Working with Mitochondrial DNA Results

I recently received this query.  It made me smile.  I receive a lot of e-mails similar to this.

“I always thought I was an intelligent woman but I am absolutely stymied on how to proceed with the DNA results from Family Tree DNA.

My mtDNA has 65 pages of HVR1 and HVR2 matches.  What does this mean?  Is there somewhere I can find a step by step procedural on how to proceed after getting DNA testing and how to apply it to genealogical research?  What should I do first?”

These are all good questions.  Unfortunately, mitochondrial DNA is more difficult to use genealogically because of the name changes in every generation.  What we really need is a big centralized data base someplace where we an enter our mitochondrial line names to see if anyone in that line has tested, but that data base doesn’t exist.  That data base would provide the same type of function for mitochondrial DNA that surname projects do for paternal lines.  If you want to know if your Johnson Y-line has tested, you just go and look in the Johnson project.  You can’t do that with mitochondrial DNA, so it’s everyone for themselves.  This means we need to be sure we do everything we can to help ourselves which gives us the best odds for success.  My Dad used to say that luck was 99% elbow grease!

What this lady didn’t say was whether or not she had tested to the full sequence level or just to the HVR1+HVR2 level.  From what she did say, I’m betting that she is haplogroup H, the most common haplogroup in Europe, carried by about 50% of the people, and that she did not get her full sequence tested.  If you are haplogroup H, and you have any HVR2 matches at all, the only reasonable way to sort out who is related in a genealogical timeframe is to take the full sequence test.  Otherwise, trying to work with 65 pages of matches is kind of like swatting at flies.

However, not everyone is reasonable, and maybe few of the people you match have taken the full sequence test.  Even if you have taken the full sequence test, there is nothing you can do about those who haven’t and you’d like to be able to use the results you have to see if anyone is a genealogical match to you.

In my experience, a short, less than one page, e-mail sent to your matches with some very specific information is the best way to garner a response.  No one wants to have to sort through a rambling e-mail, so organize it concisely so that the person receiving the e-mail can immediately see the relevant information.  What you’re hoping is that they will take a look and say “Hey, I know that ancestor,” or maybe “My ancestor is from that location too.”

In my case, I have 222 HVR1+HVR2 matches, but no full sequence matches.  Many of my HVR1+HVR2 matches have taken the full sequence test, and I know they are NOT matches to me at the full sequence level, so I don’t need to send them the e-mail.  They’ve been eliminated.

On the list above, there are only 4 people are showing as matches who did not take the full sequence (FMS) test, so they will receive the following e-mail message with relevant information about each generational ancestor, including name, birth and death years and locations, spouses name and where they lived if it wasn’t where they were born or died:

Hello <their name>,

At Family Tree DNA, you and I show as mitochondrial DNA matches at the HVR1+HVR2 level.  This means that someplace back in time, we shared a common ancestor.  I have tested at the full sequence level as well, so if you were to upgrade we could confirm that we continue to match, and in a genealogically relevant timeframe, or we would know that we don’t, and we can discontinue our search because our common ancestor was hundreds to thousands of years ago. 

I’m hopeful that perhaps we can identify our common ancestor, or perhaps just a common location. 

My ancestors on my maternal, mitochondrial line, are as follows: 

  • Me
  • My mother
  • My mother’s mother – Edith Barbara Lore born 1888 Indianapolis, Indiana, died 1960 Rochester, Indiana, married to John Ferverda, lived in Silver Lake, Indiana
  • Edith’s mother – Nora Kirsch born 1866 Dearborn County, Indiana died 1949 Lockport, NY, married Curtis Benjamin Lore, lived in Rushville and Wabash, Indiana
  • Nora’s mother – Barbara Drechsel (also spelled Drexler) born 1848 Goppmannsbuhl, Bayern, Germany, died 1930 Wabash, Indiana, married Jacob Kirsch, lived in Aurora, Indiana
  • Barbara’s mother – Barbara Mehlheimer born 1823 Goppsmannbuhl, Bayern, Germany, died 1906 Aurora, Dearborn County, Indiana, married George Drechsel
  • Barbara’s mother – Elisabetha Mehlheimer, born about 1800 probably in Goppmannsbuhl, Germany, died before 1851

Goppmannsbuhl is a small village outside Speichersdorf, close to Bayreuth and the Czech border, not too far from Nuremburg in Germany.  You can see the location on the Google map below.

