What is a Heteroplasmy and Why Do I Care?

Most people have never heard of a heteroplasmy – but you might have one.

You Might Have a Heteroplasmy If…

…You have no exact matches at the full sequence mitochondrial DNA level.

A heteroplasmy is one of the first things I think of when someone tells me they have no exact full sequence matches but several that are a genetic distance of 1, meaning one mutation difference.

That phenomenon usually means the tester has a rare mutation that no one else has, at least no one who has tested their mitochondrial DNA (yet) – and that mutation just might be a heteroplasmy.

Heteroplasmies are generally (but not always) quite recent mutations. Actually, heteroplasmies are mutations caught in the act of mutating – kind of like an insect in genetic amber – frozen in time in your generation.

By Anders L. Damgaard – http://www.amber-inclusions.dk – Baltic-amber-beetle CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=16792582

Let’s say you might have a heteroplasmy. Or maybe you want to see if you do. Even if YOU don’t have a heteroplasmy, other people’s heteroplasmies can and will affect matching.

Here’s everything you ever wanted to know about heteroplasmies but didn’t know to ask😊

Heteroplasmies are Fascinating

A heteroplasmy is actually quite interesting because it’s a genetic mutation in progress.

This means you have two versions of a DNA sequence showing in your mitochondrial DNA at a specific location.

Said another way, at a specific genetic location, you show both of two separate nucleotides. Amounts detected of a second nucleotide greater than 20% are considered a heteroplasmy. Amounts below 20% are ignored. Generally, within a few generations, the mutation will resolve in one direction or the other – although some heteroplasmies persist for several generations and can sometimes define family branches.

If you’d like to read more about mitochondrial DNA, I wrote a series of step-by-step articles and combined them into one resource page, here.

Show Me!

You can easily check to see if you have a heteroplasmy by signing on to your FamilyTreeDNA account. Hopefully, you’ve taken the full sequence test.

Today, new testers, thankfully, can only purchase full sequence tests, so HVR1 results don’t present quite the same challenges when combined with heteroplasmies as they used to. We’ll talk about that in a minute.

If you have only taken the HVR1 or HVR1+HVR2 “Plus” test, as opposed to the Full Sequence, you can upgrade by signing on here and clicking on the “Full” button on the Maternal Ancestry section of your personal page.

These buttons will be pink if you’ve taken that test already, and grey if you need to upgrade. If you have an account at FamilyTreeDNA, you can add a mitochondrial DNA test to that same account by clicking on “Add Ons and Upgrades” at the top of your personal page. You can order a test if you’re a new customer, here.

How Do I Know if I Have a Heteroplasmy?

Your mitochondrial DNA has a total of 16,569 locations that you can think of as addresses. If your DNA at those locations is normal, meaning no mutations, they won’t be listed in your results.

Mutations are shown in your mitochondrial DNA results by a different letter at the end of the location.

For example, here are my mutations for my HVR1 region. Each of these locations in the HVR1 region has a mutation.

For locations that are shown in your results, meaning those where you have a mutation, you’ll see, in order:

  • A letter, either T, A, C or G
  • The location number
  • A different letter, typically another one of T, A, C or G, but sometimes a small d

For the first mutation, C16069T, the location address is 16069, the normal value is C, the mutation that occurred is T.

Heteroplasmies are shown in your mitochondrial DNA results by letters other than T, A, C, G or d at the end of the location.

I don’t have any heteroplasmies, so I’m switching to the results of a cousin who has a heteroplasmic mutation at location T16362Y to use as an example. The trailing Y means they have a heteroplasmy at location 16362.

But first, what do those letters mean?

The Letters

The letters stand for the nucleotide bases that comprise DNA, as follows:

  • T – Thymine
  • A – Adenine
  • C – Cytosine
  • G – Guanine
  • d – a deletion has occurred. There is no nucleotide at this location.

For location T16362Y, the first letter, T, is the “normal” value found at this location. If a mutation has occurred, the second letter is the mutated value. Normally, this is one of the other nucleotides, A, C or G.

Any other letter after the location has a specific meaning; in this case, Y means that both a C and a T were found, per the chart below.

