Not always a three-way match
Reader Katherine Sutherland is struggling to understand how to work with her autosomal DNA matches. “I compare my matches on 23andMe with each other to try to sort branches if possible,” she wrote. “It seems to me that finding any matches is like a needle in a haystack, and thus if you match with 2 different people, it is likely you all share a common relative.”
The problem is that Katherine’s theory — that her matches would “all share a common relative” — may basically work with two of the three types of DNA testing — but it doesn’t work at all with the third type, the type of testing that Katherine has done with 23andMe.
Let’s review the test types:
YDNA is the kind of DNA contained only in the Y chromosome that only men have, and is passed down from father to son to son with relatively few changes through the generations.1 Because it doesn’t change much from generation to generation, any men who share enough of the same YDNA markers to be considered a match (and particularly if they’re considered a close match) are very likely to be descended from the same direct male-line ancestor somewhere back in time.
Mitochondrial DNA (mtDNA) is the kind of DNA passed down from a mother to all of her children and that only her daughters can pass on to their children.2 Like YDNA, mtDNA doesn’t change much from generation to generation. So any women who share the same mtDNA markers (and particularly if they’ve tested all of their mtDNA and not just parts of it) are very likely to be descended from the same direct female-line ancestor somewhere back in time.
Autosomal DNA is the kind of DNA that you inherit equally from both of your parents: you get 22 autosomes3 (plus one gender-determinative chromosome) from your father and 22 autosomes (plus one gender-determinative chromosome) from your mother, for a total of 23 pairs of chromosomes. So this is a test that works across genders to locate relatives — cousins — from all parts of your family tree.4
The type of test Katherine did at 23andMe is an autosomal DNA test, and therein lies the rub. Because it’s testing so much more of our DNA, and because it’s finding cousins from all parts of our family trees — not just our father’s father’s father’s line like YDNA or our mother’s mother’s mother’s line like mtDNA — the fact that we match two people like AB and CD doesn’t mean AB and CD will match each other and it doesn’t mean that all three of us will share a common ancestor.
In fact, given the way autosomal DNA works, even if AB and CD do match us and each other, we may not all be matching in the same line, descending from the same ancestor.
It may help to understand this a little better to look at the chart above (you can click on it to enlarge it). This is a chromosome browser image from Family Tree DNA and it shows two people who both match me. Under Katherine’s theory, because there is so very much overlap between the segments I share with the person shown in orange and the segments I share with the person shown in blue, it would have to be true that these two people also match each other.
But they don’t. They’re not related to each other at all.
The person in orange is my mother’s brother. All of the orange segments come from the autosomal DNA I received from my mother, and all of her family lines have been in America since at least the late 1700s.
The person in blue is my half-brother. All of the blue segments come from the autosomal DNA I received from the father I share with that brother. Our father was born in Germany, all of his family lines are in Germany as far back as we can find records, and there isn’t even a hint that our father and my mother could have any ancestors in common within the last four or five hundred years at a minimum.
So what you’re seeing here is my mother’s side in orange, and my father’s side in blue. Even though the segments are labeled with the same chromosome number, remember that we all have pairs of chromosomes: two chromosomes 1, two chromosomes 2 and so on. One came from our mothers, one came from our fathers.
When Katherine looks at the same tool on 23andMe, what she’s seeing there — and what we’re seeing here — is how each person matches her. AB could be a cousin in the line she shares with her paternal grandfather. CD could be a cousin in the line she shares with her maternal grandmother. And even if the segments appear to be on the same chromosome, they may be on different chromosomes in that pair of chromosomes, so the results don’t say a thing about whether AB and CD are related to each other.
Only when you look at how AB matches CD (something you can do in your own chromosome browser on 23andMe but have to get your match to do in his or her chromosome browser on Family Tree DNA) can you see if AB and CD are related to each other. And that won’t tell you if whatever line AB and CD have in common with each other is the same one each of them shares with you.
So when working with autosomal DNA, it’s important to keep in mind how very different it is from the other types of DNA. Matching two people with YDNA or mtDNA almost invariably means a three-way match, and a common shared ancestor somewhere back in time. But matching two people with autosomal DNA means you have a lot of work left to do before you can even theorize that there could be a three-way match, and even more before you can say what ancestral line you might share.
- ISOGG Wiki (http://www.isogg.org/wiki), “Y chromosome,” rev. 20 Jul 2013. See also ibid., “Y chromosome DNA tests,” rev. 31 Aug 2013. ↩
- ISOGG Wiki (http://www.isogg.org/wiki), “Mitochondrial DNA,” rev. 20 Jul 2013. See also ibid., “Mitochondrial DNA tests,” rev. 19 Jul 2013. ↩
- “An autosome is any of the numbered chromosomes, as opposed to the sex chromosomes. Humans have 22 pairs of autosomes and one pair of sex chromosomes (the X and Y).” Glossary, Genetics Home Reference, U.S. National Library of Medicine (http://ghr.nlm.nih.gov/glossary=Glossary : accessed 31 Aug 2013), “autosome.” ↩
- See generally Judy G. Russell, “Autosomal DNA testing,” National Genealogical Society Magazine, October-December 2011, 38-43. ↩