hefty profit.

"It is totally bogus," said Gail Geller of the Berman Institute of Bioethics at Johns Hopkins University, whose research has documented how easily the public can be bamboozled by genetic test results.

Then there is ScientificMatch.com, which "uses your DNA to maximize the chances of finding chemistry -- actual, physical chemistry -- with your matches," according to the company's Web site.

At the heart of that claim is a hypothesis that people are most attracted to others whose immune systems differ most from their own. A few studies have found evidence supporting the idea (it may be an evolutionary strategy for maintaining genetic diversity). But at best, geneticists say, it is a narrow basis upon which to choose a mate.

"It creates an air of charlatanism that doesn't help the field," Venter said.

All told, concluded a study in this month's issue of the American Journal of Human Genetics, "There is insufficient scientific evidence to conclude that genomic profiles are useful in measuring genetic risk for common diseases or in developing personalized diet and lifestyle recommendations for disease prevention."

The Science Is Still Young

Despite today's limitations, the day will come, experts agree, when enough will be known about human genetics so that a scan of an individual's genome will convincingly predict that person's medical risks and behavioral foibles -- perhaps with enough assuredness to dictate preemptive therapy or even extend disability rights to some whose behavior falls outside societal norms. But the only way to get there is to collect massive amounts of data from a wide array of people so computers can find those correlations.

That task is underway, but the work takes time, which is why direct-to-consumer genomics companies say they should be welcomed. Most people are disinclined to sign up for research that offers nothing in return, Wojcicki said, but at 23andMe, "they get something back right away, and they are also part of something really powerful."

Wojcicki predicts that as members share information about their genes, their health and their personalities -- an irresistible option for many in this age of electronic "friending" -- the new enterprises will revolutionize health care "the way YouTube revolutionized media."

"I call it the democratization of the genome," Venter said.

Concerns persist. If people want to use their information in a meaningful way, they will probably want to share it with their physician, said Francis S. Collins, director of the National Human Genome Research Institute and a leader of the Human Genome Project, completed in 2003, which cobbled together the first complete human DNA sequence. And medical records are not totally opaque to prying eyes.

"People ought to think about that," said Collins, who confessed to feeling both excited and concerned about consumer-driven genomics. "We don't want employers to use genetic information to make hiring or firing or promotion decisions on the basis of fears that an employee may get sick." It is "enormously frustrating," he added, that bills prohibiting genetic discrimination have been passed by both chambers of Congress but are stalled because of an unrelated power struggle on Capitol Hill.

One subtle but potentially insidious downside of the new trend, Collins said, is that people may slip into the DNA-deterministic thinking that fed the early 20th-century eugenics movement, in which people with "undesirable" traits underwent forced sterilizations.

"I very much worry that all this emphasis on a 'gene for this' and 'gene for that' raises the risk that people will conclude that that's the whole story," Collins said. Instead of empowering people to make healthful changes in their lives, that could simply make them "more fatalistic," he said, "in which case, what's the point?"

At the same time, he and others acknowledged, by identifying people with similar genes but different health outcomes, genomics companies' databases could help scientists identify the specific environmental influences -- diet, exposure to certain chemicals, even stress or abuse -- that interact with particular genes to make people into the individuals they are.

"By disentangling the genetics, we'll get a much deeper appreciation of both nature and nurture," said Church, the Harvard geneticist. "I would be surprised if it didn't change our view of ourselves pretty significantly."