Meet your unborn child – before it's even conceived

New Scientist, 9 April 2014

    A service that creates digital embryos by virtually mixing two people's DNA will allow parents to screen out genetic disorders – and perhaps much more

     Credit:  Hannah Gal/Science Photo Library

    Credit: Hannah Gal/Science Photo Library

     

    WILL my baby be healthy? It's a question that concerns every prospective parent. Now a service that creates digital embryos by virtually mixing two people's DNA will give a clearer glimpse of their possible child's health, and perhaps much more – before it has been conceived.The Matchright technology will be available in two US fertility clinics later this month, allowing people to screen out sperm donors who, when their genes are combined with those of the intended mother, could increase the risk of a child inheriting genetic diseases. The company that markets the technology, GenePeeks, hopes to expand worldwide.

    But the technology's patent also includes a list of traits that aren't necessarily related to health – such as eye and skin pigmentation, height and waist size – raising concerns that it could be used to select embryos on the basis of more superficial characteristics. "It covers any disease or any trait that has a genetic influence," says Lee Silver at Princeton University, who co-foundedGenePeeks – even those where the genetic basis has yet to be discovered.

    To find out how the technology may affect parents' future choices, New Scientist sent the patent to people working in reproductive health.

    The priority should be medical problems, says Martina Cornel of theEuropean Society of Human Genetics. This is what GenePeeks plans to do. It intends to use the system to identify rare conditions such as cystic fibrosis and Tay-Sachs disease, which are passed on to a child when both parents carry a mutation in a single gene.

    Screening for genetic disorders usually involves sequencing the DNA of the prospective parents. GenePeeks takes this a step further: algorithms are fed this information and use it to digitally recreate the process of genetic recombination – the mixing of genetic information between a sperm and an egg. This allows them to look at the genetic make-up of the possible embryos.

    Before a woman selects a donor from a participating fertility clinic, the Matchright algorithms, which Silver developed, are run thousands of times for each donor. This produces up to 10,000 simulated embryos per pairing. These are sequenced to look for mutations in single genes that can cause some 500 rare diseases, and then used to work out the disease risk in the hypothetical child that would develop from that particular partnership. The woman, who pays $1995 for the service, then gets a list of "safer" donors from which to choose.

    To know which mutations to look for, the software searches databases of genes linked to different conditions. To validate the method, the company used the software to digitally pair anonymous men and women whose genomes were sequenced as part of the 1000 Genomes Project. GenePeeks then screened their virtual embryos and compared the incidence of predicated disease to that in the general population.

    "These studies confirmed the system's ability to accurately predict a future child's risk profile," says co-founder Anne Morriss. However, only when a critical mass of children are born using the system will its true power become apparent. And even then it's still a game of probabilities – you might shift the likelihood of passing on traits by screening out certain donors, but it doesn't rule out the effect of spontaneous mutations that might arise during development.