FamilyTreeDNA’s My Origins offers a glimpse of geographic ancestry

Where do I come from? The DNA analysis aspect of genealogy keeps trying to answer this question. By comparing an individual’s DNA with the same from a number of control groups, science is trying really hard to come up with an answer. The more I look into these DNA analyses, the less I think they hold any relevance to identity. And the explanation for this lack of relevance is completely clear, and if more people understood this, they’d probably stop using DNA as a gateway to how they are composed as an individual.

The problem is that every human inherits about half of his or her DNA from each of his or her parents, but it’s impossible to predict which half, or what DNA markers are included in that half. This is how this problem manifests itself over two generations in an incredibly simplified example.

Let’s say you know the birth locations of your four grandparents, and they happen to be Germany (father’s father), England (father’s mother), Lithuania (mother’s father), and Romania (mother’s mother). If you were to describe myself with just this information, taking the notion of Jewish heritage out of the mix, you would say you’re quarter German, quarter English, quarter Lithuanian, and quarter Romanian.

This assumes that each of these grandparents are completely “full booded” in their geographic make-up — and that’s not just unlikely, but impossible, and is yet another fault with searching for some kind of genetic explanation of geographic heritage.

Even if the grandparents represented some kind of perfection, your genetic makeup would more than likely not contain four equal components. Your father inherited roughly half his DNA from each of his parents, so he is roughly half German and half English. Your mother did the same, so she is roughly half Lithuanian and Romanian. That’s only because in this strange hypothetical example, the grandparents are each homogeneous. The succeeding generation, everything falls apart.

You inherit half your DNA from your father, who is half German and half English. But the half you inherit is not necessarily split evenly between your father’s German half and his English half. The DNA you inherit — your admixture — could be only his English DNA, it could be only his German DNA, it could be precisely 50% each side, but it most likely is some kind of unbalanced mix. So half of your DNA is most likely an unbalanced mix of German and English, and the other half is most likely an unbalanced mix of Lithuanian and Romanian. Your sibling could have inherited a different mix from each of your parents, and could thus have a significantly different geographical genetic fingerprint than you. But why should your sibling be different? He or she had the precisely the same family history through the centuries and the same migration patterns.

By the time you look back to your sixty-four fourth great grandparents, the chances are high that one is not reflected at all in your DNA today. I increasingly believe that while ancestral maps based on autosomal DNA tests are interesting to look at, they can’t really tell you the full story people really want to see.

FamilyTreeDNA My Origins
FamilyTreeDNA My Origins

Above is what FamilyTreeDNA’s new My Origins analysis has determined for my geographical make-up based on my autosomal (FamilyFinder) DNA test. It’s not terribly different from the analysis provided by AncestryDNA with that company’s autosomal test.


These might reflect an accurate approximation of the geographic locations some of my ancestors lived at some point in their lives, but it doesn’t tell nearly the complete story of my ancestors. Some might be completely missing from the calculation. The geographic regions may only extend to several hundred years ago. The genetic material that represents a focus in Poland may come from a different century than the material that represents Afghanistan. There is no way to turn this information into a description that has a strong relevance to who I am.

So if you decide to take a DNA test looking for some sort of insight into your genetic composition, keep these caveats in mind. These results seem to exist only to satisfy curiousity with a scientific “answer,” even if that answer doesn’t exactly answer the question people really have about who they are.

View the family tree without paying a membership fee

Over the past year, I’ve found to be an excellent tool for organizing and researching my family’s history. Unfortunately, sharing the information isn’t quite as easy. I’d like to be able to share the family tree with family members while allowing the rest of the world to see public examples I write about here.

I’ve invited family members to my tree on, and they can view some of the information, but they can’t view everything I’ve discovered without paying a fee. And if they don’t pay the fee, they are bombarded with advertisements and other annoyances in order to convince them to sign up for a paid membership.

