Every now and then, when I’m doing DNA reports, I run across the perfect example of a DNA phenomenon. Today, it was a mitochondrial DNA mutation in motion. Let’s take a look at what happened, how it was discovered and what it means.
I was contacted a few weeks ago by someone I had been working with on another project. This woman, we’ll call her June, was concerned because both she and her maternal first cousin, Doris, had both taken mitochondrial DNA tests at Family Tree DNA and they didn’t match each other. I took a look, of course, and sure enough, at the HVR1 level, there was one mutation difference, at location T16271C.
This was particularly interesting, because at the first cousin level, these women shared a maternal grandmother, which means that either June’s mother or Doris’s mother had had a mutation in their mitochondrial DNA, or June or Doris did. June asked me how she could tell who had the mutation.
I asked if either June or Doris had siblings. June had a brother, John, so she ordered a kit for John. If John matched June, then their mother is the one who had the mutation. If John matched Doris, then June herself had the mutation.
How do I know this, that the mutation didn’t happen in Doris or her mother? Because the mutation is not “normal” and is listed in the RSRS values in the “extra mutations.”
Furthermore, Doris, who did not carry the extra mutation, had 13,204 matches at the HVR1 level (haplogroup H), where June who did carry the extra mutation only had 41. Clearly to be useful, genealogically, this test would need to be expanded to the full sequence level.
So June’s brother, John, tested and he matched his sister June, telling us that their mother carried this mutation, and gave it to both of her children. So the mutation occurred between June’s mother and June’s grandmother.
Are These Matches Valid?
June asked me if her matches were valid.
That’s a tough question to answer, because convergence has occurred.
So let me answer this in two ways.
The matches are technically accurate. This means that indeed she matches all 41 of the people that the matching routine shows as her exact HVR1 matches. So in that way, those matches are accurate, but they aren’t valid or meaningful for genealogy.
They aren’t useful, because we know, beyond a doubt that these matches are not related to her in a very long time, probably back into prehistory, because the reason she matches them at the HVR1 level is because she just happened to have the same mutation that all 41 of them carry. Carrying the same mutation does NOT absolutely mean you share a common ancestor who carried that mutation. Mutations can occur at any time, and if a mutation happens at this location in the mitochondrial DNA, there is a 1 in 3 chance the person who has the mutation will have the same value as you, since there are only 4 choices, T, A, C, and G, to begin with. This is what we call convergence, and you’ve just seen it happen. People match each other, but because they happened to have the same spontaneous mutation, not because they share a common ancestor who had that mutation. Most of the time, we don’t know whether we are looking at real matches or matches by convergence, but this time, we know for sure, because we can prove that June’s grandmother did not have the mutation, because June’s first cousin, Doris, does not.
So, if June’s HVR1 results aren’t useful to her, whose are? That’s easy, her cousin Doris’s results are representative of the mitochondrial DNA of their mutual grandmother, so Doris’s matches are actually June and John’s ancestral matches as well.
Could There Be A Fly in the Ointment?
Not matching someone you thought you should match is unsettling. Could we test someone else to be absolutely positive we’re not dealing with a back mutation?
Certainly, if grandmother had another female child who had children, or if grandmother has a living male child, they can be tested too. The test on the third child would positively confirm grandmother’s mitochondrial DNA values.
Could we prove positively that the first cousins are actually first cousins, to remove any nagging doubt?
Certainly, using the Family Finder test.
I receive a small contribution when you click on some of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.
Thank you so much.
DNA Purchases and Free Transfers
- Family Tree DNA
- MyHeritage DNA only
- MyHeritage DNA plus Health
- MyHeritage FREE DNA file upload
- 23andMe Ancestry
- 23andMe Ancestry Plus Health
- Legacy Tree Genealogists for genealogy research
Excellent and very helpful posting, Roberta! Goes in the category of “I shoulda thought of that”!
Sometime could you walk us through the results from FF that would confirm that they are cousins, or not?
