The chances of DNA

Speaking of science

If there’s anything a genealogist learns after really getting into family history, it’s the need for precision.

We don’t say that John Smith Sr. on that 1780 tax list is the father of John Smith Jr. on that tax list just because of the letters after those names. Because we’ve learned that senior and junior back then meant nothing more than older and younger. They might have been father and son, but that’s not why the terms were used.

And we don’t say that two people listed one after the other on a census were living next door to each other just because of the order of their names on the enumeration page. We’ve learned that some census records are alphabetical, and some census takers took some very odd routes when calling on the people in their territories.

And when we venture into the realm of genetic genealogy, we need to bring that same precision to bear on our work — and it sure would be nice if we could expect the same precision from those in the media who tell us about developments in genetic genealogy.

Alas, the Richard III story of last week showed us that we can’t rely even on science reporters — or their photo caption writers — or their editors — to get it straight.

Case in point: an article appearing in the online edition of The Guardian, a British newspaper, last Monday under the headline “Richard III skeleton raises bone of contention over DNA evidence.”1

Let’s deal with the photo caption writer and the editor first. The article was illustrated with a photo of Michael Ibsen, the Canadian-born craftsman who is the 17th great grand-nephew of the last of the Plantagenet Kings. And the caption read: “Michael Ibsen, seen here using an oral swab to give a DNA sample, is believed to be a direct descendant of the king.”

Um…

No.

I hate to break it to The Guardian, but the term descendant has a precise meaning, and adding direct to the term — while somewhat redundant — makes it very precise indeed. You’re a direct descendant only if you are, in fact, “one descended from another.”2 In other words, for Michael Ibsen to be a direct descendant of Richard III, then Richard III would have to be his ancestor — “one from whom a person is descended and who is usually more remote in the line of descent than a grandparent.”3

Only one hitch.

Richard III had no known surviving children, certainly not by his queen consort. His one known son, Edward of Middleham, died while Richard was King.4

And nobody is claiming — except, apparently, The Guardian‘s photo caption writer and editor — that Michael is descended from Richard III. The genealogy puts Ibsen into the line of descent from Richard III’s mother, Cecily Neville, through Richard’s sister Anne of York, Duchess of Exeter, 1439-1476.5

So here’s a lesson for The Guardian — a nephew is not a descendant, an uncle is not an ancestor.

Moving along to the science writer… The article asserts that “Even if there is good circumstantial evidence to suggest two people are related, they might still share the same mtDNA by chance.”

Um…

No.

Not unless you’re planning on revising everything we know about mitochondrial DNA — mtDNA — the kind of DNA that a mother passes to all of her children, but that only a daughter can pass to the next generation.6

Here’s a short course in what this non-scientist understands of the science (and I’m sure if I’m wrong somebody who is a scientist will jump in and tell me so!).

Somewhere back in time, probably 100,000 to 200,000 years ago, a woman lived somewhere on earth. Because of chance — luck on her part, bad luck on the part of other women living at the time, disease, food supply and who all knows what else — her descendants managed to be fruitful and multiply … and the direct female lines of all the others died out somewhere down the line.

That woman is what’s known as the Mitochondrial Eve (“mtEve”): the most-recent common ancestor of all people alive today but only when counted along the maternal (matrilineal, direct female descent) line.7

Her mitochondrial haplogroup — her genetic population group — was L. Over time, her daughters’ daughters’ daughters’ mtDNA changed a little here and a little there until there were enough differences to recognize different haplogroups. The first was probably what’s known as haplogroup N, around 75,000 years ago; the most recent, haplogroup V, around 15,000 years ago.8

Anyone whose haplogroup is N shares a common matrilineal ancestor. Let’s call her mtEve-N. And anybody in haplogroup V shares a different common matrilineal ancestor — mtEve-V. Now understand: all these folks are still related to mitochondrial Eve. Not very closely, for sure, but you wouldn’t be here today if you didn’t descend from that woman. But all the V folks descend from a different one of mtEve’s daughters than all the N folks.

Got that? Good.

Within each of these big haplogroups are what are called subclades — and those represent the few additional differences that help distinguish one person in a haplogroup from someone else in the same haplogroup. You can think of that as descending from one of mtEve’s granddaughters.

Those subclades get pretty darned specific, and may have shown up with enough clarity to be recognized as recently as just a couple of thousand years ago. Michael Ibsen, Richard III’s 17th great grand-nephew, is J1c2c,9 a subclade that appeared about 4500 years ago.10

So mtEve had a daughter mtEve-J. And somewhere down the road, she had a daughter mtEve-J1. And somewhere further down the road that daughter had mtEve-J1c and so on. And everybody who is J1c2c descends by definition from whichever of mtEve’s however-many-great granddaughters first passed the defining characteristics for J1c2c on to her daughters.

So the bottom line here is that we don’t roll the DNA dice with everybody having a “chance” of ending up with the “same mtDNA.” By definition, if you have the same mtDNA as someone else, you’re related. The only question is how closely you’re related, not whether you’re related.

Even the very best, most sophisticated mtDNA test available only gives you a 90% chance of being related within the last 16 generations — about 400 years — if you’re an exact match. Being off by even one step could put your common ancestor back as much as 120 generations — 3,000 years.11

So you’re still related. It just may be that you’re related back in the mists of time, far beyond where there’s even a hope and a prayer of being able to document the relationship on paper.

