How can we make genealogical matches with mitochondrial DNA? Can we extend our family trees with mtDNA? This used to be very difficult, but with the advanced tools recently released, it is now much easier. We will compare the current tools from two companies, Family Tree Dna and YFull, and see how to use them to their best advantage. We will also see how to get results into a relatively new database called mitoYDNA.
What are the origins of my great-great grandmother?
| Women in my mitochondrial DNA line |
My mother was not raised with her biological family. She was separated from them when she was six years old, and she did not see her parents or siblings again. We knew nothing about my mother's family until, after exhaustive research, I located her mother. At that time, my mother and grandmother had been separated for 52 years, so it was quite a reunion. My grandmother asked me, "If you can find me, can you find my other children?" It took awhile because this was before DNA testing, but I did find them. Then I started asking about our origins. My grandmother gave me the above photo of her maternal line. My great-great-grandmother is in the front-center of the photo. Her daughter, my great-grandmother, is on the right side of the photo. My grandmother's sister is standing in front of her mother and is holding flowers. The other women in the photo were married to my great-grandmother's brothers, so they are not biologically related to me. Even though many of these people were alive in my lifetime I never saw them. However, two of those women are my direct maternal line. I feel a real connection to them because I know that I inherited their mitochondrial DNA.
My grandmother told me that her grandmother was Native American. Is this another of those erroneous family stories about Native American heritage? Can I prove this and find out more through a mitochondrial DNA (mtDNA) test? I decided to order some mtDNA tests and find out.
What is mitochondrial DNA?
Mitochondria are organelles found outside the nucleus of the cell. They carry their own DNA. Mitochondria are always inherited from the mother. Both males and females inherit mitochondrial DNA from their mother, but only females can pass the mtDNA to their children. So your mitochondrial DNA can be used to find out more about your strictly maternal line: your mother's mother's mother . . . .
| mtDNA inheritance is shown in red |
Mitochondrial DNA has far fewer mutations than nuclear DNA, and it mutates much less frequently. Therefore, you may have mtDNA matches where the common ancestor is beyond the genealogical time frame. Most people have not tested their mtDNA, so it may take a long time to make a genealogical match.
We will trace the history of mtDNA testing, and see why it used to be considered almost impossible to extend maternal lines with this test. But due primarily to the innovation of Family Tree DNA (FTDNA), and with new tools from YFull and mitoYDNA, it is now very possible.
As you will see in this post, when I did early mtDNA testing there were few, if any, matches. It could have been considered a complete waste of time and money. But the entire point of this blog post is to show that even if you don't get immediate results with mtDNA, don't give up. Lead the way. Recruit people. Get into more databases. Follow the progress below. You are leaving a lasting legacy of your maternal line.
Early mtDNA testing
When I first ordered my mitochondrial DNA tests, all companies only tested the regions that are called hypervariable regions. These two mitochondrial regions are the ones that mutate the most, so they were considered best for tracing ancestors. The two hypervariable regions are called HVR1 and HVR2. The test results are compared to a reference sequence, and the differences are reported. The early mtDNA results were compared to the revised Cambridge Reference Sequence (rCRS).
MITOCHONDRIAL DNA RESULTS AT FTDNA
HVR1 and HVR1 results
My MTDNA results from Family Tree DNA looked like this:
In addition to receiving the mutations I was assigned a basic haplogroup, U5b. A mtDNA haplogroup is a broad grouping of people who share similar mtDNA results and have a common ancestor in the distant past. At the time I received these results, haplogroup U5b was mostly known for being associated with Finland. It appeared that U5b was definitely not a Native American haplogroup! What were my maternal origins?
| HVR1 and HVR2 mutations |
In addition to receiving the mutations I was assigned a basic haplogroup, U5b. A mtDNA haplogroup is a broad grouping of people who share similar mtDNA results and have a common ancestor in the distant past. At the time I received these results, haplogroup U5b was mostly known for being associated with Finland. It appeared that U5b was definitely not a Native American haplogroup! What were my maternal origins?
