Mitochondrial DNA Build 17 Update at Family Tree DNA

I knew the mitochondrial DNA update at Family Tree DNA was coming, I just didn’t know when. The “when” was earlier this week.

Take a look at your mitochondrial DNA haplogroup – it maybe different!

Today, this announcement arrived from Family Tree DNA.

We’re excited to announce the release of mtDNA Build 17, the most up-to-date scientific understanding of the human genome, haplogroups and branches of the mitochondrial DNA haplotree.

As a result of these updates and enhancements—the most advanced available for tracing your direct maternal lineage—some customers may see a change to their existing mtDNA haplogroup. This simply means that in applying the latest research, we are able to further refine your mtDNA haplogroup designation, giving you even more anthropological insight into your maternal genetic ancestry.

With the world’s largest mtDNA database, your mitochondrial DNA is of great value in expanding the overall knowledge of each maternal branch’s history and origins. So take your maternal genetic ancestry a step further—sign in to your account now and discover what’s new in your mtDNA!

This is great news. It means that your haplogroup designation is the most up to date according to Phylotree.

I’d like to take this opportunity to answer a few questions that you might have.

What is Phylotree?

Phylotree is, in essence, the mitochondrial tree of humanity. It tracks the mutations that formed the various mutations from “Mitochondrial Eve,” the original ancestor of all females living today, forward in time…to you.

You can view the Phylotree here.

For example, if your haplogroup is J1c2f, for example, on Phylotree, you would click on haplogroup JT, which includes J. You would then scroll down through all the subgroups to find J1c2f. But that’s after your haplgroup is already determined. Phylotree is the reference source that testing companies use to identify the mutations that define haplogroups in order to assign your haplogroup to you.

It’s All About Mutations

For example, J1c2f has the following mutations at each level, meaning that each mutation(s) further defines a subgroup of haplogroup J.

As you can see, each mutation(s) further refines the haplogroup from J through J1c2f. In other words, if the person didn’t have the mutation G9055A, they would not be J1c2f, but would only be J1c2. If new clusters are discovered in future versions of Phylotree, then someday this person might be J1c2f3z.

Family Tree DNA provides an easy reference mutations chart here.

What is Build 17?

Research in mitochondrial DNA is ongoing. As additional people test, it becomes clear that new subgroups need to be identified, and in some cases, entire groups are moved to different branches of the tree. For example, if you were previously haplogroup A4a, you are now A1, and if you were previously A4a1 you are now A1a.

Build 17 was released in February of 2016. The previous version, Build 16, was released in February 2014 and Build 15 in September of 2012. Prior to that, there were often multiple releases per year, beginning in 2008.

Vendors and Haplogroups

Unfortunately, because some haplogroups are split, meaning they were previously a single haplogroup that now has multiple branches, a haplogroup update is not simply changing the name of the haplogroup. Some people that were previously all one haplogroup are now members of three different descendant haplogroups. I’m using haplogroup Z6 as an example, because it doesn’t exist, and I don’t want to confuse anyone.

Obviously, the vendors can’t just change Z6 to Z6a, because people that were previously Z6 might still be Z6 or might be Z6a, Z6b or Z6c.

Each vendor that provides haplogroups to clients has to rerun their entire data base, so a mitochondrial DNA haplogroup update is not a trivial undertaking and requires a lot of planning.

For those of you who also work with Y DNA, this is exactly why the Y haplotree went from haplogroup names like R1b1c to R-M269, where the terminal SNP, or mutation furthest down the tree (that the participant has tested for) is what defines the haplogroup.

If that same approach were applied to mitochondrial DNA, then J1c2f would be known as J-G9055A or maybe J-9055.

Why Version Matters

When comparing haplogroups between people who tested at various vendors, it’s important to understand that they may not be the same. For example, 23andMe, who reports a haplogroup prediction based not on full sequence testing, but on a group of probes, is still using Phylotree Build 12 from 2011.

Probe based vendors can update their client’s haplogroup to some extent, based on the probes they use which test only specific locations, but they cannot fully refine a haplogroup based on new locations, because their probes never tested those locations. They weren’t known to be haplogroup defining at the time their probes were designed. Even if they redefine their probes, they would have to rerun the actual tests of all of their clients on the new test platform with the new probes.

Full sequence testing at Family Tree DNA eliminates that problem, because they test the entire mitochondria at every location.

Therefore, it’s important to be familiar with your haplogroup, because you might match someone it doesn’t appear that you match. For example, our haplogroup A4a=A1 example. At 23andMe the person would still be A4a but at Family Tree DNA they would be A1.

