The datasets for those assignments aren’t very complete, really they just assign you to the groups that match their reference data-sets. When you consider the power of two growth in potential ancestor counts it becomes pretty insane anyway.
Like everyone of European decent I can trace myself as being a direct descendant of Charlemagne. Through happenstance I happened to stumble on a connection that I knew was a decedent and traced it back for fun. Charlemagne is one of 2^42 or 4,398,046,511,104 potential ancestors at that time. Obviously that is more people that were alive back then and for most of human history marriage between cousins was the rule but it does demonstrate just how much admixture there was.
When it comes to autosomal DNA matches and assignment, it is quite possible to have been from one area, but that the testing companies reference population that also had those STRs was in another location. (STRs are selection bias too BTW and they are only testing a tiny part of your DNA)
Mathematically, for those of European decent, anyone who was alive in around the year 900CE who has living descendants is also a direct ancestor to you too.
These tests can give general clues, and are entertaining but on the autosomal side it is generally best to consider them purely entertainment.
I should also note that some of those assignments are based off of archaeology assumptions or even old racial theories that are commonly being proved to be false due to DNA testing.
Basically language and cultural influences are not as tightly coupled to gene flow as was assumed. There have been a few studies that have actually delayed publication because of the cognitive dissonance related to their findings compared to the cultural assumptions. I would expect that these models will dramatically change over the next few years as these surprising results are reproduced and validated.
TL;DR, New empirical data shows that there was a lot more intermixing that traditional archaeological methods accounted for, and most of the commercial services still make assumptions based on those traditional archaeological methods.
I should not that I am not saying it is useless, my tests returned a lot of Basque DNA that was confusing, But I had one grandparent who was thought to be Welsh and Irish. While tracing back the surname of “Sage” I discovered it was “Le Sage” from Charles V. (we all are descended from royalty due to the effect described in my previous post)
It ended up that both sides were actually Huguenots, who had emigrated to the UK under Queen Elizabeth.
So I am not claiming it is not of value, just that it is not definitive.
I don’t remember where I found it (and can’t come up with a way to Google it), but I recently read a story where a pair of families actually did bring the wrong baby home from the hospital (much easier to do back in the day). One family was mostly Irish, the other Jewish/Eastern European.
I believe it was the not-really-Irish person who ran the first test, and then spent a lot of time trying to track down biological relatives.
Why doesn’t Elizabeth Warren just get a DNA test (along with any convenient cousins)? Tests will most likely detect as little as 1% Native American ancestry, right? If not, “Sorry, family history was a myth after all; move on; nothing more to see here.”
For one thing, because she doesn’t need to. Her claim was that family legend said that she had some Cherokee in her. Evidence for her claim, if any, would come in the form of a statement from some older family member, but even that’s hardly necessary, since it’s such an unextraordinary claim. A DNA test might be able to reveal actual Native American ancestry, but she never claimed any actual Native American ancestry, just that there was a family legend of it, and DNA won’t tell you anything about what legends your family was telling.
I think it stretches credulity to say that she did not claim some Native American ancestry. It can be true that she both claimed to have Native American ancestry and that her evidence is what she was told by her parents and grandparents. I don’t think we require people to provide DNA evidence every time they make a claim about their ancestry in order to say they made a claim.
I’m not trying to turn this into a political discussion, but I think it’s clear that she did claim Native American ancestry. Whether she should get a DNA test or not is a political question, and not one that I would answer in this forum.
Let’s say parent A is half blue, half green. Parent B is half red, half yellow. They have two kids together.
You would expect each child, on average, to be 25% blue, 25% green, 25% red, 25% yellow. And if you had a large number of kids, that’s what the numbers would average out as. However, in theory, the number for each color could vary from 0% - 50%. It would be possible to get one child that’s half blue and half red, and another child that’s half green and half yellow, thus sharing essentially no DNA. In practice, the odds against this are astronomical (1 over 2^23, if you ignore recombination; much much much less likely if you take recombination into account), but it is technically possible.
Even identical twins can get back different ethnicity percentages, or the same person having the test run multiple times. They shouldn’t be taken too seriously.
Since this is GQ, let’s just say you can’t know that. Very often, claims of NA heritage turn out to be not true, but plenty of Americans do have such ancestry. IIRC, her family lore is that there is NA ancestry on both sides of her family, so who knows?
It depends, The highest end commercial tests check for about 700,000 SNPs, but this is only a portion of the roughly 10,000,000 SNPs in the human genome. This also depends on the quality of the testing companies data set, where they may only have a few dozen reference individuals among millions of a population.
Also consider if her “pure” Native American lines Daughtered or Son-ed out, and there were no samples on some form of line that is believed to be pure. That would result in non-matches for untested SNPs, or if that family line was corrupted before collection could be even tested.
This is just one reason why this concept of race is not a viable scientific concept, and is only considered a biological trait in pop-science. There is a good reason why lineage is purely considered based on a reporting members beliefs, as anything else is really trying to assign more value to it than actually exists at a biological level (which is almost none).
The reality is that the only thing real about race from an objective level is the impact that the social construct that created it.
I should also note that there is a problem with assigning genetic markers to being Native American, because our understanding on how that migration happened is far from being incomplete. I have well established evidence that I am a direct decedent of Thayendanegea or Joseph Brant, but the DNA companies will associate those markers with Pacific islanders and not dare associate that with Native Americans due to this limitation.
There is also an issue where I have a lot of Finnish and Siberian ancestry which closely ties me to ancient samples like Clovis, but I would have that association without a direct relation to Thayendanegea.
Even if you consider mtDNA or Y-DNA the picture is far from exact, where I have a shared ancestor with the Rurikids and lots of modern day Tibetans.
The problem is that pop-science is trying to attach modern technology to cultural myths and the disproven hypothesis of eugenicists.
Race is a purely a social construct, and is not a biological trait.