Along with the mtDNA results, I got a list of people who had the same, or very similar, mutations. Here is an example of how these results would have appeared:
| mtDNA match list |
The name of the person taking the mtDNA test is listed in the first column. In the second column is a link to email the person, a place for notes, and the notation HVR2 which means that the person tested regions HVR1 and HVR2. The third column contains the name of the earliest known ancestor, and the fourth column is the mitochondrial DNA haplogroup.
To be perfectly honest, most of this information was pretty useless. There is no way with HVR1 and HVR2 results to tell how closely these people are related because there is no genetic distance indication or any way to compare the actual DNA mutations. The Earliest Known Ancestor column was often of little value because in many cultures the woman's surname changes upon marriage, so every generation there is a new surname. Most people did not enter enough information in this column.
The biggest problem was that there were so few mutations, and they were shared by so many people.
Full Sequence mtDNA at Family Tree DNA
One day I was looking at my Family Tree DNA account and noticed something that was not there the day before. Family Tree DNA was offering a full-sequence mitochondrial DNA test! No other company offered this test. I knew that I needed more mutations, so I was very excited. I was less excited when I saw that the price was $895 [It has since come down considerably!!], but I immediately ordered the test because these mtDNA results might actually help break through this maternal brick wall.
I was told by FTDNA that I was the first person to ever order the full-sequence mtDNA test, so when the results came back, of course, I had no matches at all. In fact, I was assigned a new haplogroup that had never been seen: U5b1c.
You may wonder, if I knew I would have no matches, what's the point of upgrading? Because even with meager beginnings, eventually you may make a great match. If you want to encourage other people to test your mtDNA line, somebody has to go first.
With no matches at all, what could I do with these results?
The first-ever haplogroup project
It was (and still is) far easier to trace ancestors with Y-DNA because the male surname in a family may have remained intact for multiple generations. Also, there are many more mutations, and the Y-DNA mutations could be compared by joining a surname project. At the time I first tested mtDNA, the only kind of project or group that existed at any company was the surname project. The display of Y-STR mutations in a Y-DNA surname looked something like this:
| Y-DNA STR comparison in FTDNA surname group |
These men all share the same surname. We can see that these three men appear to be related, and one of them has one mutation that the other two men don't have. We needed a way to do the same thing with mtDNA mutations. But surname projects would not be useful for mtDNA because of the continual changing of maternal surnames. So I was trying to come up with some way to compare mtDNA results.
I came up with the idea of a haplogroup project. This kind of grouping did not exist at any DNA company. I contacted three of the companies where I had tested my mtDNA and asked if I could start a haplogroup project. Two of the companies said no. One company, Family Tree DNA (FTDNA), said they had never done this before, but they would consider it. In 2005, the first-ever haplogroup project was formed. I joined my new U5b project, and, of course, I was the only one in it. Not too useful! So I contacted my HVR1 and HVR2 matches and asked them to join. I posted in the relatively few relevant online forums that existed at the time, and pretty soon more mtDNA haplogroup projects started forming, people started joining them, then Y-DNA haplogroups started. From that small beginning, haplogroup projects are now considered by many to be an essential part of our ancestral research.
Display of mtDNA results in a FTDNA haplogroup project
Now that we had a way to compare mutations, we could begin to do something with mtDNA. As the price went down many people started testing their full mtDNA sequence, and the haplogroups started refining.
Here are results from a Family Tree DNA project. In mitochondrial DNA projects at Family Tree DNA, only the HVR1 and HVR2 mutations are shown, even for people who tested their full sequence, because some people would not want their full list of mutations to be displayed. The columns below show the earliest known maternal ancestor, the country of origin, the haplogroup, and the HVR1 and HVR2 mutations.
| Comparing mutations in FTDNA haplogroup project |
Making mitochondrial DNA matches with the above information can be very difficult:
1. Many people do not enter enough information about the earliest-known maternal ancestor.
2. The Country field can be very misleading. For example, people have been told in the past not to select "United States" unless the ancestor was Native American. Next, what does "origin" really mean? If someone has traced her ancestry to the 1850s in North Carolina, but heard that the ancestor might have immigrated from Ireland, did this person enter United States or Ireland in the country field? Were the ancestors in Scotland before they emigrated to Ireland? Without more precise information about the earliest known ancestor, we can't tell much from the Country field. Furthermore, the country of origin is not specific enough. For example, there's quite a difference between Alabama and Massachusetts in the United States, or between the Fujian Province and the Gansu Province in China.
