... by two different companies has left me well, dumbfounded in a good way.

Initially I purchased the 23 and Me kit for Mother's Day a couple years ago. Was happy to get an inkling of my family's origins.

Then yesterday I reconnected with a genealogist cousin who'd had hers done by Ancestry and was skeptical of its findings. So she uploaded the same test to her My Heritage account and told me its findings make so much more sense to her.

I didn't realize I'd already uploaded mine some time ago if not for her sharing that with me. So when I checked its pie chart of my origins it included some findings 23 and Me hadn't. For example, I learned I'm 5.4% Greek (news to me.)

I'm 10% East European, from the Balkans. Also news to me. And I'm more than 25% Scandinavian! That is another surprise. My Ashkenazi Jewish took an upward slide from 11.4% to 12.7%. I seem to be more Welsh, Scottish, and Irish than English per My Heritage results. 23 doesn't include the Welsh or Scottish at all.

What a pleasant surprise to learn of my new findings thanks to my Montana cousin. She and I are first cousins who met through Findagrave research. Small world! :0

Don't take them as "gospel" or anything...

https://now.tufts.edu/articles/pulling-back-curtain-dna-ancestry-tests

("...you have to look at the percentages you receive back with skepticism.")

The interesting comparison about this one is its specificity the other one lacks.

23's breakdown wasn't as fine tuned, and left out some ethnicities altogether. Like having Greek, or Balkans for example.

That's kind of the point of the article I linked to -- that the "specificity" is quite possibly artificial.

Mine gave me 1% Southeast Asian ancestry. It's a total crock. There's no Southeast Asian in my ancestry for at least 20 generations that is known. So what probably happened was they mistakenly matched up some SNP that was either an artifact, an error, or actually there with some SNP that their (limited) database said indicates "Southeast Asian," and assigned me that background -- erroneously.

Lots of reasons not to take these results with anything other than skepticism, and certainly good reason to assume their "percentages" are shaky at best, completely false at worst.

But, hey, have fun with it :)

“at least 20 generations that is known“

Fortune cookies in the cookie jar seems like something you’d keep quiet.

https://www.deseretnews.com/article/900043742/why-do-dna-results-change-ancestry-says-new-data-is-driving-better-origin-drill-downs.html

University of Utah professor of human genetics Aaron Quinlan said when it comes to tracking ethnicity via DNA, it's important to distinguish between a person's genealogical records and genetic evidence.

"Our sense of genealogy is a narrative which can be supported with historical and family records," Quinlan said. "And, while many people think we can track DNA back many generations, we really can't draw useful information past about seven generations."

Quinlan explained that any individual's genetic makeup represents about half a contribution from each parent. One tier up the family tree, and each of those parents is getting half of their genes from a parent, and so forth. Trying to assemble a picture of our ethnic histories, based on that tracking, gets hazy very quickly.

"The way I like to think about it is like a patchwork quilt," Quinlan said. "The further we reach back, the smaller the patches of genetic information remains. Companies like Ancestry and others are trying to compare one patchwork to other patchworks."

Funny but I've always felt a little Greek if only for the reason that my dad gave me a Greek name at birth. My middle name is Hebrew after my Jewish grandmother. My Greek name translated into Hebrew means Naomi &/or David. Then with my middle name of Ruth, they're "all in the same family!"

I doubt he or mom knew that, but it is what it is. :)



Edited 1 time(s). Last edit at 12/04/2018 02:42PM by Amyjo.

Is with the data bases. I'm hopeful Simon Southerton will see this and elaborate.

My take is these companies are more interested in "marketing the drama," as as evidence, the reality that they differ in their interpretations, and that should be illuminating.

Some of these commercial companies are of varying accuracy in evaluating your DNA

This is a big difference from DNA being read in criminal cases.
There, DNA readings get more accurate with each passing year.

My family therapist told me this yesterday when I shared my "new" findings with him. He said the DNA companies have made tremendous progress in just the past couple of years, providing more specificity than ever before just from advances they've made technologically. And he was talking about Ancestry, My Heritage, 23 and Me, et al, not just the DNA tests used in criminal cases.

It is from the genealogical databases some cold cases have been solved in fact, from the DNA submitted by relatives of criminals.

