On the (mis-) use of DNA

Fascinating piece in the Wall Street Journal recently about the English/Welsh DNA database. We'll sneak up on it.

The movie GATTACA always bugged me, or rather people's reaction to it did. Whenever I'd mention biometrics, people would say, "Have you seen GATTACA? Isn't that a frightening view of where DNA identification is going?" Well, no, actually. It's a frightening view of where DNA analysis could go, but identification is a different story.

So What?

Let's start with analysis. The concern that's always put forward is that insurers will use DNA to discriminate against customers---to force higher premiums on people with inherited medical conditions (or even to deny them coverage altogether). Also that employers will use it to vet employees.

Misuse of DNA by insurers is an easy concern to allay. If insurers could use certain demographic data (e.g., religion, race), they'd be able to write more profitable policies. This would, of course, constitute something bad (maybe an invasion of privacy, maybe discrimination). But many years ago, Congress passed a law making it illegal for them to use that information (even if they have it) to set premiums. Problem solved.

Employee discrimination is another DNA analysis threat, and this one isn't so easily dismissed. You can pass a law against use of genetic information in recruiting, but enforcement is going to become very, very difficult. When a hip white Apple iDNA analysis kit costs $200 and weighs less than eight ounces, it'll be easy for prospective employers to pick up saliva from your discarded Coke can, run an analysis, and reject you for any number of reasons. Violations of any law against this activity would be rife.

DNA-based identification is an entirely different matter, however. The ability to identify people accurately will aid apprehension of the guilty and exonerate the innocent. (And so we come to the WSJ article.) Look no further than England and Wales for evidence. "Cold" cases are being reopened successfully as much as 35 years after the fact as DNA samples (presciently preserved) are matched to still-living perpetrators.

England and Wales are a leading indicator, I believe. Nearly eight percent of their population has its DNA stored in a central government database---among the highest proportions in the world. How did this happen? Well, anyone in those countries who's arrested (for anything) has a sample taken, and it's retained even if the person is freed without charge. Young children's samples are in the database, too. (I can't imagine that 8 percent of the adult population has been in trouble with the law; there must be some other force at work. Maybe the authorities have a little understanding with the Tooth Fairy.)

There's concern that acquiring the DNA of a non-suspect constitutes treating him like a suspect, and that this is bad. I'm not so sure: It strikes me that a house-to-house search or a traffic checkpoint (as for drunk drivers) treats everyone like a suspect, too, and yet we accept them without a murmur.

The other concern is that the samples could be hacked and then analyzed. Experts assure us that the samples are small subsets of your entire genome and cannot be used to determine genetic predispositions of any sort. If you think these claims are solid, you can relax.

In fact, I have yet to hear a good, solid civil liberties case against DNA identification (as opposed to analysis). I'd welcome one: I, like many others, have an instinctive touch-a-hot-stove reaction to it. But I haven't heard one yet.