3. The main problem with HRV1 and HVR2 mtDNA mutations is that there are so few of them, and some of them can be quite old. Are any of the above mutations shared by everybody within haplogroup U5? Are any of them shared in another haplogroup? We can't tell by looking at these results.
Even with the above shortcomings, limited results can still be useful. The first person in the chart above tested both HVR1 and HVR2, so there are more mutations to compare. The second and third people only tested HVR1, and hundreds of people in the U5 group may share those mutations. You really need a more complete list of mtDNA mutations. However, there is one result above that is useful for tracing ancestry: the second one.
1. Many people do not enter enough information about the earliest-known maternal ancestor.
2. The Country field can be very misleading. For example, people have been told in the past not to select "United States" unless the ancestor was Native American. Next, what does "origin" really mean? If someone has traced her ancestry to the 1850s in North Carolina, but heard that the ancestor might have immigrated from Ireland, did this person enter United States or Ireland in the country field? Were the ancestors in Scotland before they emigrated to Ireland? Without more precise information about the earliest known ancestor, we can't tell much from the Country field. Furthermore, the country of origin is not specific enough. For example, there's quite a difference between Alabama and Massachusetts in the United States, or between the Fujian Province and the Gansu Province in China.
3. The main problem with HRV1 and HVR2 mtDNA mutations is that there are so few of them, and some of them can be quite old. Are any of the above mutations shared by everybody within haplogroup U5? Are any of them shared in another haplogroup? We can't tell by looking at these results.
Even with the above shortcomings, limited results can still be useful. The first person in the chart above tested both HVR1 and HVR2, so there are more mutations to compare. The second and third people only tested HVR1, and hundreds of people in the U5 group may share those mutations. You really need a more complete list of mtDNA mutations. However, there is one result above that is useful for tracing ancestry: the second one.
How can you use limited mtDNA results to trace ancestors?
To extend your mtDNA line at FTDNA, having matching mutations is not enough. Matching mutations with an ancestral surname are not enough. What you need is the information found in the second row in the example above. Along with the list of mutations you need a name, date, and place where the earliest known ancestor lived. For example, if you had a maternal ancestor who was born about 1865 in or near Broome County, New York, her parents could have lived there around the 1840s when Rhoda Ann Collar was born. If your HVR1 mutations match, you have somewhere to start. You might try to connect your family tree to Rhoda Ann Collar, and you would definitely want to encourage the descendant of Rhoda to upgrade her results to a full mitochondrial DNA sequence to see if the two of you are still matches.
Here's an example of such mtDNA success. Shortly after the U5b project was formed at Family Tree DNA, I noticed that one woman had matching HVR1 and HVR2 results with another woman in the project, and their earliest known ancestors had lived in the same county. One of the women had traced her family much further back than the other woman. So I traced the ancestry of the woman who had less information and was able to extend her ancestral line three generations further when I connected her line with one of the other woman's ancestors. This was only possible because we were able to compare their results in a haplogroup project.
Yes, tracing ancestry with mitochondrial DNA can be difficult, but it's definitely possible!
More refined haplogroup with full-sequence results
Hypervariable region results can only estimate a broad haplogroup like "U5." But the full sequence results can provide a more precise haplogroup that may get further refined as more people test their mtDNA.
GenBank: Advancing scientific research
In 2005 nobody with haplogroup U5b1c was in GenBank which is the database of the National Center for Biotechnology Information (NCBI). GenBank is very important for researchers. I wondered if there were any way to get my results into that database. I found that there was a man, Ted Kandell, who was doing programming to try to make FTDNA submissions compatible with the GenBank database, but GenBank had not yet agreed to the submissions. So, Ted did the programming, and I helped negotiate with GenBank. In 2006, both of us submitted our full-sequence results.
Submitting to GenBank helps refine the mtDNA haplogroups for all further testers, and it is possible for you to contribute to scientific research by submitting your full mtDNA sequence.