If you utilize MitoSearch or if you are looking at mtDNA haplogroups recorded in GedMatch, for example, be aware of the source of the information. If you are utilizing other vendors who provide haplogroup estimates, ask which Phylotree build they are using so you know what to expect and how to compare.

Knowing the history of your haplogroup’s naming will allow you to better evaluate haplogroups found outside of Family Tree DNA matchs.

Build History

You can view the Phylotree Update History at this link, but Built 17 information is not yet available. However, since Family Tree DNA went from Built 14 to Build 17, and other vendors are further behind, the information here is still quite relevant.


If you’re wondering how much the tree grew, Build 14 defined 3550 haplogroups and Built 17 identified 5437. Build 14 utilized and analyzed 8,216 modern mitochondrial sequences, reflected in the 2012 Copernicus paper by Behar et al. Build 17 utilized 24,275 mitochondrial sequences. I certainly hope that the authors will update the Copernicus paper to reflect Build 17. Individuals utilizing the Copernicus paper for haplogroup aging today will have to be cognizant of the difference in haplogroup names.


If your haplogroup changed, or the haplogroup of any of your matches, your matches may change. Family Tree DNA utilizes something called SmartMatching which means that they will not show you as a match to someone who has taken the full sequence test and is not a member of your exact haplogroup. In other words, they will not show a haplogroup J1c2 as a match to a J1c2f, because their common ancestors are separated by thousands of years.

However, if someone has only tested at the HVR1 or HVR1+HVR2 (current mtDNA Plus test) levels and is predicted to be haplogroup J or J1, and they match you exactly on the locations in the regions where you both tested, then you will be shown as a match. If they upgrade and are discovered to be a different haplogroup, then you will no longer be shown as a match at any level.

Genographic Project

If you tested with the Genographic Project prior to November of 2016, your haplogroup may be different than the Family Tree DNA haplogroup. Family Tree DNA provided the following information:

The differences can be caused by the level of testing done, which phase of the Genographic project that you tested, and when.

  • Geno 1 tested all of HVR1.
  • Geno 2 tested a selection of SNPs across the mitochondrial genome to give a more refined haplogroup using Build 14.
  • Geno 2+ used an updated selection of SNPs across the mitochondrial genome using Build 16.

If you have HVR1 either transferred from the Genographic Project or from the FTDNA product mtDNA, you will have a basic, upper-level haplogroup.

If you tested mtDNA Plus with FTDNA, which is HVR1 + HVR2, you will have a basic, upper-level haplogroup.

If you tested the Full Mitochondrial Sequence with Family Tree DNA, your haplogroup will reflect the full Build 17 haplogroup, which may be different from either the Geno 2 or Geno 2+ haplogroup because of the number and selection of SNPs tested in the Genographic Project, or because of the build difference between Geno 2+ and FTDNA.

Thank You

I want to say a special thank you to Family Tree DNA.

I know that there is a lot of chatter about the cost of mitochondrial DNA testing as compared to autosomal, which is probe testing. It’s difficult for a vendor to maintain a higher quality, more refined product when competing against a lower cost competitor that appears, at first glance, to give the same thing for less money. The key of course is that it’s not really the same thing.

The higher cost is reflective of the fact that the full sequence mitochondrial test uses different technology to test all of the 16,569 mitochondrial DNA locations individually to determine whether the expected reference value is found, a mutation, a deletion or an insertion of other DNA.

Because Family Tree DNA tests every location individually, when new haplogroups are defined, your mitochondrial DNA haplogroup can be updated to reflect any new haplogroup definition, based on any of those 16,569 locations, or combinations of locations. Probe testing in conjunction with autosomal DNA testing can’t do this because the nature of probe testing is to test only specific locations for a value, meaning that probe tests test only known haplogroup defining locations at the time the probe test was designed.

So, thank you, Family Tree DNA, for continuing to test the full mitochondrial sequence, thank you for the updated Build 17 for refined haplogroups, and thank you for answering additional questions about the update.


If you haven’t yet tested your mitochondrial DNA at the full sequence level, now’s a great time!

If you have tested at the HVR1 or the HVR1+HVR2 levels, you can upgrade to the full sequence test directly from your account. For the next week, upgrades are only $99.

There are two mtDNA tests available today, the mtPlus which only tests through the HVR1+HVR2 level, or about 7% of your mitochondrial DNA locations, or the mtFull Sequence that tests your entire mitochondria, all 16,569 locations.

Click here to order or upgrade.



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

Genealogy Services

Genealogy Research

45 thoughts on “Mitochondrial DNA Build 17 Update at Family Tree DNA

    • Individuals mutations, other than the haplogroup defining mutations don’t represent any specific thing. Haplogroup defining mutations define haplogroups. Some other mutations wind up defining match group. It’s all individual.