I have no interest in seeing my family part with their money, so I’ve been looking for a better solution. I’ll continue to use services and software like, Family Tree Maker, and MyHeritage for my own management and research pertaining to my family history, but I will also host a public, free-to-access, version of the tree here on The interface is not as polished as it is on or any other commercial tool, but it provides access to the information to visitors and members, free of charge.

You can access the family tree website here. Currently, it contains almost 5,000 individuals, focusing mainly on my ancestors and their descendants, but I’ve also researched somewhat distant branches of the family and added them to the tree if I felt there was a good enough reason to do so.

In order to see details about living persons and other information like sources and certain media, you must log in after requesting a new account. Because of the nature of private information available, I’ll approve new accounts for only family members. Otherwise, only certain basic information will be visible.

There have been a few notable updates since my last posting:

After writing about piecing together more details about the Lepianski branch, I was contacted by a cousin who provided more details. I knew that several of Anna (Khiene Liba) Lepianski Kerstman’s brothers and sisters found their way from Yanowa, Poland (now Lithuania) to the United States, but I hadn’t discovered their whereabouts. The cousin was closer to this family and provided more information on Moisei Isaak Lepianski, who became known in the United States as Morris Lopinsky. I’m still on the look-out for Anna’s sister Esther (Ester Leia).

I most recently wrote about my relatives who perished in the Holocaust. Sophie Heimbach, my second cousin three times removed and a Sturmwald descendant, and her father both perished. Sophie’s two sisters, Frieda and Ida, both left Europe for the United States during the first decade of the twentieth century. I’ve been able to find more information about Frieda Heimbach, who married Christian Schaefer.

While reviewing my DNA matches using FamilyTreeDNA and Gedmatch, I’ve encountered several matches who are related to the Oppenheimer family, a prominent Jewish family from Germany. The Oppenheimers have many living descendants, and more than just a few are interested in tracking their family history. One branch of the Oppenheimers (Feygele, daughter of Samson Oppenheimer) consists of ancestors of a Royal Dutch family. I’ve known of my own connection to the Oppenheimers as one several marriages away from my own lineage, but I’m wondering if DNA evidence is pointing to a closer connection.

There’s no definitive conclusion as a result of DNA analysis yet, but I’m continuing to try to make sense of what is known.

In my research, I found this diagram of the Oppenheimer family tree courtesy of the Loeb Family Tree website. According to the author of the site, this tree was drawn by Samuel Dokow of Hemsbach in 1900.

Oppenheimer Family Tree c 1900

FamilyTreeDNA mitochondrial DNA results

In an effort to learn more about my family’s history, I’ve ordered several DNA analyses.

First, I ordered an autosomal DNA analysis from This test was an attempt to identify missing links in my family tree. Almost every week, provides me with a list of new potential matches — other customers who tested their DNA who may be third, fourth, or more distant cousins. Since the company provided me with the initial results, I’ve received hundreds of these new potential matches.

I have yet to confirm a relationship with anyone listed as a potential relative. These false positives may be a common problem among those with certain DNA associated with Ashkenazic Jewish ancestry.

The Family Finder DNA test from FamilyTreeDNA provides a similar type of analysis. Again, I have hundreds of matches; in fact, I sent a few emails to those listed as potentially being third cousins, but I’ve received no responses. This is an old service; those who received their results many years ago may no longer be checking the email addresses they provided originally, or just aren’t interested in determining whether there is a relation.

FamilyTreeDNA also offers testing of Y-chromosome and mitochondrial DNA. The Y-DNA test focuses on patrilineal ancestry. Assuming the analysis is correct, I fall within haplogroup G2c. Or at least, that is what I assumed based on the markers when I received the data from the test. Since then, FamilyTreeDNA has changed the way it describes haplogroups. I am now categorized as G-M201. My testing did not go far enough to determine my specific subgroup within haplogroup G — that would require a more expensive test. Part of haplogroup G migrated from the Middle East into Europe while another part migrated to South Asia.