https://maps.google.com/maps?f=q&source=s_q&hl=en&geocode=&q=G%C3%B6ppmannsb%C3%BChl,+Speichersdorf,+Germany&aq=1&oq=goppmann&sll=37.0625,-95.677068&sspn=43.25835,101.513672&vpsrc=0&t=h&ie=UTF8&split=0&hq=&hnear=G%C3%B6ppmannsb%C3%BChl,+95469+Speichersdorf,+Oberfranken,+Bayern,+Germany&z=16&iwloc=A

Do any of these families or locations look familiar to you?  Sometimes even if we can’t find a common ancestor, we discover that our ancestors were from the same general area.  Where does your mitochondrial DNA line come from?

Roberta Estes

You’ll note that I did three things here.  I mentioned major landmarks nearby that might be familiar to people, including the Czech border.  At least one of my matches is from Czech Republic and if I don’t mention how close my ancestors lived to that border, people from there will see Germany and dismiss any possible match.  I also included a map that people can click on.  Sometimes that helps.  Lastly, I clearly show the mitochondrial path so that if they don’t understand how that works, they can use my example – me to mother to her mother, etc.  You’d be amazed at how many people are unclear about this.

Oh, and one last thing, I don’t include the information about my mother.  She is deceased, but they just don’t need that.

While we are waiting for replies, we can upload our  information to Mitosearch and continue our search there.  You can do that by clicking on the “upload to Mitosearch” link on the bottom of your Matches page at Family Tree DNA, or you can enter your results manually if you tested elsewhere.  We can also upload our GEDCOM files to both locations.  That makes it easier for potential matches to see if there is anything relevant.

For the most part, you’ll find the same people at Mitosearch that you’ll find at Family Tree DNA.  There are a few exceptions, but generally, people who test elsewhere either don’t know about Mitosearch or aren’t motivated to add their information there.  In some cases, I think people get discouraged and don’t do what they can to find out about their matches.  Case in point is that I seldom receive query e-mails about potential matches, and no, it’s not because I send them an e-mail immediately.  You know, the cobblers kids and no shoes:)

The great thing about Mitosearch is that you can click on the User ID to see information provided by your matches when they were uploading or entering their information.  There are various search criteria. I always select the option to compare me only to those who have tested both the HVR1 and HVR2 regions, and only show me people who match in both.

Here’s my entry.

Unfortunately, in Y-search, Mitosearch’s companion data base, you can search by surname, but Mitosearch doesn’t contain that feature.  Not only does YSearch give you matches, but it also provides you with a list of pedigree charts that the name appears in.  For names like Smith, this probably isn’t terribly useful, but for Mehlheimer, one match would be a goldmine.

I click through the User Ids of all my exact matches.  An exact match is when both “differences” columns equal zero.

There are two other alternatives as well to find matches, although neither is wonderful.

One is the Sorenson data base, www.smgf.org.  The problem with this data base is that you can’t contact any of the matches, so aside from the genealogy that they provided and is shown with their DNA information, you’re dead in the water.  Recently, since being acquired by Ancestry, this data base has been experiencing a significant amount of down time when it is unavailable.

The other alternative is Ancestry.com.  You can enter your results there and compare to their data base.  However, before you do, read this blog about what to expect and the problems with Ancestry mitochondrial matches.  Ironically, they provide no genealogical information unless the participant has connected their tree.  Furthermore, you must contact your matches there though Ancestry’s message system and very few people reply.

If you want to know more about your mitochondrial DNA and the secrets it holds for you, you can purchase the Personalized DNA Report on your personal page at Family Tree DNA under the orange “Order an Upgrade” button or on my webpage at http://www.dnaxplain.com/shop/features.aspx.  I write these reports regardless of whether you purchase through Family Tree DNA or my website, and it’s the same report in either case.

Hopefully, by now, your matches that you sent e-mails to will be replying!  Happy mito hunting.

34 thoughts on “Working with Mitochondrial DNA Results

  1. Hi Roberta

    Thanks for your blogs. You’ve increased my understanding of genetic genealogy three-fold and saved me innumerable hours of wasted time now that I know what to look at, and when.

    Could you clarify the comment in this column about the DNA reports – do you write this for FTdna or as an independent contractor/individual business owner?

    Lastly, one day I’d like to see a blog from you on your wish list for genetic genealogy – what are the companies missing in terms of what they provide to us consumer genetic genealogist – that can be achieved in the short term (say 3-5 years) and what is your longer term, bigger picture wish list (big, and bodacious without thought to time/cost/reality).