That’s easy. Given how closely they should be related, they will either match or not on Family Finder. If they match, they confirmed. If they don’t, then there is an adoption someplace. First cousins should share about 12.5% of their DNA. Here’s a page about relatedness and percentages. http://www.isogg.org/wiki/Autosomal_DNA_statistics
Thanks. When I get results from FF they will say 2nd to 4th, does this mean that the range we share 3.25% or is it a range from 3.25 to .81 % ?
It says 2nd to 4th. You can see an example of the new format results here. http://dna-explained.com/2013/10/03/family-tree-dna-updates-family-finder-and-adds-triangulation/
Learning the ropes of DNA more with each blog! Thanks for sharing your knowledge!
Have any of June’s HVR1 matches taken the Full Mitochondrial Sequence test?
Yes, several had. They were scattered between H7 (mostly), H43 and H69. When her brother, John, tested, he was H43. At the HVR2 level, June only has 4 matches, and only one of those has the FMS, which is her brother.
It’s really interesting that a parallel mutation occurred even within the same haplogroup. 16271 is not one of the most notorious hotspots (Table S3 in Soares’ Mitochondrial Molecular Clock paper):
Yes, it is. Also, just to tie up loose ends, I heard from June today and she and Doris did order the autosomal test and they do match there, so there is no question about parentage. Just for good measure, she also tested her brother too.
You know, this makes me wonder how many parallel mutations there really are. Unless something like this happens, we have no way of recognizing them. Food for thought.
I just recently did the full sequence mtDNA and was disappointed to not have a single match. Closest was off by 2. I thought no one from my branch had tested and now I see I may have my own private branch and need not wait for someone else with the same configuration to test. I already got my mother’s maternal first cousin to do a Family Finder test. Could I now get just an HVR1 test for her or will I have to get full sequence?
The only way to know if she matches entirely is to purchase the full sequence test. It’s a judgment/money call on your part.
I have these mutations: 309.1C, 315.1C, 522.1A, 522.2C, 522.3A, 522.4C, G7805A, T10166C,
C10757T, T16140C and C16260T. My question is how can a person with no training tell which of these mutation are rare. I see a lot of different haplogroups with 309.1C, 315.1C and some of the others. I have never seen a person with 16260T. Also is it the number or letter that causes it to be a mutation.
That’s part of what I do in the reports I write for people. But if you’ve had your full sequence done, which I see you have, you can look at the results and at the top it says “extra mutations.” Those are the ones that are outside of the “norm” although they are not always “rare” per se. The number is the location and the first letter is the normal value and the second letter is the value that you have at that location.
Thank you, Roberta. That makes sense. I just had it in my mind that all mutations were rare.
Mutations are forever, so the ones that occurred first, a long time ago, are carried by lots of people and those are the mutations that define the haplogroups. Mutations that happened more recently and are carried by fewer people are subhaplogroups. Mutations continue to happen all of the time.
Have; 315.1C 522.1A 522.2C mysellf.
I’m so glad I found this blog. I have a heteroplasmy and since I’m so new to genetic genealogy I really didn’t understand it at all. I got the full testing done on myself only (HVR1, HVR2, and full coding region) and I match a handful of people exactly (with 0 genetic distance) at H7c1 Haplogroup. But this heteroplasmy concerns me. I manage my brother’s kit, he had Y-DNA done. After reading this blog, I may have an HVR1 and HVR2 done on him to see if he has the heteroplasmy as well. I’m assuming the getting the full coding sequence would be pointless right, since I already had that done?
It depends on where your heteroplasmy occurs. If it is in the coding region, then testing your brother on the HVR1 and HVR2 only won’t tell you if he carries that mutation too.
Pingback: DNAeXplain Archives – Intermediate DNA Articles | DNAeXplained – Genetic Genealogy
what does the d1 after U6 mean?
Each letter and number after U is a smaller branch.