Haplogroup K, FamilyTreeDNA
Haplogroup K, FamilyTreeDNA
When I ordered the mitochondrial DNA test from FamilyTreeDNA, I chose to full package, which analyzes the complete mitochondrial DNA sequence. This gives me the fullest picture of my matrilineal ancestry, which I share with my mother, her sister, and her mother’s sister. The designation for my mtDNA haplogroup subclade is K2a2a1. The individual in charge of the haplogroup K community on FamilyTreeDNA contacted me soon after, helping me navigate all the information available.

The image on the right shows the migration path for haplogroup K. The “Eve” indicated on the map isn’t the biblical Eve, it’s Mitocondrial Eve. All mitochondrial DNA can be traced back to this theoretical woman, the most recent common ancestor of all living women, who lived in Africa about 200,000 years ago. The K2a2a1 subclade of haplogroup K is identified as being exclusively Ashkenazi.

In addition to the proposed relatives listed among the Family Finder matches, the Y-DNA and mtDNA tests also provide a list of potential matches. Very few people provide their family trees, however, so it would be almost impossible to determine if there are any close relatives in the system without sending out mass emails.

Taking all the DNA tests into account, I appreciate the idea that I can trace the historical migration of my patrilineal and matrilineal ancestors back to practically the origin of the species, but those two lines make up a small portion of me. Going back four generations, my Y-DNA and mtDNA represent 12.5% of my ancestry (two ancestors of sixteen). Going back eight generations, the same DNA represents only 0.4% of my ancestry (two ancestors of 256). Go back 1,000 years, about 50 generations, and my Y-DNA and mtDNA represent about one five-hundred-trillionth of my total ancestry (assuming there has been no duplication of ancestors, but is also impossible because there were only about 400 million people alive in the year 1000).

That’s the extent of my satisfaction with these DNA tests. So far, the results have done little to identify relatives who might help me build my family tree. There has been one potential close call, but my skeptical nature prevents me from agreeing to a potential relationship without some sort of documentation, or at least some fairly convincing evidence.

FamilyTreeDNA Family Finder results

In order to determine more about my ancestry and genealogy, I ordered several DNA tests last year. The first was an autosomal DNA test from The purpose of this test was to find potential relatives by comparing a large portion of my DNA fingerprint to others who have also provided their DNA to the same project.

The results haven’t been very encouraging. Although the DNA database seems to be growing very quickly, and I receive a list of new potential every week, since July last year the best match I’ve found so far is a potential third cousin with a confidence level of 98%. As far as I can tell, this does not mean there is a 98% chance of this individual being my third cousin. I don’t know what these numbers really mean, and as I mentioned in an earlier post, I’m part of a community — Ashkenazic Jews — whose DNA tests are more likely to present false matches.

And that’s why I have hundreds of potential matches with at a confidence level of 96% or higher, listed as potential fourth to sixth cousins. I have checked the families trees of every potential match at this confidence level, for those who make their family trees available to others in the community, and I have yet to find a single confirmed match. The results are a little disappointing, but that doesn’t mean some of these matches aren’t cousins. There are a few obstacles:

  • I have details going back only four or five generations for most of my ancestors.
  • Most of my potential DNA matches have even less information listed on their trees.
  • To confirm a fourth cousin you need to have a great great great grandparent in common, and I only have information about a few of mine.

That may be because many other testers have not provided enough information about their trees, or that my ancestors may have siblings I have not yet discovered, but the results are a little disappointing.

One potential DNA match is a neighbor’s family from my childhood, and we may have found a link based on what might be my second great grandmother’s maiden name, but I have several conflicting records, and there are some other location discrepancies. It might be interesting to find out the girl next door was my third cousin once removed, but this is a long way from being a verifiable fact.

