Ances­try sites could soon ex­pose nearly any­one’s iden­tity, re­searchers say

Tehran Times - - SCIENCE -

Ge­netic test­ing has helped plenty of peo­ple gain in­sight into their ances­try, and some ser­vices even help users find their long-lost rel­a­tives. But a new study pub­lished this week in Sci­ence sug­gests that the in­for­ma­tion up­loaded to these ser­vices can be used to fig­ure out your iden­tity, re­gard­less of whether you vol­un­teered your DNA in the first place.

Ear­lier this year, Sacra­mento po­lice ar­rested 72-year-old Joseph James DeAn­gelo for a wave of rapes and mur­ders al­legedly com­mit­ted by DeAn­gelo in the 1970s and 1980s. And they claimed to have iden­ti­fied DeAn­gelo with the help of ge­neal­ogy data­bases.

Tra­di­tional foren­sic in­ves­ti­ga­tion re­lies on match­ing cer­tain snip­pets of DNA, called short tan­dem re­peats, to a po­ten­tial sus­pect. But these snip­pets only al­low po­lice to iden­tify a per­son or their close rel­a­tives in a heav­ily reg­u­lated data­base. Thanks to new tech­nol­ogy, the in­ves­ti­ga­tors in the Golden State Killer case iso­lated the ge­netic ma­te­rial that’s now col­lected by con­sumer ge­netic test­ing com­pa­nies from the sus­pected killer’s DNA left be­hind at a crime scene.

His­tor­i­cal records

This in­for­ma­tion, cou­pled with other his­tor­i­cal records, such as news­pa­per obit­u­ar­ies, helped in­ves­ti­ga­tors cre­ate a fam­ily tree of the sus­pect’s an­ces­tors and other rel­a­tives. Af­ter ze­ro­ing in on po­ten­tial sus­pects, in­clud­ing DeAn­gelo, the in­ves­ti­ga­tors col­lected a fresh DNA sam­ple from DeAn­gelo—one that matched the crime scene DNA per­fectly.

But while the de­tec­tive work used to un­cover DeAn­gelo’s al­leged crimes was cer­tainly clever, some ex­perts in ge­netic pri­vacy have been wor­ried about the grander im­pli­ca­tions of this method.

Er­lich and his team wanted to see how easy it would be in gen­eral to use the method to find some­one’s iden­tity by re­ly­ing on the DNA of dis­tant and pos­si­bly un­known fam­ily mem­bers. So they looked at more than 1.2 mil­lion anony­mous peo­ple who had got­ten test­ing from MyHerit- age, and specif­i­cally ex­cluded any­one who had im­me­di­ate fam­ily mem­bers also in the data­base.

They found that more than half of these peo­ple had dis­tant rel­a­tives — mean­ing third cousins or fur­ther — who could be spot­ted in their searches. For peo­ple of Eu­ro­pean de­scent, who made up 75 per­cent of the sam­ple, the hit rate was closer to 60 per­cent. And for about 15 per­cent of the to­tal sam­ple, the au­thors were also able to find a sec­ond cousin.

Much like the Golden State in­ves­ti­ga­tors, the team found they could trace back some­one’s iden­tity in the data­base with rel­a­tive ease by us­ing these dis­tant rel­a­tives and other de­mo­graphic but not overly spe­cific in­for­ma­tion, such as the tar­get’s age or pos­si­ble state res­i­dence.

The re­searchers say it will take only about 2 per­cent of an adult pop­u­la­tion hav­ing their DNA pro­filed in a data­base be­fore it be­comes the­o­ret­i­cally pos­si­ble to trace any per­son’s dis­tant rel­a­tives — and there­fore, to un­cover their iden­tity.

Newspapers in English

Newspapers from Iran

© PressReader. All rights reserved.