Now in fairness it’s clear that the writer was trying to convey the concept we’re talking about here. He does say at one point that having the same mitochondrial type “would not necessarily mean they shared a common ancestor at the time of Richard III.”

The emphasis, of course, is mine. And the point is, it should have been his.


 
SOURCES

  1. Alok Jha, “Richard III skeleton raises bone of contention over DNA evidence,” The Guardian, online edition, posted 4 Feb 2013 (http://www.guardian.co.uk/ : accessed 9 Feb 2013).
  2. Merriam-Webster Online Dictionary (http://www.m-w.com : accessed 9 Feb 2013), “descendant.”
  3. Ibid., “ancestor.”
  4. See Wikipedia (http://www.wikipedia.com), “Edward of Middleham,” rev. 6 Feb 2013.
  5. See Judy G. Russell, “Rewriting history through DNA,” The Legal Genealogist, posted 3 Feb 2013 (http://www.legalgenealogist.com/blog : accessed 9 Feb 2013).
  6. See ISOGG Wiki (http://www.isogg.org/wiki), “Mitochondrial DNA,” rev. 30 Jul 2010.
  7. See ISOGG Wiki (http://www.isogg.org/wiki), “Mitochondrial Eve,” rev. 5 Jul 2010. Also, Krishna Kunchithapadam, “What, if anything, is a Mitochondrial Eve?,” Talkorigins.org (http://www.talkorigins.org : accessed 9 Feb 2013).
  8. See Wikipedia (http://www.wikipedia.com), “Human mitochondrial DNA haplogroup,” rev. 16 Oct 2012.
  9. Rachel Ehrenberg, “A king’s final hours, told by his mortal remains,” Science News, web edition, 6 Feb 2013 (http://www.sciencenews.org/ : accessed 9 Feb 2013).
  10. See James Lick, “Analyzing the mtDNA of descendants of Richard III’s mother,” posted 8 Feb 2013 (http://blog.jameslick.com/ : accessed 9 Feb 2013).
  11. How do I tell how closely I am related to an mtDNA match?,” Frequently Asked Questions, Family Tree DNA (http://www.familytreedna.com/faq/ : accessed 9 Feb 2013).
Print Friendly
This entry was posted in DNA. Bookmark the permalink.

16 Responses to The chances of DNA

  1. michael barrett says:

    Terrific article.

  2. Elizabeth says:

    I have encountered a lot of people who use the term “descendant” or “ancestor” when what they mean is that they are related to someone but collaterally only. Sometimes I try to correct them, but it is hard to get through sometimes.

    • Judy G. Russell says:

      The collaterals can be hard to reference, sometimes, Eilzabeth — but that’s sure no excuse for misusing the terms! Maybe we just need better terms for them.

  3. David Condit says:

    Thanks for a good article. I have read quite a bit about my mtDNA matching level but had missed the chart you reference in footnote 11. It makes sense now that of the 14 exact mtFullSequence matches I have, my genealogy back to my GGGrandmother, 4 generations, doesn’t match any of the matches. Any of them could still be within another 12 generations, 300 years. Brick walls a son-of-a-gun to break down. More digging ahead.

    • Judy G. Russell says:

      Tell me about it, David. I have a bunch of full sequence matches that went from 0 steps to 1 step in the most recent re-evaluation of mtDNA results — which means my chances of ever finding the common ancestor dropped like a rock. Sigh…

  4. You did a great job explaining this topic Judy. I sometimes think we spend an awful lot of time as a community fixing intentional or unintentional “misrepresenations” of genetic results.

    • Judy G. Russell says:

      Coming from you, that’s high praise indeed, Roberta! Thank you!

      And it isn’t just misrepresentations — sometimes it’s just misunderstandings. Makes you wish people would spend some time fact-checking and asking questions… and fact-checking and asking more questions… and…

  5. JoF says:

    Today’s journalists are in such a hurry to print/post something that they generally don’t spend enough time understanding the subject they’re supposed to be talking about. Most higher class media outlets have “fact checkers” but they usually know even less about a particular topic so they can’t catch most of the mistakes that require interpretation of info. I have no idea how to get people to slow down and think. Actually, I have no idea how to get people to slow down OR think!

  6. The article appeared in The Guardian, a paper so famous for typos and other examples of sloppy editorial quality control that it was long ago renamed The Grauniad by the satirical magazine Private Eye. My favourite Grauniad typo of all time was a a review of the opera “Doris Godunov”. More recently they published a bar chart showing a £5 billion profit by British bank RBS as a £44.7 billion loss.

  7. CeCe Moore says:

    Oh boy, here we go again: aolbroadband.welcomescreen.aol.co.uk/video/public-offered-dna-testing-to-find-richard-iiis-descendants/517669315/. Take a look at the headline.

  8. Karla says:

    Not so long ago Ancestry sent me an email (I’m sure everyone else here got it too) which asked, among other things, whether my ancestors had had children. There was much merriment on Facebook when I posted that. Yeah, I’m the product of spontaneous generation and so are all of my cousins. As Hamlet observed, “If the sun breed maggots in a dead dog…”

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>