  1. Hi Roberta, a little bit on and off topic, but is there any advantage in choosing to do the full sequence test on the mother, the daughter or the son? All will have the same exact sequence, I presume, but is there any tactical advantage to be had by testing one over the others when all are available?

  2. Difficult to understand that I was T2b2a and I am now T2b2b, while T2b2a has dissapeared. Mystery! I also that my personnal sequence was used to define T2b2b.

    • Below is what is mentioned in email from FTDNA

      “In a very few cases haplogroups may have reverted to a higher branch on the tree. Usually, this is because in Build 14, some of the branches of the tree were predicted, not confirmed. The additional sequences added between Build14 and Build 17 did not provide supporting data to justify their existence, so these branches have been removed.”

  3. Hi Roberta, would you happen to know why the pace of updates to PhyloTree has slowed down so much over the years? Thanks. I ask because I’m waiting for a particular new subclade that’ll be in the queue.

    • Not specifically, but I suspect it has to do with the inherent complexity and the fact that it takes so much effort for vendors to rerun their entire data bases to conform to new versions.

  4. I’m still plain H2a1, maybe I’ll get more lucky with BUILD 18.

    So updates are at $99 right now, I need to tell one of my HVR1+2 match. Up to which day does this sale works?

  5. Thank you for an easy readable article. I appreciate your good explanations. I will share the cost for a mtDNA FULL with my sister. Thank you.

    Ralf Wrangö R-M269

    Ekerö Sweden

    > 24 mars 2017 kl. 22:33 skrev DNAeXplained – Genetic Genealogy : > >

  6. Good that FTDNA has updated to Build 17 however it appears they may not have re-run all mtDNA results as according to James Lick’s Build 17 anaysis I should have moved from K1-T16362C -> K1d. Not K1d1 as I mis-match 6890G as can be seen on the phylotree with 6890A. Is there a reason why FTDNA’s and James Lick’s Build17 results are different?

  7. I tested both my mom and dad with the mtFull sequence. My dad’s came out fine with lots of matches and a haplogroup that didn’t change with the newest build, but my mother has no matches and no assigned haplogroup. Is this because not enough people have tested from her group?

  8. Thank you for this, which helps as I continue to try to wrap my head around mtDNA. I particularly appreciate you describing what Geno 2+ is, given that’s where my father’s result is from. Not upgradeable, I don’t think, but I’ll keep an eye out for other sales, as I like the idea of getting matches even if they’ll likely be beyond our paper trail (as I understand it).

    Some of my female ancestors are rather shadowy figures, some with only first names, that I can’t help but want more information on them!

  9. I am concerned if it is necessary for my or my sister to upgrade to newest built 17. Mom had mtDNA full sequence done a year ago. I later had mine done but only the mtDNA plus (also had family finder done at the same time). My sister also had mtDNA plus analysis. We are in haplogroup H4a1. Should any of us spend money to have upgrades to the newest build 17 with the mtDNA full sequence that includes the coding region? We do have a family tree of the Mother’s line back to early 1800’s. For the ancestor on the maternal line born 1795 we have the first name but not clear about the last name as spellings have four versions. mtDNA reanalysis would not help us with the paper trail. Mom is 97 with her full analysis and family finder on the former build. If my sister and I need to upgrade to build 17 we need to do so right away. Should or could we have Mom upgraded? any benefit for her results?

    • There are two separate things here. 1. Everyone was upgraded to build 17. It’s free. 2. If you have not tested to the full sequence level, you can upgrade to that level of testing. If your Mom already had it done, there is no reason for you or your sister to do it too, as you will both be the same as your mother.

      So it sounds like you are all set.

  10. Same subclade, less cumbersome name, for me and my mom. Old version had the defining back mutation in the name… T2b2-C16304T! is now T2b2b. Thanks for the article, Roberta.

  11. FTDNA sounds very advanced, but makes it sounds like their testing is better even though they don’t do mt or y testing.

    Who is correct? Here are a couple quotes:

    ..How does the new AncestryDNA test differ from other DNA tests?
    It’s more comprehensive…

    …information is more relevant and recent—targeting your family history a few hundred or even a thousand years ago, as compared to the Y and mtDNA tests, which have a 10,000 to 50,000 year time focus.


    • T and mtDNA CAN reach back that far in time, but they can also be very useful by matching precisely with contemporary people. For example, if you’re a male and you want to know if your Y DNA matches the same surname line as the ancestor you think you descend from, autosomal DNA can’t tell you that. It’s very specific for the Y line and the mtDNA line. Autosomal can’t be broken down like that. It depends on what you want to know. Here’s a short article explaining the difference between the different kinds of DNA. Both kinds of tests have their place.

