I don’t think that we actively think about our DNA segments as nested ancestors, like Russian Matryoshka dolls, but they are.
That’s exactly why segment information is critical for genealogists. Every segment, and every portion of a segment, has an incredibly important history. In fact, you could say that the further back in time we can track a segment, the more important it becomes.
Let’s see how to unveil nested segments. I’ll use my chromosome 20 as an example because it’s a smaller chromosome. But first, let’s start with my pedigree chart.
Before we talk about nested segments that originated with specific ancestors, it’s important to take a look at the closest portion of my maternal pedigree chart. My DNA segments came from and through these people. I’ll be working with the first 5 generations, beginning with my mother as generation #1.
Generation 1 – Parents
In the first generation, we receive a copy of each chromosome from each parent. I have a copy of chromosome 20 from my mother and a copy from my father.
At FamilyTreeDNA, you can see that I match my mother on the entire tested region of each chromosome.
Therefore, the entire length of each of my chromosomes is assigned to both mother and father because I received a copy from each parent. I’m fortunate that my mother’s DNA was able to be tested before she passed away.
We see that each copy of chromosome 20 is a total of 110.20 cM long with 17,695 SNPs.
Of course, my mother inherited the DNA on her chromosome 20 from multiple ancestors whose DNA combined in her parents, a portion of which was inherited by my mother. Mom received one chromosome from each of her parents.
I inherited only one copy of each chromosome (In this case, chromosome 20) from Mom, so the DNA of her two parents was divided and recombined so that I inherited a portion of my maternal chromosome 20 from both of my maternal grandparents.
Identifying Maternal and Paternal Matches
Associating matches with your maternal or paternal side is easy at FamilyTreeDNA because their Family Finder matching does it automatically for you if you upload (or create) a tree and link matches that you can identify to their proper place in your tree.
FamilyTreeDNA then uses that matching segment information from known, identified relatives in your tree to place people who match you both on at least one significant-sized segment in the correct maternal, paternal, (or both) buckets. That’s triangulation, and it happens automatically. All you have to do is click on the Maternal tab to view your triangulated maternal matches. As you can see, I have 1432 matches identified as maternal.
Some other DNA testing companies and third-party tools provide segment information and various types of triangulation information, but they aren’t automated for your entire match list like Family Finder matching at FamilyTreeDNA.
You can read about triangulation in action at MyHeritage, here, 23andMe, here, GEDmatch, here, and DNAPainter, which we’ll use, here. Genetic Affairs AutoKinship tool incorporates triangulation, as does their AutoSegment Triangulation Cluster Tool at GEDmatch. I’ve compiled a reference resource for triangulation, here.
Every DNA testing vendor has people in their database that haven’t tested anyplace else. Your best strategy for finding nested segments and identifying matches to specific ancestors is to test at or transfer your DNA file to every vendor plus GEDmatch where people who test at Ancestry sometimes upload for matching. Ancestry does not provide segment information or a chromosome browser so you’ll sometimes find Ancestry testers have uploaded to GEDmatch, FamilyTreeDNA or MyHeritage where segment information is readily available. I’ve created step-by-step download/upload instructions for all vendors, here.
Generation 2 – Grandparents
In the second generation, meaning that of my grandparents, I inherited portions of my maternal and paternal grandmother’s and grandfather’s chromosomes.
My maternal and paternal chromosomes can be divided into two pieces or groups each, one for each grandparent.
Using DNAPainter, we can see my father’s chromosome 20 on top and my mother’s on the bottom. I have previously identified segments assigned to specific ancestors which are represented by different colors on these chromosomes. You can read more about how to use DNAPainter, here.
We can divide the DNA inherited from each parent into the DNA inherited from each grandparent based on the trees of people we match. If we test cousins from each side, assigning segments maternally or paternally becomes much, much easier. That’s exactly why I’ve tested several.
For the rest of this article, I’m focusing only on my mother’s side because the concepts and methods are the same regardless of whether you’re working on your maternal side or your paternal side.
Using DNAPainter, I expanded my mother’s chromosome 20 in order to see all of the people I’ve painted on my mother’s side.
DNAPainter allows us to paint matching segments from multiple testing vendors and assign them to specific ancestors as we identify common ancestors with our matches.
Based on these matches, I’ve divided these maternal matches into two categories:
- Maternal grandmother, meaning my mother’s mother, bracketed in red boxes
- Maternal grandfather, meaning my mother’s father, bracketed in black boxes.
