National Post (National Edition)

The DNA danger in the hunt for a killer.

- Colby Cosh National Post ccosh@nationalpo­st.com Twitter.com/ColbyCosh

If I ask you to imagine a genealogis­t, what kind of person comes to mind? Someone older, I bet. Genealogy is one of those interests that tends to sprout later in life, like gardening. Probably you’re picturing an innocuous sort of person in a cardigan and old-fashioned eyeglasses. Someone who looks like they would enjoy haunting a library and using words like “ahnentafel” in conversati­on.

Maybe you should be imagining Clark Kent, that mild-mannered legend of my profession. Genealogis­ts are the latest species of nerd to be endowed with unexpected superpower­s by the magic of computing. They know this already, just as Clark Kent is aware he is Superman, but it has now come to the public’s attention through the arrest of a strong suspect in the case of California’s “Golden State Killer.”

The “Golden State Killer” committed a dozen murders and 50+ rapes up and down the state in the years 19761986. Since he committed his atrocities in clusters, he was regarded by investigat­ors for a long time as being several different people: the “Original Night Stalker,” the “East Area Rapist,” and so on. These media monikers were merged by homicide detectives in the 21st century, when DNA samples from crime scenes showed that several rapist-murderers were one monstrous person. The suspect now in custody, Joseph James DeAngelo, is an ex-cop who fits the non-genetic behavioura­l profile of the killer. (One element of which was “He might be a cop.”)

We are all familiar by now with the notion of DNA as an identifyin­g personal fingerprin­t — a sequence of bits unique to each of us, unless we have a geneticall­y identical sibling. But DeAngelo was identified by a process of detection that may be without precedent, even though genealogis­ts have been discussing the possibilit­y for a long time and were decidedly unastonish­ed by the news.

Police used old samples of the killer’s blood to sequence his DNA, then looked for possible relatives in a large public database of DNA sequences called “GEDmatch.” GEDmatch is a volunteer opensource website for genealogy hobbyists: about 900,000 people have uploaded their DNA sequences in order to find unknown relatives or learn about their deep ancestry.

Police in California have tough statutory rules about using their own databases of convicts’ DNA samples, and commercial DNA-analysis companies would be likely to make a fuss before allowing the sequences of identifiab­le paying customers to be used in such a way. But the GEDmatch data, which is ordinarily uploaded with familytree informatio­n attached, is broadly available to anyone who signs up for membership.

The site’s terms of service — as it has been quick to point out to reporters — do not explicitly outlaw investigat­ive uses. From a legal standpoint, the data was pretty much just lying around in the open.

Police told the San Jose Mercury News that their probe into the GEDmatch database yielded matches with “third and fourth cousins.” This, in turn, allowed them to identify remote ancestors of the unknown killer and create a list of living suspects descended from the right people.

Applying the database approach to the DNA of unidentifi­ed victims has already helped detectives crack a few cold homicide cases. It is a fairly easy matter to throw a lot of John and Jane Does into DNA databases in the hope of finding a close relative, such as a first cousin: from that point to a positive ID it can be a matter of minutes. The situation with the Golden State Killer was different. This was a well-resourced manhunt put together after several homicide and sex assault files turned out to be one unusually awful career of crime. Since the matching relatives were distant from the killer, it required many man-hours to create the full family tree and check out the various branches.

The police still had to get a warrant to collect the actual DNA of their statistica­lly identified suspect-candidate, using discarded items in his household trash. Some people are feeling uncomforta­ble about this novel use of a genetic database, although one wonders if they worry as much about what is going into their garbage bins.

Few GEDmatch users will have imagined having their personal genetic informatio­n roped into a murder hunt, and that raises a possible question of informed consent. However, gene-gazers who chose GEDmatch will have mostly understood it was an open, free-to-explore database, with the freedom being reciprocal. The whole point of this kind of thing is to provide for unanticipa­ted uses by others, acting without advance permission. I doubt any of DeAngelo’s distant cousins are too upset about his arrest, if they are aware of it.

Still, the general principles are worth considerin­g. You own the informatio­n in your own DNA: I see no one arguing against that, nor do I see how one would. But that informatio­n has some pertinence to your kinfolk, near and far, familiar and unsuspecte­d. It could potentiall­y be used, by the law or other institutio­ns, to help identify persons wishing to remain anonymous.

And we are still far from a complete, assured understand­ing of what sort of inferentia­l power DNA sequencing may ultimately enable. Which is to say: no one at all can possibly know the full consequenc­es of sharing one’s DNA. But, then again, most of you reading this are parents, so you knew that already.

 ??  ??

Newspapers in English

Newspapers from Canada