Note – if you have a small letter t, a, c or g, it’s not a heteroplasmy, and I wrote about small letters and what they mean in the article, Mitochondrial DNA Part 2: What Do Those Numbers Mean?

Check Your Results

On your FamilyTreeDNA personal page in the mtDNA section, click on the Mutations tab.

If you’ve taken the full sequence test, you’ll see Extra Mutations. You’re looking for any mutation that ends in any letter other than T, A, C, G or d.

If you haven’t taken the full sequence test, you don’t have “Extra” mutations listed, but you can still view your mutations for the HVR1 and HVR2 regions.

Look for any value that has any letter other than T, A, C, G or lower case d at the end of the location.

The Y tells us that this location is a heteroplasmy.

Heteroplasmy Matching

Ok, let’s look at a heteroplasmy mutation at location 16326. A heteroplasmy can occur at any mitochondrial location. I’ve selected this location because it occurs in the HVR1 region of the mitochondrial DNA, so even people who haven’t tested at the full sequence level will see results for this location. Plus, the location at which the heteroplasmy occurs affects matching in different ways.

Using the example of T16362Y, the Y tells us that both nucleotides C and T were found. This location should match against anyone carrying the following values in the same location:

  • Y (letter indicating a C/T heteroplasmy)
  • T (standard or normal value)
  • C (mutated value)

However, currently at Family Tree DNA, the heteroplasmy only counts as a match to anyone with a Y, the specific heteroplasmy indicator, and the “normal” value of T, but not the mutated value of C.

This table shows how heteroplasmies are counted at FamilyTreeDNA. For heteroplasmy T16362Y, based on the value your potential match has at this location, you either will or will not be considered a match at that location.

Scenario Other Person’s Value Your Result – T16362Y
1 T16362Y – heteroplasmy indicator Match to you at this location
2 T16362T – normal value, not a mutation Match to you at this location
3 T16362C – mutated value Not counted as match to you at this location
  • If your match has a value of Y, the heteroplasmic C/T value, they are counted as a match to you, so no problem.
  • If your match has a value of T, the normal value, this location won’t be shown on their mutation list at all. They WILL be counted as a match to you so there’s no issue.
  • If your match has a value of C, the mutated value, in my opinion they should also be counted as a match to you, but they aren’t today. The logic, I believe, was that the most likely value is the standard or normal value and that the mutated value is much less likely to be accurate. Regardless, I’ve requested this change and am hoping for a matching adjustment in a future release for heteroplasmies.

Heteroplasmies do affect matching at the different levels.

Viewing Your Matches

Mitochondrial DNA, for testing purposes, is broken into three regions, HVR1 (hyper-variable region 1), HVR2 and the Coding Region.

At FamilyTreeDNA, you can view your matches at each level. The matches are cumulative, meaning that the HVR2 level includes the HVR1 level information, and the Coding Region level includes the HVR1 and HVR2 regions. That highest level which includes all three regions shows information from your entire your entire full mitochondrial DNA sequence.

Heteroplasmy Effects on Matching

If you otherwise match someone exactly, but one of you has a heteroplasmy and the other person carries the mutated value, you will be counted as a mismatch of 1 at the full sequence level.

A mismatch has different effects when it occurs in the HVR1, HVR2 or Coding Regions, respectively.

GD is an abbreviation for Genetic Distance which is how mutations are counted. A GD of 1 means the two people have one mutation difference between them.

In the following chart, the effects of you having a nonmatch, heteroplasmic or otherwise, in each of the regions is shown at each level. The region in which the mismatch occurs is shown in the first column, at left, and the effect the mismatch has on matching in each region is shown in columns 2-4.

The red sections are not counted as matches.

Mismatch Occurs in this Region HVR1 Level Match to Someone Else HVR2 Level Match to Someone Else Coding Region Level Match to Someone Else
HVR1 region nonmatch GD of 1 means no match GD of 1 means no match GD of 1 is a match
HVR2 region nonmatch Does not affect HVR1 – so you are a match GD of 1 means no match GD of 1 is a match
Coding Region nonmatch Does not affect HVR1 – so you are a match Does not affect HVR2 – so you are a match GD of 1 is a match

For purposes of this discussion, we’re assuming our two people being compared in the chart above match exactly on every other location so matching is not otherwise affected.