The text and arrows in these graphics refer to the colors of the brackets/boxes, and NOT the colors of the segments beside people’s names. For example, if you look at the large black box at far right, you’ll see several people, with their matching segments identified by multiple colored bars. The different colored segments (bars) mean I’ve associated the match with different ancestors in multiple or various levels of generations.
Generation 3 – Great-grandparents
Within those maternal and paternal grandparent segments, more nested information is available.
The black Ferverda grandfather segments are further divided into black, from Hiram Ferverda, and gold from his wife Eva Miller. The same concept applies to the red grandmother segments which are now divided into red representing Nora Kirsch and purple representing Curtis Lore, her husband.
While I have only been able to assign the first four segments (at the top) to one person/ancestor, there’s an entire group of matches who share the grouping of segments at right, in gold, descended through Eva Miller. The Miller line is Brethren and Mennonite with lots of testers, so this is a common pattern in my DNA matches.
Eva Miller, the gold ancestor, has two parents, Margaret Elizabeth Lentz and John David Miller, so her segments would come from those two sides.
Generation 4 and 5 – Fuschia Segment
I was able to track the segment shown in fuschia indicated by the blue arrow to Jacob Lentz and his wife Fredericka Ruhle, German immigrant ancestors. Other people in this same match (triangulation) group descend from Margaret Elizabeth Lentz and John David Miller – but that fuschia match is the one that shows us where that segment originated. This allows us to assign that entire gold/blue bracketed set of segments to a specific ancestor or ancestral couple because they triangulate, meaning they all match me and each other.
Therefore, all of the segments that match with the fuschia segment also track back to Jacob Lentz and Fredericka Ruhle, or to their ancestors. We would need people who descend from Jacob’s parents and/or Fredericka’s parents to determine the origins of that segment.
In other words, we know all of these people share a common source of that segment, even if we don’t yet know exactly who that common ancestor was or when they lived. That’s what the process of tracking back discovers.
To be very clear, I received that segment through Jacob and Fredericka, but some of those matches who I have not been able to associate with either Jacob or Fredericka may descend from either Jacob or Fredericka’s ancestors, not Jacob and Fredericka themselves. Connecting the dots between Jacob/Fredericka and their ancestors may be enlightening as to the even older source of that segment.
Let’s take a look at nested segments on my pedigree chart.
You can see the progression of nesting on my pedigree chart, using the same colors for the brackets/boxes. The black Ferverda box at the grandparent level encompasses the entire paternal side of my mother’s ancestry, and the red includes her mother’s entire side. This is identical to the DNAPainter graphic, just expressed on my pedigree chart instead of my chromosome 20.
Then the black gets broken into smaller nested segments of black, gold and fuschia, while the red gets broken into red and purple.
If I had more matches that could be assigned to ancestors, I would have even more nested levels. Of course, if I was using all of my chromosomes, not just 20, I would be able to go back further as well.
You can see that as we move further back in time, the bracketed areas assigned to each color become smaller and smaller, as do the actual segments as viewed on my DNAPainter chromosomes.
Segments Get Progressively Smaller
You can see in the pedigree chart and segment painting above that the segments we inherit from specific ancestors divide over time. As we move further and further back in our tree, the segments inherited from any specific ancestor get smaller and smaller too.
Dr. Paul Maier in the MyOrigins 3.0 White Paper provides this informative graphic that shows the reduction in segments and the number of ancestors whose DNA we carry reaching back in time.
I refer to this as a porcupine chart.
Eventually, we inherit no segments from red ancestors, and the pieces of DNA that we inherit from the distant blue ancestors become so small and fragmented that they cannot be positively identified as coming from a specific ancestor when compared to and matched with other people. That’s why vendors don’t show small segment matches, although different vendors utilize different segment thresholds.
The debate about how small is too small continues, but the answer is not simply segment size alone. There is no one-size-fits-all answer.
As segments become smaller, the probability, or chances that we match another person by chance (IBC) increases. Proof that someone shares a specific ancestor, especially when dealing with increasingly smaller segments is a function of multiple factors, such as tree completeness for both people, shared matches, parental match confirmation, and more. I wrote about What Constitutes Proof, here.
In the Family Finder Matching White Paper, Dr. Maier provides this chart reflecting IBD (Identical By Descent) and IBC (Identical By Chance) segments and the associated false positivity rate. That means how likely you are to match someone on a segment of that size by chance and NOT because you both share the DNA from a common ancestor.
