Roots

DNA test­ing, some­times a gim­mick, can also be a god­send for ge­neal­o­gists.

Canada's History - - CONTENTS - by Paul Jones Paul Jones, a for­mer pub­lisher, is a writer, a con­sul­tant, and an avid ge­neal­o­gist.

DNA test­ing, some­times a gim­mick, can be a god­send for ge­neal­o­gists.

The “bestest best boy in the land” re­cently had his DNA tested. It turns out he’s a bea­gle mix with a gazil­lion un­dif­fer­en­ti­ated hound va­ri­eties — ba­si­cally a nose on legs.

Is DNA test­ing a gim­mick or the real deal? That’s the ques­tion most of­ten put these days to ex­pe­ri­enced ge­neal­o­gists by mem­bers of the in­ter­ested pub­lic — for ex­am­ple, the readers of this mag­a­zine and their ilk. Even dogs are do­ing it. Should you?

As dis­cussed last time (“Ori­gin sto­ries,” Oc­to­ber-Novem­ber 2016), DNA test­ing to de­ter­mine eth­nic­ity has lim­ited value for any­one who knows the ori­gins of all four grand­par­ents (un­like my dog!).

But if you have gaps in your fam­ily tree within the past five or six gen­er­a­tions, or if you have some other rea­son to con­tact dis­tant cousins, then au­to­so­mal DNA test­ing is for you. And yes, it’s the real deal.

Don’t be put off by the jar­gon. “Au­to­so­mal DNA” sim­ply refers to the ge­netic ma­te­rial in your twenty-two non-sex chro­mo­somes. This is the DNA you in­her­ited, fifty per cent from each parent, and, on av­er­age, twen­ty­five per cent from each grand­par­ent, 12.5 per cent from each great-grand­par­ent, and so on. Ac­tual per­cent­ages fluc­tu­ate, some­times greatly, due to ran­dom­ness in DNA re­com­bi­na­tion in suc­ces­sive gen­er­a­tions.

Other kinds of DNA, which we’ll dis­cuss in a fu­ture col­umn, fol­low dif­fer­ent in­her­i­tance pat­terns and can­not be an­a­lyzed sta­tis­ti­cally in the same way.

Test­ing com­pa­nies look at hun­dreds of thou­sands of mark­ers on your au­to­so­mal DNA, com­pare you with ev­ery other per­son in their data­bases, and ap­ply sta­tis­ti­cal al­go­rithms to es­ti­mate re­la­tion­ships, such as sib­ling, first cousin once re­moved, fourth cousin, what­ever. You, the con­sumer, re­ceive a list of your matches, such as the one il­lus­trated, in­clud­ing es­ti­mated re­la­tion­ships and an­cil­lary in­for­ma­tion that varies from one test­ing com­pany to an­other. The re­sults can be mag­i­cal.

An adoptee iden­ti­fies birth par­ents and half sib­lings she never knew she had (nor they her). A dis­tant cousin sends an an­tique photo of an an­ces­tor whose face you thought was lost to hu­man mem­ory. New rel­a­tives from half­way around the world in­vite you to visit the home­land.

First, though, you need to con­tact your matches to com­pare notes. Don’t be afraid to bug them if you don’t hear back. Fail­ure to re­spond can be a nag­ging prob­lem with this kind of re­search.

Au­to­so­mal test­ing does have other lim­i­ta­tions. While false pos­i­tives are not com­mon if re­sults are in­ter­preted with care, the close­ness of re­la­tion­ships can be over­es­ti­mated be­tween two peo­ple who de­scend from the same cul­tur­ally or ge­o­graph­i­cally iso­lated pop­u­la­tion, such as Ashke­nazi Jews, Ice­landers, New­found­lan­ders, Men­non­ites, and oth­ers. In such cases, the sta­tis­ti­cal al­go­rithms have dif­fi­culty dis­tin­guish­ing be­tween match­ing bits of DNA that de­rive from a com­mon an­ces­tor (“IBD,” or “iden­ti­cal by de­scent”) and those that are a fea­ture of the source pop­u­la­tion (“IBS,” or “iden­ti­cal by state”).

For most of us, false neg­a­tives will be a big­ger con­cern. Once we reach back a few gen­er­a­tions, we in­herit rel­a­tively lit­tle iden­ti­fi­able DNA from each an­ces­tor and none from some. As a con­se­quence, cur­rent au­to­so­mal tests miss about ten per cent of third cousins, fifty per cent of fourth cousins, and ex­po­nen­tially higher pro­por­tions of more re­mote re­la­tions. Even full-genome se­quenc­ing, per­haps an af­ford­able op­tion by 2020, may not im­prove these sta­tis­tics by more than a frac­tion.

How do we fight sta­tis­ti­cal un­cer­tainty? With weight of num­bers. We need to test not only our­selves but also our sib­lings, first cousins, sec­ond cousins, and so on. This way you max­i­mize the odds that some­one close to you will match the lost cousins you’re seek­ing, even if you do not.

Even be­fore test­ing you and your cousins, though, there’s an­other group that should be ac­corded the high­est pri­or­ity: any sur­viv­ing mem­bers of your par­ents’ gen­er­a­tion. Once they’re gone, so is their DNA.

To avoid need­less blun­ders and to answer ques­tions you may have, do some read­ing be­fore pur­chas­ing your first test.

The In­ter­na­tional So­ci­ety of Ge­netic Ge­neal­ogy pro­vides an in­for­ma­tion-rich wiki. Check it out here: http://isogg.org/wiki/ Ge­net­ic_­ge­neal­ogy. Next time: some case stud­ies that il­lus­trate the power of au­to­so­mal DNA test­ing — among hu­mans.

The au­thor’s DNA test re­sults from Ances­tryDNA iden­ti­fied thou­sands of po­ten­tial rel­a­tives. Shown here are sev­eral that can be ex­plored in de­tail.

Newspapers in English

Newspapers from Canada

© PressReader. All rights reserved.