  • If your heteroplasmic nonmatch occurs in the HVR1 region – in other words, scenario 3 – you’ll fall into the HVR1 nonmatch row. That means you won’t be shown as a match at the HVR1 or HVR1+HVR2 levels, but you WILL be shown as a full sequence match.
  • If your heteroplasmic nonmatch is in the HVR2 region of addresses, it won’t affect your HVR1 matches, but it will affect your HVR2 and Coding Region matches. This means you will be shown as HVR1 match, not an HVR2 match, but will be a full sequence match.
  • If your heteroplasmic nonmatch is in the Coding Region, it won’t affect your HVR1 or HVR2 matches, but it will affect your Coding Region matches. However, it won’t preclude matches and you’ll be shown as a match in all three regions.

To be very clear, I have no issue with these match thresholds. It’s important to understand how this works, and therefore why heteroplasmic (and other) mismatches in specific regions affect our matches in the way they do.

Why Aren’t Mismatches of 1 Counted as Matches in the HVR1 or HVR2 Regions?

The match threshold at FamilyTreeDNA for the HVR1 and the HVR1+HVR2 regions, both small regions of about 1000 locations each, is that only an exact match is considered a match. Therefore, a heteroplasmic nonmatch in this region can really be confusing and sometimes misleading, especially if either or BOTH people have NOT tested at the full sequence level.

These are the match thresholds in effect today.

HVR1 GD or # of Mutations Allowed for a Match HVR2 GD or # of Mutations Allowed for a Match Coding Region GD or # of Mutations Allowed for a Match
0 – no mutations allowed 0 – no mutations allowed 3 mutations allowed

If both people match on either the heteroplasmy identified (Y in our case) or one person has the normal value – all is fine. But if one person has a heteroplasmy and the other has the mutated value – then a mismatch occurs. This is really only problematic when:

  • The heteroplasmy mismatch is in the HVR1 region and both people have only tested at that level, causing the two people to not match at all.
  • The heteroplasmy mismatch occurs in combination with other mutations that, cumulatively, push the two people over the GD 3 full sequence matching threshold.

The second scenario happens rarely, but I have seen situations where people don’t match their mothers, aunts, siblings, or other close relatives because of multiple heteroplasmic mutations occurring in different people.

And yes, this is hen’s teeth rare – but it does occasionally happen.

So, what’s the bottom line about heteroplasmies?

Heteroplasmy Bottom Line

  1. You can suspect a heteroplasmy if you have full sequence matches, but no exact matches.
  2. If you have a heteroplasmy in the HVR1 region, understand that you may not have many or any matches in the HVR1 and HVR2 regions. The remedy is to test at the full sequence level and check matches there.
  3. If you have a heteroplasmy and don’t match someone you expect to match – reach out to them and ask about their value at that specific location. If that location isn’t listed for them in their results, then they have no mutation there and your heteroplasmy is NOT the cause of you not matching with them.
  4. If you don’t match someone you expect to match, reach out to them and ask if THEY have any heteroplasmies. The easiest way to ask is, “Do you have any mutations listed that end with anything other than T, A, C, G or d?” Feel free to link to this article so that they’ll know where to look, and why you’re asking.

Do you have any heteroplasmies?



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 Products and Services


Genealogy Research

47 thoughts on “What is a Heteroplasmy and Why Do I Care?

  1. Very interesting article. So if in the Coding Region I have a lower-case t at a location I have a heteroplasmy there? I have no exact matches, and no 1 step matches, just 2.

  2. Thank you for this information. What does a small ‘t” rather than a capital ‘T’ mean at the end of a mutation? I did not see this annotation in your list and it occurs on a coding region mutation on a kit that I manage. As you mentioned, this kit has no full mtDNA matches at the full sequence level and only 2 matches that are a genetic distance of 1.

  3. In my Coding Region I have “A825t” – note the lower case t, not the upper case T. What does this mean?

  4. I have some mutations that end in either a lower case c or lower case t rather than capital letters. Does that means anything?

  5. I have two people who state my own fourth great grandmother as their oldest known female line, and match me at a genetic distance of 3. I have no heteroplasmy, and my siblings and I share probably more autosomal dna than what I’d expect, although that’s probably irrelevant, although we are definitely related. I’m puzzled though, about whether this is consistent or not. I have about 165 gd 1 matches, about the same number with gd 2, and the remaining of about 550 total matches at gd 3. Does this make sense?

