We should be less frightened of wider access to our DNA profile — those in the know are offering theirs up for public view
The 1997 movie Gattaca sets out one of the great dystopias of science fiction. Genetic technology has divided humanity. Social class is determined by DNA. The “valids”, with sound genetic profiles, dominate the top jobs and political power. The “in-valids”, with flawed genomes, form a genetic underclass.
The film’s vision highlights a deep concern about the coming genomic age: that as science reveals more about the human genetic code it will not only herald new approaches to medicine but also create new ways to discriminate and invade privacy. Employers might select candidates according to genetic aptitude. Health insurers could refuse cover to people with high-risk genomes. And personal information about our health and personalities might be revealed for all to see.
Such fears have built a consensus that ethical use of genetic information must be founded on strict privacy. If DNA profiles are to improve treatment and prevention of disease without compromising liberty, access must be controlled. They belong to individuals and should not be disclosed without their consent. The prevailing assumption is that we should guard such data closely.
It is far from clear, however, that this assumption is correct. True genetic privacy may be impossible to achieve — and many of the scientists whose work is driving the DNA revolution believe that the costs of trying will outweigh the benefits. Some are even backing their words with action, by publishing their own genetic data on the internet.
The fantasy that it is possible to prevent unauthorised access to a person’s DNA was exposed in March by Peter Aldhous and Michael Reilly, of New Scientist. Aldhous drank from a glass and gave it to Reilly, who used commercial services to extract his colleague’s genetic profile. It is illegal in Britain to test DNA like this without consent. But the process is so cheap and simple, and the chance of detection so slim, that this offers little practical protection. We leave so much DNA wherever we go that we cannot expect to keep it to ourselves. Legislation might be a deterrent, but it will not prevent unauthorised use.
Even if genetic data could be comprehensively protected, it doesn’t automatically follow that it should be. In making a fetish of DNA, we may be limiting its usefulness. If science is to unravel the medical implications of genetic variations between individuals, to personalise and improve healthcare, it will be necessary to compare the DNA sequences of hundreds of thousands of people. A culture of privacy will hold this back while secreting information that may be less sensitive than it seems.
Certainly, the great and the good of human genetics do not feel that they have much to lose by sharing this data. Some are challenging the privacy assumption by example. When the genetics pioneer Craig Venter became the first person to have all six billion DNA letters of his genome sequenced he published the lot. James Watson, the co-discoverer of the double helix, has done likewise, redacting only the status of a gene linked to Alzheimer’s because he did not want to know the information himself. Kari Stefánsson, of deCODE Genetics, which sells personal DNA tests, allows his customers to compare their data with his.
George Church, of Harvard University, has gone a stage further. The founder of the Personal Genome Project (PGP), and its first participant, has published not only his complete genome sequence but also his medical records and other physical details. He aims to recruit 100,000 volunteers to do the same.
The nine other recruits to the PGP’s pilot phase are all people who know their stuff: they include the psychologist Steven Pinker, John Halamka, Dean of Technology at Harvard Medical School, and Misha Angrist, a geneticist at Duke University in North Carolina. Another 13,000 have joined a waiting list to take part. By putting their information in the public domain, participants hope to exploit “crowd-sourcing”, the notion that allowing anyone to work on a problem will lead to it being solved more swiftly. The phenomenon underlies the success of Linux open-source software and Wikipedia. It could soon be driving genetic discoveries as well.
This will have individual benefits as well as social ones. When Church placed his medical records on the internet he was contacted by a doctor who suggested a useful change to his cholesterol medication. Interpretations of open data are more likely to help than harm.
The prophets of open-source genomics are unruffled because the perceived need for absolute secrecy is also based on a misunderstanding of heritability. While most people think of genetics as a deterministic science, this is rarely so. Most genes that influence health have a probabilistic effect, raising or lowering risk by small margins. Most people with the Alzheimer’s risk gene , for example, will not get Alzheimer’s. It would not only be wrong to judge somebody’s intellect, skills or good health by their ownership of particular genetic variations — it would be profoundly misleading.
All of us carry risky variants: there is no such thing as a perfect genome. That should help to prevent discrimination: if insurers were to exclude everyone with a raised genetic risk of this or that they would rapidly go bankrupt. If the market fails, companies can be banned by law from demanding genetic test results; the US has already done this, and the UK has a voluntary moratorium.
Nobody should be forced to reveal genetic data, which should be published only with the informed consent of the owner; the PGP takes the “informed” part so seriously that it requires participants to pass a genetic literacy test. But it is instructive that so many of those who best understand the mechanics of the genome see so little to fear. Given wider understanding of the probabilistic nature of genetics, their choice to go public is one that many more of us could emulate.