I wrote Concepts: Identical by Descent, State, Population and Chance to help you better understand how this works.
In the chart below, I’ve combined the generations, relationships, # of ancestors, assuming no duplicates, birth year range based on an approximate 30-year generation, percent of DNA assuming exactly half of each ancestor’s DNA descends in each generation (which we know isn’t exactly accurate), and the average amount of total inherited cMs using that same assumption.
Note that beginning with the 7th generation, on average, we can expect to inherit less than 1% of the DNA of an ancestor, or approximately 55 total cM which may be inherited in multiple segments.
The amount of actual cMs inherited in each generation can vary widely and explains why, beginning with third cousins, some people won’t share DNA from a common ancestor above the various vendor matching thresholds. Yet, other cousins several generations removed will match. Inheritance is random.
In order to match someone else descended from that 11th generation ancestor, BOTH you AND your match will need to have inherited the exact SAME DNA segment, across 11 generations EACH in order to match. This means that 11 transmission events for each person will need to have taken place in parallel with that identical segment being passed from parent to child in each line. For 22 rolls of the genetic dice in a row, the same segment gets selected to be passed on.
You can see why we all need to work to prove that distant matches are valid.
The further back in time we work, the more factors we must take into consideration, and the more confirming proof is needed that a match with another individual is a result of a shared ancestor.
Having said that, shared distant matches ARE the key to breaking through brick-wall ancestors. We just need to be sure we are chasing the real deal and not a red herring.
The most exciting possibility is that some segments are actually passed intact for several generations, meaning those segments don’t divide into segments too small for matching.
For example, the 22 cM fuschia segment that tracks through generations 4 and 5 to Jacob Lentz and Fredericka Ruhle has been passed either intact or nearly intact to all of those people who stack up and match each other and me on that segment. 22 cM is definitely NOT a small segment and we know that it descended from either Jacob or Fredericka, or perhaps combined segments from each. In any case, if someone from the Lentz line in Germany tested and matched me on that segment (and by inference, the rest of these people too), we would know that segment descended to me from Jacob Lentz – or at least the part we match on if we don’t match on the entire segment.
This is exactly what nested segments are…breadcrumbs to ancestors.
Part of that 22cM segment could be descended from Jacob and part from Fredericka. Then of Jacob’s portion, for example, pieces could descend from both his mother and father.
This is why we track individual segments back in time to discern their origin.
The Promise of the Future
The promise of the future is when a group of other people triangulate on a reasonably sized segment AND know where it came from. When we match that triangulation group, their identified segment may well help break down our brick walls because we match all of them on that same segment.
It is exactly this technique that has helped me identify a Womack segment on my paternal line. I still haven’t identified our common ancestor, but I have confirmed that the Womacks and my Moore/Rice family interacted as neighbors 8 generations ago and likely settled together in Amelia county, migrating from eastern Virginia. In time, perhaps I’ll be able to identify the common Womack ancestor and the link into either my Moore or Rice lines.
I’m hoping for a similar breakthrough on my mother’s side for Philip Jacob Miller’s wife, Magdalena, 7 generations back in my tree. We know Magdalena was Brethren and where they lived when they took up housekeeping. We don’t know who her parents were. However, there are thousands of Miller descendants, so it’s possible that eventually, we will be able to break down that brick wall by using nested segments – ours and people who descend from Magdalena’s siblings, aunts, and uncles.
Whoever those people were, at least some of their descendants will likely match me and/or my cousins on at least one nested Miller segment that will be the same segment identified to their ancestors.
Genealogy is a team sport and solving puzzles using nested segments requires that someone out there is working on identifying triangulated segments that track to their common ancestors – which will be my ancestors too. I have my fingers crossed that someone is working on that triangulation group and I find them or they find me. Of course, I’m working to triangulate and identify my segments to specific ancestors – hoping for a meeting in the middle – that much-desired bridge to the past.
By the time you’ve run out of other records, nested segments are your last chance to identify those elusive ancestors.
Do you have genealogical brick walls that nested segments could solve?
Follow DNAexplain on Facebook, here or follow me on Twitter, here. You can also subscribe to receive emails when I publish articles by clicking the “Follow” button at www.DNAexplain.com.
You’re always welcome to forward articles or links to friends.