    • I would suggest reaching out to them and looking to see if they have heteroplasmies. Ad how closely they match each other. It’s certainly not impossible but statistically, it’s not likely either.

  6. As you stated in the Blog.
    I have NO 0 GD matches. I do however have several 1, 2, 3 GD matches. Each of my matches have C14167T and I have C14167Y

    Closest known relative is traced to about 6 generations back.

    I have tried unsucessfully for a long time to get someone closer than 6 generations to test FMS.

    Have not given up but anxious to find out more. I therefore recently upgraded my brother to FMS. Results will be after several weeks.
    We are now in waiting mode to see if my brother has the heteroplasmy also or if it starts past my mother.

  7. Thank you Roberta. However, your thoughtful explanation still does not explain my situation with my first cousin, about which I have been round after round with FTDNA for a proper explanation during the past 9 nine years. Last I heard months ago, FTDNA was going to review results and explain.
    As there has been no explanation forthcoming, and confusing results remain as before, I hope you won’t mind my venting my frustration here.
    Background: Our mothers were sisters, identical female lineage. I have heteroplasmy C16296Y. Cousin has mutation C16296T. She has 818 FMS matches — three with GD0. I have 816 matches, most with GD1 some GD2 / 3. Cousin is GD2!!
    One might think we mismatch at another location, but as far as I can tell we are otherwise identical. If you have an explanation, I’d love to have it. Thanks again for your time and effort.

    • No, and I clearly can’t analyze this without looking at both results. There could be a matching bug of some sort too. I do know that some updates and improvements are scheduled for later this year, so hopefully that will resolve any situation.

  8. I have a heteroplasmy in the coding region, it took a long time to research the reasons and implications of this. I understand that a mutation in the coding region can cause health issues. I was able to get my niece, daughter of my sister, to test and her results confirmed that my mutation is mine alone.

    • I can’t comment on medical issues except to say that if you have concerns, please contact a genetic counselor.

    • It’s entirely possible that your niece also has heteroplasmy, but the percentage of mtDNA molecules with the mutation is below the threshold for reporting.

      Medical implications of a heteroplasmy depend on the position of the mutation. My main reference for my custom reports is


      • To be clear, Ann is a medical doctor and does interpret mitochondrial DNA for medical implications.

      • Hey Ann,

        I posted my FASTA file to MITOMAP and it did not pick up my Heteroplasmy. Does this mean there may be an issue for one of the sites in presenting my results? It missed another of my transitions as well. Is it becuase the mustation is too new?

        Coding Region – C11089 Y


        • Mags, I checked my personal database of sequences from GenBank for heteroplasmies in submissions from people who had tested at FTDNA. There were a few hundred such records where a base was reported as Y or R instead of A/G/T/C. That could imply that it’s not a matter of policy to omit heteroplasmies in a FASTA file, but it’s also possible that people used Ian Logan’s method to create a GenBank record. In that case, they would manually be entering the heteroplasmy based on their list of differences from the CRS. You may need to do a little survey and have people with known heteroplasmies check their FASTA file.

  9. Hi Roberta,
    Great article and great timing. I’ve been working on documenting my mtDNA results with my brother (B), sister (S) and maternal uncle (U). All these were done with FTDNA.
    My mother’s test was done postmortem by SecuriGene out of Vancouver, Canada. I’ve included her results as (M), but I don’t have raw data to import into FTDNA.

    Location 1555 (Coding Region):
    A1555G – B, U, M
    A1555R – S, me
    Location 185 (HVR2):
    G185A – B, S, U, M
    G185R – me
    I’m still working on understanding the last section of your article where you discuss how FTDNA figures out the GD for matches.
    I match S with a GD of 2. I don’t see B or U because it only shows up to GD=3. Does this fit the scenario which you call as rare as hen’s teeth?


