Did your DNA test results change when using a second company to do the test?

[Clarki]

TL;DR Summary

My DNA tests are not the same. 23&M and MyHeritage did these, and the lower percentages in the make-up differ.

Here's the story:
I did the 23&Me and it told me that I was about 90% X but then there was among the smaller %ages a Y = French
Later, about a year later, I checked it again. The French ancestry had disappeared. There was still 90% X, but now I had a Y = Scandinavia.

So, I ran the test again with MyHeritage

X was about the same at about 90% X. Scandinavia had vanished. New Ys = Finnish and Iberian.

Is it normal to get this sort of variation on the lower %ages between tests?

(I also wonder about the basis of 23&Me changing French to Scand.)

What are you thoughts?

 

[Clarki]

Follow up,..I just read somewhere else that 23&Me changed their system of assessment. This meant that results changed. But...if a system change means that French vanishes and is replaced by Scandinavia then what is the point of it at all? The entire point is to know origins. France and Scandinavia are not the same...ugh! See my facepalm.
I still don't know what Iberia has to do with any of it, they are totally different genomes in this thing.

 

[Keith_McClary]

It would be nice if they gave the raw data and the algorithms used to compute X and Y (including their definition of French, Scandinavia, etc.) .

 

[Clarki]

It would be nice if they gave the raw data and the algorithms used to compute X and Y (including their definition of French, Scandinavia, etc.) .

Yes it would indeed. My old Scandinavia DNA could be my new Finnish DNA. But, where does the Iberian stuff suddenly appear from. I will do a 3rd test with AncestryDNA.

 

[Ygggdrasil]

Here's a good piece about DNA ancestry testing:

When it comes to ancestry, DNA is very good at determining close family relations such as siblings or parents, and dozens of stories are emerging that reunite or identify lost close family members (or indeed criminals). For deeper family roots, these tests do not really tell you where your ancestors came from. They say where DNA like yours can be found on Earth today. By inference, we are to assume that significant proportions of our deep family came from those places. But to say that you are 20 percent Irish, 4 percent Native American or 12 percent Scandinavian is fun, trivial and has very little scientific meaning. We all have thousands of ancestors, and our family trees become matted webs as we go back in time, which means that before long, our ancestors become everyone’s ancestors. Humankind is fascinatingly closely related, and DNA will tell you little about your culture, history and identity.

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

There has likely been sufficient mixing between human populations (especially populations in Europe), that it would not surprise me if most European populations were very similar and it would be hard to distinguish ancestry between specific regions of Europe.

 

[Keith_McClary]

A.J. JACOBS: Well it depends on which company you're talking about. I took all of them and they were not all the same. One of them I liked was because it said I was 97 percent Jewish, which I knew. I knew I was predominantly do it. But it also said I had two and a half percent Arab. So I liked the fact that I had the Middle East conflict in my body. I embody that literally.

 

[Clarki]

OK, but again.
Two different companies. One said Northern European DNA but no Southern DNA.
One said Northern European DNA and Southern European DNA
Does lab testing miss such a major difference? North and South are different genomes.
"Iberian" DNA should appear on both tests because it is a different genome - why didn't it show on both tests? Is it so easy to miss? So, I dunno, it goes around and around. I'll do a third test just to see if it falls one way or the other.

 

[jim mcnamara]

Okay. You are not getting the point. Your assumptions need some tweaking.

200 years ago almost all people in the world were born, grew up, and died within about a 50 mile radius.
Because of unrest, famines, and wars people in the 20th and 21st centuries have moved and are still moving. An incredible amount. Look at the recent flood of refugees from Lebanon.

So pretend I am a geneticist and I sample genes in people from, say, a very isolated population. I would expect to see certain marker genes, right? Nope. That does not happen anymore. I could see all kinds of anomalous, out of place genes. Example: Genes from Northern Germany are very common in La Cumbrecita Argentina. People there are fluent in German. Can figure out why? Is it magic? So what if your Mom said she was originally from Argentina? Then what? Then there is Dearborn MI. Arabic. Or central Market Street in San Francisco. Chinese. The list goes on.

These tests work on probabilities - X haplotype implies Y parentage about Z% of the time. And with more datasets or different datasets sometimes you do not get identical answers to that question. Plus, the tests can only look at a relatively small number of the 100K genes you have. And still be affordable. Ancestry.com has the largest data set as of 2019.

Per:
https://thednageek.com/genealogical-database-growth-slows/

I oversimplified but so did you. Testing companies explain markers they use and the probability of them being perfectly correct. Go do some reading.