  12. mtDNA sometimes gets a negative rep for not being particularly useful for figuring out ancestors within a genealogical timeframe… however, I wonder if that rep might change with more folks utilizing full sequence mtDNA testing? mtDNA testing of my paternal aunt revealed that our brick wall ancestor (on her matrilineal line) is likely of British ancestry, and not the Pennsylvania Dutch we had long thought.

  13. Dear, Miss Roberta Estes
    Today, i know more about my Cousins, my Aunt’s from my Father’s side, an my Deep Maternal Ancestors from FTDNA mtFullSequence (FMS). My mtDNA belongs to Hg B4c (Southern Chinese and Southeast Adian), my Father’s and his sister’s mtDNA belongs to Hg M7c3c — today M7c1c3 (an Austronesians including Filipino, Borneo and Indonesian). But my Cousins and my Mother’s Sister in Law are belongs to mtDNA Hg M8a3a (Native Siberian, The Kazakhstan people?). I never imagine when i can learned more specifics about the DNA test from an Anthropological perspective. Thanks!

  14. Roberta, you are my “go-to” person for DNA answers. My first test at was the basic mitochondrial which didn’t yield much information. I then added autosomal, but during the recent sale I upgraded to the full mitochondrial, expecting it to have more useful information on matches. I just got my results, and there seems to be no change at all in the reporting I am getting from the original small test and the full sequence test.
    I understand that I can’t get matches to people who haven’t tested. But when i log in and click the mitochondrial match results, I get the notice “No matches found at this level of testing. You are being automatically directed to the previous level of testing” which takes me back to HVR1 level only, which I think is the same as my previous results without the full mitochondrial test. Needless to say I am disappointed to not have more information than I had in the past. Is this normal? To be expected??? Do I just need to be patient and wait for more people to do full mitochondrial tests?

    • You may have some very unusual mutations, or your DNA may be so rare that very few people will match you at the full sequence level. There’s a message in that as well.

      • Do you have a post on what no matches at mtFull Sequence could imply? My mother has no matches. Her mother and grandmother were Germans born in Romania.

  15. Pingback: Which DNA Test is Best? | DNAeXplained – Genetic Genealogy

  16. Pingback: Native American Mitochondrial Haplogroups | DNAeXplained – Genetic Genealogy

  17. I took an mtFullSequence to know more about my deep maternal ancestors and i saw my mtDNA results was quite similar with a build 17’s data, but i have missing mutation at 16235 position. A B4c2’s mutation must have A16235G, but i have A16235A. It is okay for me to claimed my mtDNA belong to Haplogroup B4c2?

  18. I am T152C…… and if I am reading correctly it says “no label”..can someone explain this to me please.

  19. I am Mtdna B2l according to family Tree 5% american indian however according to family history there is no Indian records or history in my family they know they have some jewish ancestry
    through the mother side and the rest from Spain. Seville and the Canary Islands. I really want to know more about my Mtdna type if there is more information available. Thanks.

  20. I just got my FTDNA full sequence mitochondrial results back. I wanted information on my 3rd great grandmother who was born in Bordeaux, France ca. 1780. I found a living cousin who was in the direct mother to daughter line who agreed to take the DNA test. My closest match (in HRV1, HRV2, Coding Regions) showed 4 matches with Genetic Distance 2. It also showed 9 matches with Genetic Distance 3. They all trace down to Haplogroup H1t. All of my Genetic Distance 2’s and 3’s who have responded to my emails show more recent family trees coming from the Azores, Hawaii, Portugal, Brazil or the Basque region. Could this mean that my 3rd great grandmother could have come from this general region? All of the trees I have seen from these matches have most recent ancestors from the 1800’s, much more recent than the 1780’s. I am stumped and don’t know if the genetic distance is too far away (although I read that a genetic distance of 3 could be more closely related than a genetic distance of 2.) I am thoroughly confused. Could anyone help?

    • Your closest matches can be in the past few generations or further back in time. Looking at your exact matches and their locations can provide clues about where your ancestors lived or migrated from.

  21. Any talk of build 18, its been 4 years and the data has got to be exponential? FTDNA has my at V13 (~1050 y.o.) So my Russian Viking VK253 might be an actual grandmother, ;). YFull has as of today added 2 more layers @ V13b1 and only ~150 y.o. This would be closer than my Big Y matches that I have met. I’m currently back to 1800 Ilawa, Poland on my MtDna line.

Leave a Reply to MarjorieCancel reply