    Thanks for all you do.

    Hausa

    • Hi Hausa,

      I’m an independent contractor and own my own company which I started in 2004. It’s hard to believe it’s been 8 years now. Family Tree DNA offers my Personal DNA Reports to their clients.

      I’ll work on that wish list:)

      Roberta

  2. You write, ” I always select the option to compare me only to those who have tested both the HVR1 and HVR2 regions, and only show me people who match in both.”

    Why do you do this? It seems to me that you may have a match, particularly with those who tested only HVR1 at a company other than FTDNA – but you won’t see them. It may be “like swatting at flies”, but some of those flies may be your match. It seems to me that eliminating a subset of your matches is a little like finding out that you have MOORE or SMITH ancestry, and deciding that you won’t pursue it at all because it will be like swatting at flies.

    You also write, ” If you are haplogroup H, and you have any HVR2 matches at all, the only reasonable way to sort out who is related in a genealogical timeframe is to take the full sequence test.”

    I had a cousin test for me who was haplogroup H, and had 3 HVR2 matches. I emailed the 3 matches, and one turned out to be genealogically connected. No full mitochondrial DNA sequence was done. Why was it unreasonable to just email the 3 matches and ask, rather than do the FMS?

    • Some people literally have thousands of matches a the HVR1 level. Think about those who match the CRS exactly or have only 16519c. There are just too many people (matches) to work through. That’s why I set my threshold to HVR1+HVR2 only. However, if someone actually wants to do this, there is certainly no issue with it. Most people want to know how to work most effectively. I would certainly contact the closest matches first, then work back through the HVR1 only.

      Most, but not all, haplogroup H still have a lot of matches at the HVR2 level. Many other haplogroups do too, but H is the most common so I used them as an example. Most people see something like 200 or 300 matches and their eyes glaze over. At some point, and that point probably differs for each person, the matches just become overwhelming and they want them to be as succinct as they can be. If they have only 3 HVR2 matches, then great, it’s easy to e-mail 3 people. 33 matches, probably still OK. If they have 333, then maybe they are willing to try to work through that. If they have 3,333, probably not.

  3. Roberta,
    I have NO “0” matches on mitosearch. I tested through SMGF and belong to the K1a4a1a2a haplogroup (K Project member). None of my K matches have tested to HVR2 and all of them are “-1” – even those at K1a4a1.
    Is this common? Am I just unlucky? I’m trying to decide whether it is worthwhile to test to FMS – our maternal surname is Fulton from Ulster, Ireland.

    Kathleen

  4. Roberta, I have been encouraging people to upload their FMS data to http://www.mtdnacommunity.org. There is a tab on your personal results page at FTDNA to expedite the process. This is a new site, and not all features have been implemented, but it will show your “close” matches, in contrast to FTDNA which will show you only your exact matches. It’s not common, but close matches can have genealogical significance. In my case, I have a “private” mutation that must have occurred sometime within the last 300 years, which prevents me from seeing an interesting match at FTDNA (a common ancestor born in 1640).

    Your name and e-mail address will be visible to your matches but not the positions where you differ. This preserves some confidentiality in the event that some of your mutations have medical significance. My custom reports for people with FMS data provide details about medical implications.

    Ann Turner DNACousins@aol.com

      • I have the opposite problem, in a way. I tested with Genographic in 2005 and uploaded to Family Tree. I had then, and will never have, any matches because of a back mutation in the coding region. I am an R*, and my “journey” ends in the middle of nowhere. A few years later, I heard of Sorenson, filled out all their paperwork and did their test. I did this because they upload the actual family tree and I found a match there! Descendants of one of my great grandmother’s sisters. From there back, the tree is the same. So the back mutation happened within the last four generations, which is like right now in the time frame we’re working with. Of course, there is no way to get in touch with those people, and now Sorenson is dead in the water.
        I was excited at first to read about Genographic 2, but now I’m thinking, “What’s the point for me, really?” Other than finding out about Neanderthal heritage, which will just be something for my husband to laugh about. And I really really don’t want to know about medical implications. That’s dangerous ground if your insurance company ever gets hold of it!
        This has been unexpected, disappointing, and frustrating.

      • Hi Cindy,

        A couple of things. First, just so you know, Geno 2.0 does NOT return medical information. It’s entirely stripped out, so that isn’t a concern.