Although my full-sequence mtDNA results were originally assigned to haplogroup U5b1c, there have been many more testers in the fifteen years since that test. I now have 19 exact full-sequence matches at Family Tree DNA, and the haplogroup has been further refined to U5b1c2. Here are some of the matches:
The ability to add a family tree is an extremely important advantage for making matches.
Progress at FTDNA with Full sequence matches
Although my full-sequence mtDNA results were originally assigned to haplogroup U5b1c, there have been many more testers in the fifteen years since that test. I now have 19 exact full-sequence matches at Family Tree DNA, and the haplogroup has been further refined to U5b1c2. Here are some of the matches:
| Full-sequence mtDNA matches at FTDNA |
Notice that with full-sequence results there is a Genetic Distance column. The 0 in the Genetic Distance column shows that this is an exact match. The name of the person taking the test is next, and if you click on the name you will see the person's profile information and email address. Some of these people actually have family trees as shown by the
| Family Tree symbol |
The ability to add a family tree is an extremely important advantage for making matches.
To compare the individual mutations you must still join a project.
Differences in Full-sequence mtDNA haplogroups at FTDNA
Today, in FTDNA mitochondrial DNA haplogroup projects the display of results looks the way it did in the beginning with one exception: the names of the haplogroups have been expanded beyond the basic U5 or U5b. [I removed the first few columns in the screenshot below to focus on the one that changed.]
| Refined mtDNA haplogroup |
Although the results have not changed, the number of people in many of the mtDNA groups has expanded dramatically.
Precise locations for mtDNA matches
The most useful feature for making mtDNA matches is not found in your FTDNA match list or in your haplogroup project. By far, the most important piece of information for mtDNA matching is the place where your ancestor lived. The best matching information is found in the Matches Maps section of your FTDNA account.
| Matches Maps at FTDNA |
When you click on this link you will be taken to a map. Many of my matches have added the exact location coordinates to their most distant ancestor. While this precise location does not show up in the match list or in the haplogroup project, it does show up in the Matches Maps.
| Map of mtDNA origins |
You can click on any pin to find more information. Your own pin is the white one. The locations for the most distant maternal ancestor of your exact mtDNA matches are colored red. Remember, however, that these are locations reported by your matches. Look at their family tree to verify this information. For example, I clicked on the red pin at the bottom left, and here is one of my exact matches:
This appears to be a paternal ancestor, not a maternal ancestor. This person is on my match list and has no family tree. So I will disregard this pin unless it is verified that the maternal ancestor was named John or that the most distant known maternal ancestor was John's daughter who lived in this location.
These pins have the most potential for allowing you to make great matches, so be sure to update the most precise location for your ancestor, if known. Instructions appear in the "How to make matches with Mitochondrial DNA" section below.
A second mitochondrial DNA test at FTDNA
In 2007 I ordered a full-sequence mtDNA test for a Chinese relative of mine. We will call her Mary. Mary has always lived in China, and most people who test at Family Tree DNA are from the United States. Mary is in haplogroup M8a2b. Since Mary didn't have any close matches, I went no further until YFull started mtDNA groups (discussed below). So those results have been sitting for 13 years.
Grouping in mtDNA projects at FTDNA
There is no M8 or M8a project at Family Tree DNA, but there is a haplogroup M project, so Mary has recently been put in that project. Projects at FTDNA require manual grouping by the project administrators. Mary is in the last row of the Ungrouped section:
| Ungrouped results in FTDNA haplogroup project |
If we look at the M8 section of the haplogroup M project we can see that when Mary's results are moved to the appropriate subclade of haplogroup M, Mary will still have no matches in the project. There are only two people in the M haplogroup project who are within subclade M8. Neither of them is in haplogroup M8a2b.
| Subclade M8 in mtDNA haplogroup project M |
With no matches to compare, Mary's results will still have to wait until more people test. Mary needs to be in more than one database. Her results were submitted to YFull.
MITOCHONDRIAL DNA RESULTS AT YFULL
What is YFull?
YFull is not a DNA testing company. YFull is an analysis and comparison service for Y-DNA Next Generation Sequencing and full mitochondrial DNA sequences. These interpretation and comparison services are quite different from those found at any DNA testing company.
