Researchers at the University of Washington have assembled the first comprehensive genetic map of an unborn child — a development that could help usher in a new era of prenatal testing.

By analyzing fetal DNA circulating in the mother's blood, the scientists were able to sequence the baby's genome 18 weeks into the pregnancy. The technique also worked at eight weeks, with slightly lower sensitivity.

Because the approach requires only a blood sample from the mother and saliva from the father, it poses none of the miscarriage risk associated with invasive tests such as amniocentesis. And while most existing prenatal tests are designed to check for single disorders, including Down syndrome, a full-gene scan has the power to reveal a wide range of potential problems before birth, said lead author Jacob Kitzman, a doctoral student in genetics.

"It's much more comprehensive."

The procedure is still several years away from commercialization, project leader Jay Shendure said.

But the UW study, published in the June 6 issue of Science Translational Medicine, marks a significant step forward in technology that's been developing over the past several years — and which worries some people, said Marcy Darnovsky of the Center for Genetics and Society in Berkeley, Calif.

"I think it's a game-changer," she said. Cheap, safe genome sequencing could give parents the power to practice a kind of eugenics, preselecting children based on desirable traits.

"It could become a routine part of prenatal testing...which raises questions about what people will do with the information," Darnovsky said.

Shendure cautioned against expecting too much — at least in the near future. Scientists may be able to sequence the 3 billion DNA units that make up each person's genetic heritage, but they still don't understand the genetic basis of most common diseases.

"The capacity of genomics to generate data is outstripping our ability to interpret it in useful ways," he said.

Some genes can raise the odds that a person will develop conditions such as diabetes or certain types of cancer. But most diseases arise from a complex tangle of multiple genes and environmental factors.

What should prospective parents do, Darnovsky asked, if they're presented with a report that predicts their unborn child has a 30 percent higher chance of developing breast cancer at age 50?

"The meaning of this data is going to be unclear for a very long time," she said.

That's why the UW scientists predict genome sequencing in the womb will initially be most valuable for diagnosing the more than 3,000 diseases, including cystic fibrosis and sickle cell anemia, known to be caused by a glitch in a single gene. Though individually rare, these single-gene diseases strike about 1 percent of newborns and can account for up to 40 percent of the pediatric care in some hospitals.

If a test for those diseases is available when he becomes a father, Kitzman would probably opt for it.

"On balance, I would prefer to know if my child was at risk for a serious disorder so that I could begin to explore the options for therapy and care," he said in an email.

UW obstetrics & gynecology professor Dr. Edith Cheng, who was not involved in the study, said that more than a third of prospective parents don't take advantage of existing genetic tests and that few of her patients have been clamoring for more.

Prenatal sequencing will definitely be useful for families, particularly as the cost drops, Cheng said. But she predicted it also will be a powerful tool to study the way humans develop in the womb by tracking genes that switch on and off throughout gestation.

The scientists found no troubling genetic signatures in the baby whose genome was sequenced, nor could they have alerted his parents if they had. The couple were anonymous.

The UW study built on the work of researchers in Hong Kong, who discovered that DNA from the fetus appears in the mother's blood in varying concentrations throughout pregnancy. A prenatal blood test for Down syndrome that recently went on the market also targets that fetal DNA, and a few companies claim they can use it to predict a baby's sex.

But to sequences the fetus' entire genome, the researchers had to launch a needle-in-a-haystack search to differentiate the tiny fraction of fetal DNA in the mother's blood from her own. They did it by mapping both the mother's and father's genes, and using statistical analysis to identify sequences that differed — and therefore must have come from the fetus.

To test their results, the researchers sequenced DNA in cord blood collected after the healthy boy's birth. The sequence from the prenatal test was more than 98 percent accurate. Sequencing of another fetus eight weeks into the pregnancy, when the level of DNA in the mother's blood is lower, was about 95 percent accurate.

The researchers filed for patents on their method, which could mean royalties for the UW if it proves marketable.