        Second, today, with full sequence results, Family Tree DNA now allows up to 4 mutation differences and still shows you as a match, so your single back mutation alone will NOT prevent you from having matches.

        Roberta

  5. Pingback: What is a Haplogroup? | DNAeXplained – Genetic Genealogy

  6. Pingback: Mitochondrial DNA Tests for $49 – Wow!!! | DNAeXplained – Genetic Genealogy

  7. After emailing my 4 best FMS matches yesterday, I was wondering what to do next. I reread this post and it has given me several ideas. I am going to rework my introductory letter along the lines of your sample letter. I plan to upload my data to Mitosearch tonight after that.

    I am surprised at the lack of resolution in the FMS even with an exact match. The ftDNA web site says there is a 50/50 chance of a common ancestor in 5 generations and a 90% chance in 16 generations. In the worst case, a perfect FMS match leaves a gap in the paper trail of 250 years to fill. Ouch!

    • Yes, it is ouch. And unfortunately, all that the estimates can do is average mutation time. Your mutations might have just happened, or happened hundreds of years ago. There is just no way to know.

  8. Pingback: Deciphering the DNA of Sarah McSpadden Faires (c 1745 – c 1820) – 52 Ancestors #21 | DNAeXplained – Genetic Genealogy

  9. Hi Roberta,
    This post has really helped my thought process on mtDNA. I recently received my dad’s mtDNA results (B2a). I understand that mtDNA is passed on from mother to child, but only her daughters can pass it on to their children, but what makes up the 3rd line of DNA in women? If men get the y-dna from their father and mtDNA from their mother, and the mixture of autosomal DNA. Then women also get the autosomal DNA and mtDNA from their mother–then what makes up the third line?

  10. Both my parents were adopted so I need to know if I have Canadian native in me or Italian or German or Scottish and what percentages of different races I am made of?

    Also are your laboratories in Canada or in America?

  11. Pingback: DNAeXplain Archives – Introductory DNA | DNAeXplained – Genetic Genealogy

  12. Sorry this is coming in a little late from your writing of it.
    Both my sister and I tested FMS with FTDNA. The results give me 9 matches with a 0 Genetic Distance, including my sister.. All of them have done the FMS. I understand from ISOOG that 0 genetic distance means there are no differences in the two results and they are an exact match. I’ve read all I can find but still wonder what an exact match mean to me? Should I pay more attention to them for making contact? Thanks, Roberta.

    • Your closest matches will likely be zero differences a the full sequence level. They could be very close in time or several generations back. It just depends on the luck of the draw in terms of when mutations do or don’t happen.

  13. You are terrific to take the time to answer. It’s all stiff a bit fuzzy, but little by slowly I’m picking up what I can from you and your other great blogger buddies.
    Merry Christmas, Roberta. 🙂

      • Hehe! I meant chagrin…I am surprised that not a single match popped up. Perhaps it’s just that no one that would be a match on the planet has tested? I am interested in what is involved in a consult 🙂

      • Generally, you ask me what you want to know about your DNA and I look at your DNA and give you the best answer possible within the constraints of an hour. When I thought it was your results compared to your children, that is a perfect Quick Consult. Otherwise, a report would probably be in order to see if you have rare mutations, or what exactly, or you can just wait for matches.

  14. I just got my results and I’m seeing something curious. I have a GD 1 match to someone who also did FF. When I look him up in my FF match list he is identified as a paternal match. Does this mean that someone from my maternal line is married to his father or is there another explanation?

    • No, if he is identified as a paternal match, he matches you on your paternal line, meaning someplace in your father’s line. If your GD1 is mtDNA, that could be from far back in time, possibly, or they may match you on both sides but thee isn’t anyone on the mother’s side for them to phase to in order to be assigned to your mother’s side too.

      • Since he got his mT DNA from his mother wouldn’t that imply that he is also related on his mothers side but just too far back to register in aT DNA? I presume that would also mean that he is descended from a female sibling of someone in my direct maternal line. Is there any way to guestimate just how far back (# of generations) that might be. There doesn’t seem to be the equivalent of the Y TiP report for mT DNA.

      • Right, but the paternal match means the autosomally, he is matching that person on the father’s side. There are two different kinds of DNA involved and two different ancestors. He could be related on the mtDNA side more closely in time BUT if there is no one to phase against, there is no way to assign the maternal side. There is no mtDNA TIP report and no way to estimate, unfortunately. I’ve seen everything from 1 generation to many.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s