Last month, Jorge Conde handed out the first of his company's precious products: an engraved silver box housing a USB drive. On the drive, protected by encryption software, was the sequence of the recipient's entire genome, a proprietary analysis of his genetic risks for disease, and software for browsing the data.

The event signals a shift in the world of human genome sequencing. Until now, the handful of people who have had their genomes sequenced, including genomics pioneers Craig Venter and James Watson, have been part of government or industry efforts to study the human genome or showcase new sequence technologies. But thanks to Knome, a startup based in Cambridge, MA, genome sequencing is no longer just a research tool. Anyone with $350,000 to spare and an adventurous spirit can now have his or her own genome sequenced.

Knome is at the forefront of the push toward so-called personalized medicine. Scientists and physicians hope that when sequencing costs come down enough, genetic analysis will become a ubiquitous part of health care, helping doctors choose the best treatments for a specific patient, or helping individuals take steps to prevent diseases for which they are at risk.

While Knome has not yet disclosed the number of people who have signed up for its service, it aims to sequence 20 genomes this year. "We have individuals at every stage of the process," says Conde, Knome's president and chief executive officer. "We collected data from additional clients in Europe last week and have people being sequenced now." That number may grow as sequencing costs plummet: the company soon plans to announce a lower price. And competition from other companies, such as Illumina, based in San Diego, could push down costs further.

Cheaper personal-genomics services are already available: three companies--23andMe, Navigenics, and Decode--offer analysis of hundreds of thousands of genetic variations, to predict risk of disease or assess ancestry and other traits. But Knome sequences the entire genome--nearly all three billion bases. "That has a tremendous advantage, because the information is significantly greater and more complete than other services," says Raju Kucherlapati, scientific director of the Harvard Partners Center for Genetics and Genomics at Harvard Medical School, in Boston. "Ultimately, this is the direction to go. The only impediment is really the cost."

At this point, Knome's customers are buying a product that has aspects of both a well-equipped Lamborghini and Internet service circa 1985. The $350,000 price tag is out of reach to all but a very few. But interpreting a person's genome sequence requires comparing it with a database of human genetic information linking particular variations with health or other traits. The greater the number of genomes sequenced, the richer the analysis; but at the moment, that number is pretty low. "This information is going to be thorny and problematic in terms of interpretation," says James Evans, a professor of genetics and medicine at the University of North Carolina at Chapel Hill and editor in chief of Genetics in Medicine. "We all have mutations and alterations that we simply don't understand. As usual, the technology will be ahead of our ability to use it."