Aside of that, even the ethnic backgrounds are getting easier to trace.

I think you'd be good going with the Scandinavian part :)

I prefer being a Greek muse!

The Scandinavian was a complete surprise. I knew there was some, according to my dad. But not 25+%.

The Scandinavian DNA could well be from the times the Danes ruled most of England.

Isn't it interesting no one gets "American"?

My main blood lines have been in north America for over 450 years. Why doesn't that ever show?

Yes, Danish blood is all through eastern and southern England, also southeastern Ireland (Dublin was a Viking city) and the Isle of Man.

More generally, the Vikings spread throughout most of Western Europe and along the rivers as far as North Africa, the Black Sea, and Moscow. Given that this happened a millennium ago, virtually anyone with European derivation today has Scandinavian DNA.

That may not show up, however, in these ancestry tests since those tests essentially choose a random date and look at things then. So if they look at the situation in 1500, they would define English as including DNA that was originally Viking. The person whose genes were being tested would be told that the DNA came from England, which is true at 1500 but not at 900.

My guess is therefore that Amyjo has much more recent Scandinavian DNA. That from the Viking era would not register as "non-English" if the comparison date were anything after about 1200 or 1300.

There is an episode in the series "Adam Ruins Everything" that deals with DNA. Those who are interested might find it thought provoking.

Thanks for the suggestion. Will check it out.



Edited 1 time(s). Last edit at 12/06/2018 05:06AM by Amyjo.

My wife and I recently had our genomic DNA analysed by a European-based company (Living DNA (LD)). I chose them because they have the most comprehensive database of people in the United Kingdom, which is where we thought all of our ancestry came from. We wanted to know the breakdown in the UK. Our experience highlights some of the critical factors that influence the results.

LD's database includes thousands of individuals from across the UK whose 4 grandparents were all born within about 80 miles of each other. These people very effectively represent local populations as they existed before the 20th century when people started migrating all over the place. LD estimates the proportions of your DNA that come from different counties and regions in the UK.

My wife's parents come from working class families, her father from Northern England (Yorkshire) and her mother from southern England (Southhampton). LD predicted almost exactly half of Jane's DNA came from counties in Northern England and half from Southern England. They picked up a little Scandinavian DNA which is not surprising given the Norse migrations into northern England. The estimates for the different counties matched well with Janes paper genealogy.

My results were a little more interesting. My fathers side came from Shropshire near Wales and my mother from Scotland and East Anglia. Again, the estimates for the different counties matched well with my paper genealogy, with one notable exception. I have about 8% Northern Italian DNA. I have no northern Italians in my paper genealogy. It is almost certain this DNA came in on my father's side because when you add up the counties that fit with my mother's paper genealogy you get almost exactly 50%. While it would be easy to dismiss this as an error, I think its probably correct for a couple of reasons. Firstly 8% is a significant amount of Italian DNA. Secondly, my father's side of the family moved all over the place in England, and lived in London for a while. Then his father migrated to Shanghai in the early 20th century and his mother's parents migrated to Hong Kong in the late 19th century. They would have had far more exposure to foreigners than my mother's ancestors. It's quite possible that an Italian stallion snuck in there somewhere.

Its important to keep in mind that the quality of the database the company uses has a huge influence on the results. You also need to keep in mind that the level of non-paternity (sneaky phuckers) can be very significant in some families and often would have gone unnoticed. In one study by Brian Sykes (godfather of human genetics) he found rates of non-paternity of 2-3% per generation. In some families it could have been much higher.



Edited 1 time(s). Last edit at 12/05/2018 09:49PM by SimonS.

8% is not all that distant of an ancestor/s either.

"Non-paternity" is a new word to me, but very useful.

Thank you!!

Tevai

[biological daughter of an "inconvenient" father]

The science is fine.

The commercial products often promise more than they can deliver, present simple results when the reality is complex, and detract from a key fact: more depends on you than your ancestors:

"There is little a geneticist can tell you with this information that will outweigh standard lifestyle advice: Don’t smoke, eat a balanced diet, exercise regularly and wear sunscreen"

https://www.scientificamerican.com/article/how-accurate-are-online-dna-tests/

This is a consumer product. They'll do what gets your money.