Friday, December 31, 2010

My Full Mitochondrial Sequence, Heteroplasmy and Comparing mtDNA designations at FTDNA and 23andMe

I received my Full Mitochondrial Sequence (FMS) results from FTDNA last week. FTDNA lists my mtDNA haplogroup as U5b1, while 23andMe reports it as U5b1b2. I am fortunate in that I have four exact matches in the FTDNA database, including a close relative of Dave Dowell of Dr D Digs Up Ancestors.
Often mtDNA testing is not especially meaningful genealogically, but in my case it is more so because this subclade does seem to be predominantly Finnish, which matches my paper trail perfectly. My matrilineal great grandmother Mathilda Huhtala (1878-1950)  immigrated to the United States from the South Ostrobothnia area of Finland in 1902. Thanks to my wonderful Finnish friends, including one of my early high resolution mtDNA matches, I have a pretty solid paper trail back to Margaretta Mattsdotter Kauppi Storstraka Taipale born 21 July 1712 in Härmä, Lansi-Suomen Laani, Finland. Margaretta is my sixth great grandmother and from the same area of Finland as Mathilda.
I have ten High Resolution (HRV1 + HRV2) mtDNA matches. Eight of these are Finnish and two suspect Finnish ancestry. Eight of the ten have done the FMS test and, of those, I match four. I'd say that I got pretty lucky since so many people have no matches to their FMS in the database so far. There seems to be quite a lot of interest in DNA testing among my subclade. Perhaps this is because Finns have an unusually high participation rate in DNA testing overall compared to other countries outside of the US.

I asked Dr. Ann Turner to review my results. She was especially interested in comparing the mtDNA sequences recorded by FTDNA to my 23andMe raw data. She found 34 no-calls in my 23andMe mtDNA data, including one at 6260 where, according to FTDNA, I have a heteroplasmy (notated by a "R"). She checked my mother's 23andMe data and discovered that she also showed a no-call there. Dr. Turner noted that no-calls are usually not significant, but in this case, she theorized that my mother may also have a heteroplasmy at that location, resulting in the shared no-call.

(Note on heteroplasmy- Funny how a concept becomes so much clearer when you learn that it affects you personally.)

From my custom report prepared by Dr. Turner:
You have the mutations that define superhaplogroup UK (11467G, 12308G, and 12372A)...U5 splits off with 3197C, 9477A, and 13617. U5 in turn is divided into smaller groups, with 7768G and 14182C defining U5b, 5656G defining U5b1, and 12618A defining U5b1b. Somewhat unusually, a mutation in HVR2, 217C, defines the deepest level, U5b1b2.
...the most recent mutations (are) the ones that have not been observed often enough to be formally recognized as a subhaplogroup. Population geneticists sometimes call these “private” mutations, using the word in the sense of “confined to particular persons or groups.” Private mutations may in fact be very recent or quite old, but regardless of their absolute age, they are the ones that narrow down the pool of matrilineal relatives to your closest cousins. Not everyone will even have a private mutation. Your private mutation is 6260R.

Interestingly at 23andMe where I have tested ten close matrilineal relatives so far, one of my first cousin's mtDNA haplogroup is designated as U5b, while the rest of us are U5b1b2. Dr. Turner investigated this difference by checking my cousin's raw data for no-calls. As she suspected, my cousin has a no-call at a defining SNP, thus causing the different designation. She also noted that this cousin does not share my heteroplasmy at 6260, adding, "That's a good illustration of how heteroplasmy levels can vary within a family."

Since Dr. Turner confirmed that there are no known medical implications to my mtDNA results, I am sharing them here for anyone who is interested:
HVR1 differences from CRS
16189C, 16192T, 16270T

HVR2 differences from CRS
73G, 150T, 217C, 263G, 309.1C, 315.1C, 573.1C, 573.2C

CR differences from CRS
750G, 1438G, 2706G, 3197C, 4769G, 5656G, 6260R, 7028T, 7768G, 8860G, 9477A, 11467G, 11719A, 12308G, 12372A, 12618A, 13617C, 14182C, 14766T, 15326G

I am in the process of submitting my results to GenBank. It is a bit complicated, so I am including links and helpful instructions I received from Ian below.

If you are looking for your FASTA file to submit to Ian Logan for GenBank, it has been moved. Here is what you do:
Login to your FTDNA account.
Go to the mtDNA -> Print Certificate/Report/Data page.
Look for where it says:
mtDNA Results
Click on the FASTA File link to download your FASTA file. Further submission instructions are here.

[Disclosure - My company StudioINTV has an existing production agreement with FTDNA that has no bearing on the opinions I express. I receive no other compensation in relation to any of the companies or products referenced in my blog.]


  1. That is interesting and I am glad you have some FMS matches. I have no high resolution (HVR1+HVR2) matches unfortunately so I would not have any FMS matches either. My closest match that I know of (based on the V project listing as FTDNA only provides exact matches in our profiles unfortunately) lists her most distant direct maternal ancestor's birthplace as Germany (mine is Swiss-German, close enough).

  2. What is a "no call"? I have never seen this term before. I had a full sequence done, also asked Ann Turner for a report--very interesting.

  3. A no call is a spot in my genetic scan where the DNA "reading" equipment did not get a result. They were unable to tell me what value (A, T, C, G) I have at that specific gene.

  4. Hi, it appears that I am "U5b1b2 - predicted from HVR 1 & 2" as I did not take the mtFullSequence @ FTDNA. My mother's line is also of Finnish ancestry; I am kit #226977.

  5. How do you find out the medical implications of your differences? Thanks!

    1. You can either research your mutations by reading academic papers or you can order a report from Dr. Ann Turner (used to be about $60) and she will research for you.