      • I looked into the Advanced Matches section. I downloaded the results and looked for values in the HVR1, HVR2 and FMS columns that were other than “-” (no info available) or “x” (no match) for any of the 3 columns.
        U – 0, x, x (no match)
        B – 0, x, x (no match)
        S – 0, x, 2 (came up as GD=2)
        For people I matched, unknown genetic relationships:
        PC – 0, 0, 2 (exact match on HVR1, HVR2)
        BD – 0, 0, 3 (ditto)
        8 people with – 0, x, 3
        1 person with – x, x, 3
        So, I’m working on figuring out what this all means.

        I find it interesting though, that 2 people, PC and BD, match me on HVR2 so they also have the same heteroplasmy G185R there? I don’t know if there is any significance to this. I understand there is a lot of research with respect to what causes mtDNA mutations. Exciting times!!
        Thank you so much for this article. You straightened out my understanding of heteroplasmy. I had some concepts wrong.

      • My sister (S) matches my brother (B) and uncle (U) except for location 1555 in the Coding Region (CR).
        Yet her GD = 2.
        A1555G (B,U,M) vs A1555R (S) – mutation and heteroplasmy.
        How does FTDNA count this as a GD of 2?

        • I have the same thing. My mother had a heteroplasmy (4646Y). I took the test and it had resolved for me as there was no mutation there listed (which I suppose means I’m the normal 4646T). But, our genetic distance is 2 rather than 1, and we both match only two other people at a GD of 3.

  10. I have 2 Mutations that I should only have one or the other. I first tested on Ancestry and had a bunch of 1 Mutations. I re-tested at Family Tree DNA. I had one person who matched perfect at both HVR1 and HVR2.I had 3 persons at HVR1 that matched me perfect. One of the persons who matched at HVR1 is a cousin to the one who matched me at HVR1 and HVR2 We connected and found our line. The other 2 HVR1 persons would never respond. I am a Family Tree DNA Administrator for the Savage Group Project.

  11. Thank you, Roberta! This is so helpful. My dad has A14899R and C16365Y. My mom has T4646Y and A16183c. I now know why they have no exact matches.

  12. My sub clade ends with ! I was told some years ago that this was a heteroplasmy. It is now recognised as a sub clade of H. Is this a heteroplasmy? If so I reckon it is very old.

  13. Thank you for this. It is helping me understand a situation that has had me puzzled. I tested an mtDNA descendant of my GG grandmother (“Person 1”) in hopes of finding out who her mother was. I then tested another person (“Person 2”) who was an mtDNA descendant of someone whose surname, age and location indicated she could have been the sister of my GG grandmother.

    When the second results came in in October 2016, Person 2 had no matches closer than 1 Genetic Distance, these included Person 1. Those same matches were all indicated as exact matches in Person 1’s results.

    Then, some months later (I didn’t note down exactly when it happened) the results changed to show that Person 2 had no matches closer than a Genetic Distance of 2 (including person 1) I’m confused as to why this happened, as it seems that there is only one heteroplasmy in Person 2’s results.

    These are the extra mutations shown:

    Person 1

    Person 2

    Do you have any idea of what is going on here? Was there a change in the analysis after Oct 2016 that might have led to the change in Genetic Distance? Thanks for any ideas!

    • I don’t see anything that should cause a GD of 2. I wonder if there’s a bug that is counting the heteroplasmy as 1 and then the mutated value as 1 too. So a total of 2.

  14. Thank you for this very detailed explanation. My daughter and I are a GD of 2! I think I understand this more having read your article.

  15. I have one in the coding region, 13185 which ends in Y. I also have extra mutations 315.1C

    I have no GD0 matches but I also don’t have any atDNA matches with any mtDNA matches. My maternal great-uncle did his Y-DNA test so maybe I can see if his sample can be used for mtDNA too. He’s no longer in a position to test again.

    Honestly, though–I need to read this post a few more times before it all sinks in.

  16. Roberta, thank you for this article! My mtDNA test results came back in 2016. I thought these results would only give me haplogroup results so just never looked closely at the mutations information. Because of your article, I went back and followed your instructions.

    It looks like I may have a heteroplasmy – in the coding region G4092R. Also have 3 transversions in the coding region: A825t, C11242g, and C15452a. How can I tell if there are more than 20% of Gs or As in the G4092R so I know whether I really do have a heteroplasmy? I also have a missing mutation C152T, and 4 other mutations: 315.1C, 522.2C, G1719A, and C16527T.

