For $1,000 and up, several new companies will scan an individual’s entire genome for clues about ancestry, potential health limitations and the inheritance of traits such as lactose intolerance. Clients can compare their DNA with a celebrity’s or invite friends and family members to share genetic profiles. Despite the comprehensive reports and background data these Web-based services deliver, some observers believe the information is more recreational than relevant.

Direct-to-consumer genetic tests have existed for at least a decade, and in recent years the number of choices has exploded. Whereas most of these offerings probe for only a small number of gene variants, advances in genome chips now allow a quick, inexpensive search for a wide range of targets all at once. Navigenics in Redwood Shores, Calif., 23andMe in Mountain View, Calif., and deCODE Genetics in Reykjavik, Iceland, recently began scanning for markers associated with as many as two dozen conditions and traits. And for upward of $350,000, Knome in Cambridge, Mass., enables customers to join J. Craig Venter and James D. Watson in the elite cadre of humans who have had their entire genome sequenced, analyzed and interpreted.

With new tools, reference sequences and big study populations in hand, geneticists have found increasingly robust associations between DNA variations and disease susceptibility. But the data are still incomplete and sometimes conflicting, cautions Muin Khoury, director of the Centers for Disease Control and Prevention’s public health genomics office. For now, he says, sequencing one’s genome or scanning for susceptibility markers offers “no useful information.”

Except in the case of rare disorders caused by a single gene variant, having a genetic susceptibility is far from a guarantee of falling ill. Multiple genes interact within a complex biological system that includes many other important players, among them RNA and chemicals in the environment. Complex conditions such as diabetes or heart disease have myriad behavioral and environmental components working in concert with an unknown number of genes.

With so much still to learn, it is too early to use results from gene association studies for health-planning purposes, according to Khoury. Besides, he points out, it is unclear in the medical literature whether news about genetic susceptibility to particular conditions has any power to change people’s habits. Moreover, physicians are unsure how to apply the drug metabolism information from pharmacogenomics tests in their prescribing decisions. And in some cases, the tests offered online seem altogether disconnected from genetic medicine. A July 2006 U.S. Government Accountability Office study of services provided by nutrigenetics companies questioned whether, in some cases, any DNA was analyzed at all. The agency concluded that the firms “mislead consumers by making predictions that are medically unproven.”

The new genotyping companies claim to cut through the confusion by delivering high-quality, responsible science. They stop short of offering medical services and instead promote their scans as an innovative means to provide health information and empower consumers to act on it. While genetics research moves forward, the companies say, people can start working now with their doctors or take steps to change their diet or other behaviors. “It’s better to start than to wait an indefinite period of time until the science is perfect,” insists Mari Baker, chief executive of Navigenics. Not all of the 1.8 million points her company scans on the genome are informative today, she says, but customers will be able to access ongoing findings as they come in. Eventually Navigenics plans to incorporate other health data, such as family history and medication use.

Some experts worry that the rollout strategy could backfire. Because the relevance of whole-genome scans is so limited, warns Sharon Terry, president and chief executive of the advocacy group Genetic Alliance, “the average person may lose interest before there is enough information to have utility.” Along with creating tools such as an online guide to family health history and a WikiGenetics Web site, Genetic Alliance is pressing for regulations that would protect against discrimination, safeguard privacy, and require quality and validity testing. So far efforts to extend laws into these areas have failed.

Navigenics, 23andMe and deCODE specify that customers own their personal data. But executives keep the door open to use their growing databases for research with commercial or nonprofit partners. Such studies should take place under research protocols, not as an outgrowth of consumer marketing, the CDC’s Khoury argues. For now, he says, the best tool available to personalize medicine is low tech and low cost: family health history. It captures the effects of multiple genes, shared environment and common behavior. But less than one third of the population has actively collected such information. Genotyping is a wonderful research technology, Khoury remarks, but “it’s going to take a long time to translate gene discovery into action.”

Editor's Note: This story was originally printed with the title "Taking Genomes Personally"