You Can Help Out and It’s Free
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 Uploads
- FamilyTreeDNA – Y, mitochondrial, and autosomal DNA testing
- MyHeritage DNA – Autosomal DNA test
- MyHeritage FREE DNA file upload – Upload your DNA file from other vendors free
- AncestryDNA – Autosomal DNA test
- 23andMe Ancestry – Autosomal DNA only, no Health
- 23andMe Ancestry Plus Health
Genealogy Products and Services
- MyHeritage FREE Tree Builder – Genealogy software for your computer
- MyHeritage Subscription with Free Trial
- Legacy Family Tree Webinars – Genealogy and DNA classes, subscription-based, some free
- Legacy Family Tree Software – Genealogy software for your computer
- Charting Companion – Charts and Reports to use with your genealogy software or FamilySearch
- RootsMagic Software – Genealogy software for your computer
- Newspapers.com – Search newspapers for your ancestors
- NewspaperArchive– Search different newspapers for your ancestors
- DNA for Native American Genealogy – by Roberta Estes
- DNA for Native American Genealogy – for those ordering outside the US
- Genealogical.com – Lots of wonderful genealogy research books
- Legacy Tree Genealogists – Professional genealogy research
Hi Roberta –
Thank you for a very informative article. I am an avid “DNA Painter” and learn so much from your articles.
I am in the market for some genealogy software for my PC and so perused your links. A couple comments and a question: 1) the “Charting Companion” link does not work – seems there is a problem with the destination site; 2) the “RootsMagic Software” is no longer available via that link – discontinued perhaps; 3) do you then recommend the “Legacy Family Tree Software,” in other words is it something you use? There aren’t any reviews on their site, but I’d love to know from someone who uses it. I am aiming to have my tree “local” on my PC, so that if one day I no longer have a subscription to a tree site (or don’t have access to the internet – God forbid1), I can still navigate around in my tree and perform research. Thanks in advance for any feedback you can offer.
I will check those links. Thank you. Legacy has a trial copy you can download for free, or at least they used to. Try it and see if you like it.
Must the tree be public? I’m helping a friend find his birth parents. This might come in handy.
Do these techniques even work in endogamous populations? I don’t really need to do this in my tree for brick walls, but for figuring out or placing some unknown matches, this could come in handy. In a Leeds analysis, roughly 20% of my matches match on all 4 grand-parental lines. Another 20% only match one line. The remaining 60% match on 2 or 3 lines …. While I have both parents DNA, I don’t know how well this would work. At their closest, Mom and Dad are 3C1R, but there are 5 or 6 diffent kinship paths between them. The funny thing is that GEDmatch says they aren’t related if you do the “Are my parents related” test, but if you do a 1-to-1, they show up properly as 3C1R-ish. *sigh*
I don’t think the tree at FTDNA must be public.
Yes, you can opt out of Family Tree sharing at FTDNA. Go to Account Settings, selecting “Privacy and Sharing” and scroll down to the bottom, where you can change your Family Tree Sharing options, including who can see your Family Tree. The options are “All FamilyTreeDNA users”, “Only my matches” and “Only me.”
Thank you Roberta ….
Thank you for the wonderful article and explanations.
Great article and illustrations. I like your “Exciting Possibilities” section and your term of “nested segments.”
Thank you for another informative blog which raises a question in my mind: If I share, say, 25cM on a chromosome with a distant member of my father’s family and also a distant member of my mother’s family, does this imply an ancestor common to both sides of the family?
If you’re sure it’s from the same side of your family, yes, so long as it’s identical by descent.
Thanks, Roberta, for another very interesting article. As a summary, would you provide your definition of “nested segment”?
In each generation, above our parents, we can only assign a segment to a couple, like our maternal grandparents. Within each of those segments are segments from ancestors in the next distant generation. So those great-grandmother and great-grandfather’s segments are “nested” in the DNA of their child. That’s true of each generation, where in each segment can be tracked further upstream. That segment belongs to many ancestors. My mother, her mother, her father, his mother, etc., etc. on up the tree. The DNA of the oldest ancestors are “nested” in the DNA of the closest ancestors. Does that make sense. Think of those Russian nesting dolls.
Thanks, Roberta! That makes sense and is just what I was looking for.
In naming triangulated groups (TGs), is there a clever way to include the span of the segment so that the nested nature of the smaller ones within the larger ones would be evident? I am familiar with Jim Bartlett’s method, but I don’t think he includes the length. What TG naming scheme do you use?
I don’t but I need to think about this. Maybe Jim will too.