    Since every female fetus forms all of her own individual eggs at about 12 weeks of gestation, just how exactly do these mutations occur? I have not yet checked any of my siblings’ mtDNA nor any of my own daughters’ mtDNA, I don’t know if I am the beginning of these mtDNA mutations in my family, or if they are accumulated from mother and grandmothers. (Will have to save up the $$$ and pay for the tests.)

    My family has medical issues, including me and my daughters. Recently saw a genetic physician who informed me that ‘science has not yet caught up’ but that I do have an unknown connective tissue disorder and/or another genetic mechanism.

    If you can enlighten me in any way, or otherwise point me in the right direction, I would appreciate it very much. I have been wondering since age 28, just what is wrong genetically in my family. Am now 74 years old. Just glad there are people much smarter than me. Thanks much.

    • By definition, you do have a heteroplasmy because of the “R” — that is the code for a mixture of mtDNA molecules with G and A. Re the question of how heteroplasmy develops, this article has a diagram:


      Heteroplasmy per se has no medical implications — it depends on which location has the mutation. Mitomap.org does not show any records of medical conditions for position 4092.

      • Thank you for sending me that link, Ann, and taking time to reply. I had already found the list of mt mutations that cause disease so was aware that my heteroplasmy location is not known to cause medical issues. The diagram in your linked document explains how these mutations occur, so I have a better understanding.

        Apparently, there is only one genetic physician here in our state of CO, whom I saw in late 2020. I worked with a genetic counselor regarding my young grandson, who was hospitalized with pneumonia at the time. A WES test was done in 2019 but no significant mutations turned up in that test. I also learned that Medicare will not cover genetic testing, unless it involves cancer. I also had an Aortopathy Comprehensive Test that only turned up one Variant of Unknown Significance.

        It was frustrating that we did not find the mutation that surely must be in my dna somewhere. And it was equally frustrating when the genetic doctor said I have an unknown connective tissue disorder and/or an, as yet, unknown genetic mechanism. Looks like I have done about as much as I am able to find the cause of our numerous and varied medical issues. Thanks again.

  17. Maybe late to the conversation, but hope you can help explain something to me.

    I tested a few years ago and am K1c2, extra mutations 315.1C, 522.1A, 522.2C, and T16362C. From what I understand the T16362C might be unusual and mean a subgroup will be or is created….
    I was hoping that heteroplasmy might be the cause, but your article did not indicate this. So the question: is it normal to have 100s of matches with genetic distances of 1, 2, 3 … but only two matches 0 distance? Both these people share my mtDNA ancestor Francoise Barbary. No people 0 distance to a common woman who was further back in history?
    I do understand that in the grand scale of things not enough people have tested, but just a little frustrating to not place the origin of the line closer to the root.

    I have another question, and maybe you can point me to an article for that. I supposedly have been connected (loosely) to historical figures through my mtDNA (mitoYDNA.org)

    User ID EKA Surname EKA Birthplace Haplogroup Distance

    T19181 Barbary France K1c2 0 0 0
    16224C, 16311C, 16320T, 16362C, 16519C 73G, 146C, 152C, 263G, 315.1C, 498-

    Z10198 Roet Somme, Picardie, France J1c2c3 6 7 24 16069T, 16126C, 16519C 73G, 146C, 185A, 188G, 263G, 295T, 315.1C, 462T, 489C

    Z10200 Pfannberg T2a1a 8 3 36
    16126C, 16169Y, 16294T, 16296T, 16519C 73G, 263G, 315.1C

    Z10396 Forcalquier Forcalquier, France H1af2 6 6 22 16111T, 16357C, 16519C 263G, 315.1C, 524.1A, 524.2C

    Z10699 Forcalquier Forcalquier, France H1af 7 4 25 16111T, 16357C 263G, 315.1C

    Is any of that possible given the different haplogroups and distances? Is that how people aquire mutations by inserting pieces of a fathers maternal DNA into the mix even if the haplogroup is different?

    Thank you for your expertise.

    • Those haplogroups aren’t close to you or each other at all. I suggest you contact the organization. And it’s not unusual to have only a few exact matches. You may get your own haplogroup one day.

Leave a Reply