Close match DNA comparison, FamilyTreeDNA Family Finder
Close match DNA comparison, FamilyTreeDNA Family Finder
Late in the year, I ordered a similar test from FamilyTreeDNA, which this particular company is branding as “Family Finder.” Because the two companies have different sets of users, in theory, I could find different potential cousins by submitting my DNA to FamilyTreeDNA. Although FamilyTreeDNA supposedly corrects their matching results to take the problem with Ashkenazic DNA into account, my results, provided at the beginning of 2013, identify 230 potential distant relatives and two close relatives — second or third cousins. For the most part, FamilyTreeDNA does not link family trees to DNA, so I can’t peruse potential cousins’ families for matches. Emails to the two potential close relatives have not yet been answered.

What is the point of all this? Why spend so much money for DNA tests to discover cousins? At this point, the discovery of new blood relatives is unlikely to change much in my life. It seems to be for the sake of curiosity only. And with what seems to be such poor results so far, I’m not sure whether there’s a point in continuing to browse through strangers’ family trees looking for a possible connection. At the same time, I haven’t exhausted all resources yet. I have very little information about my ancestors who lived in Russia and Romania.

Sources from Lithuania, England, and Germany have been helpful, but these come from records that are already available online. To go further into my ancestry, I’ll need to find records on the ground in Eastern Europe. Perhaps a potential relative’s family tree provides some clues, like a brother or sister of one of my ancestors I haven’t confirmed yet. The chances seem low, however.

FamilyTreeDNA offers some insight

This summer, I ordered a DNA testing kit from FamilyTreeDNA. Having already completed autosomal DNA testing with AncestryDNA, I opted to order a Y-DNA 37 test from a different company. Testing the Y-chromosome offers different results than autosomal testing. The latter helps you carve a picture of your entire genetic composition, roughly equal parts from each ancestor in each generation. That is, each of your eight great-grandparents contributes roughly 12.5 percent to your genetic composition with these tests.

The Y-chromosome test and the mitochondrial DNA test are different. The Y chromosome provides clues only to patrilineal ancestry, your father, your father’s father, his father, and so on, back through time, through only the male contributions. The test uses mutations in the genes to determine ancestral heritage, and through this information, can classify every male into a migration group. Any males with matching genetic markers has a good chance of being related through a recent common ancestor.

I enrolled in FamilyTreeDNA and provided the details of my furthest confirmed paternal ancestor, Moses Landes. While I do have one document indicating Moses’s father’s name was Joseph, I don’t have any birth or death dates for him, so I didn’t include that information in my profile. I joined a group on the website where other Landes (or Landis) descendants meet online to discuss their findings.

My genetic composition does not fit with the other members of this Landes group. It’s not surprising, given that our surname does not extend very far. From what I’ve been told, this family did not go by the name Landes in Romania.

Haplogroup G Migration Pattern [FamilyTreeDNA]
The DNA indicates I am a member of haplogroup G2c, a branch of haplogroup G. This haplogroup represents Ashkenazi Jews, and some evidence suggests that the G2c branch reflects a migration out of Sicily to northeastern Europe in the Middle Ages.

FamilyTreeDNA links me to others who have had their DNA analyzed and whose DNA markers are similar to mine. There are no exact message for all 37 markers tested, but when limited to 12 markers, the site’s database offers me eight exact matches and 131 matches one step away. Evaluating 25 markers, I have two matches two steps away, one of whom is also two steps away when evaluating all 37 markers I’ve tested. There is a 91 percent chance according to FamilyTreeDNA than he and I share a common ancestor within 12 generations.

Unfortunately, I’m not sure where to go from here.

The Family Finder service, another autosomal DNA test, might provide me more insight into my relatives, considering FamilyTreeDNA’s database is more extensive than AncestryDNA’s at this time. The Mitochondrial DNA will provide migration information for my maternal ancestors. I’m not prepared at this time to spend more money on testing until I can glean more value for the tests I’ve already purchased.