Why would you want to submit your
mitochondrial DNA results to YFull?
At YFull, your results will be compared to scientific samples and to people who tested at different companies. This will prove to be increasingly significant because the full mitochondrial DNA sequence can be extracted from full genome tests, and people are now beginning to test their full genome at various companies and transferring the results to YFull. Furthermore, there will always be more scientific studies.
YFull only began to do mitochondrial DNA analysis in 2019, and so many people do not know about its mtDNA services. It is possible that you may not yet have many matches there. However, even though many of YFull's mtDNA services are just starting you will learn much more about your mitochondrial DNA from YFull's matching system and YFull's groups.
Mitochondrial DNA matches at YFull
When you submit results to YFull you will see a menu of options for Y-DNA and for mtDNA. Here are the current options in the mtDNA menu:
| mtDNA menu at YFull |
When you click on Hg [abbreviation for haplogroup] and SNPs, you will see a screen similar to the following. Only the first two lines are shown below.
| Haplogroup and SNPs |
Here is a closeup of the SNP information on the right of the screen:
| mtDNA haplogroup and SNPs |
For the first mutation on the list, the Reference Sequence has A at position 15052. This is called the Ancestral value (Anc). The tester has a G in this position which is called the Derived value (Der). So the mutation is listed as A15052G.
Notice at the top of the image that you can download three versions of your results at any time.
Here is a closeup of the left of the screen:
| mtDNA haplogroup |
| Click to go to YFull public mtDNA tree |
Click on it to be taken to your placement in the YFull mtDNA tree. You can scan up and down the tree to any position. I did this, so in the example below we are seeing results from haplogroup HV4.
| Some DNA samples in mtDNA haplogroup HV4 |
Each time the tree is updated, you will see the notation "New" next to samples that have recently been added. In the above screenshot there are four samples from scientific studies, and three have countries of origin. Of the three private submissions, only one has a country of origin, but it also has a specific region. Hover the mouse over the abbreviated place names to see the place in full. The specific regions of a country are especially useful.
This is the only information available to the general public, but you can find much more by clicking on the Mt matches link in your account menu.
| Click on mtDNA matches |
Mitochondrial DNA match lists at YFull
Your matches can evolve rather quickly. For example, a few months ago I was identified as U5b1c2 at YFull. I was moved to a more downstream subclade in the next version of the haplotree, and even further in the version of the tree that was released this week. Each time I get a new list of matches.
Below is part of a list of matches.
| Scientific samples in mtDNA match list |
You will not receive any identifying information about your matches. For example, you will not receive an email address, but you will be able to contact anyone in your match list or in your groups, no matter how distant the match is, by using the Private Message (PM) envelope.
The PM column is listed next to the YFull ID column. You will see an arrow
The PM column is listed next to the YFull ID column. You will see an arrow
next to scientific samples. You will see an envelope
next to individual submissions.
To me, the most interesting piece of information is where to find more about the study of this sample. The GenBank record shows that the name of the article in which this sample appeared is "Full mtGenome reference data: Development and characterization of 588 forensic-quality haplotypes representing three U.S. populations." But how do I find this article?
YFull has a list of mitochondrial DNA articles for samples found in their database. Click the name of the article to see an abstract and find how to access the full article:
Included with the name of the article is a link to Samples. When you click the Samples link on the right of the screen, you will be taken to the list of samples from the study. Here is a portion of them:
When you click the PM envelope, you will see something like this:
| Send Private Message (PM) |
When you click the arrow next to a scientific sample ID, you see much more information. In my own list of matches I clicked the arrow next to KM102073.1. The link takes me to the GenBank record. Below is only a small portion of it.
| GenBank entry |
YFull has a list of mitochondrial DNA articles for samples found in their database. Click the name of the article to see an abstract and find how to access the full article:
| Scientific study with mtDNA samples in YFull database |
| List of YFull mtDNA samples from scientific study |
This list includes the YFull ID for the sample, the ID from the scientific article, the country of origin (if reported), a link to the YFull haplogroup where it appears, and specific mutations.
In the sample list above, ID KM102073.1 is on my list of matches. This is from a hispanic sample in the United States.
YFull mtDNA Groups
