RootsChat.Com
General => Ancestral Family Tree DNA Testing => Topic started by: John Scott 1959 on Sunday 20 October 24 05:36 BST (UK)
-
My wives DNA from her father's side has me a bit bamboozled. Her surname is of German origin, and I can trace her family on her father's side back to migrants to Queensland in the mid 1850s through her 3rd GG and his family.
My wife has 12 DNA matches that also track to the 3 times GG but none from 4th onwards. What is puzzling me is that she has no German DNA at all on her paternal side. Was the 3 x GG's father, not his biological father? But then surely the 3xGG mother is of German origin.
Am I missing something obvious?
-
My wives DNA from her father's side has me a bit bamboozled. Her surname is of German origin, and I can trace her family on her father's side back to migrants to Queensland in the mid 1850s through her 3rd GG and his family.
My wife has 12 DNA matches that also track to the 3 times GG but none from 4th onwards. What is puzzling me is that she has no German DNA at all on her paternal side. Was the 3 x GG's father, not his biological father? But then surely the 3 x GG mother is of German origin.
Am I missing something obvious?
-
Hi,
There are two main parts to my response, and i will start with the simpler one first which is on you wives 'German DNA'. Basically the 'Ancestry ethnicity estimates' really should not be taken to seriously, if you look at the post on the recent changes to them, a lot of people don't think much of the break downs they give. Despite what ancestry say they are not an exact science, but instead are just a view of how the DNA compare to reference panels of other testers.
On the second part, the no matches to 4th GGP, are you basing this on the thrulines/common ancestors or you own research into the matches ? Based on the DNA inheritance process, it is quite possible that you wife is just unlucky not to have inherited much DNA from that individual set of relatives, which means you wont see any matches as they are below the accepted thresholds. However it is also possible that nobody who is descend from them (other than on your wife's 3 GGP lines) has tested, There is real no simple way to tell. The best you can do is to make sure you investigate as many of the shared matches of those 12 matches you have identified, and try and work out how they are related.
Richard
-
It is very possible that no “German DNA” has been inherited as Richard writes.
I keep posting this chart as it visually shows how DNA can be inherited.
Do note that each DNA tester will have their own, unique version of this chart.
-
If you tested with Ancestry, upload your raw data to MyHeritage. They sometimes throw up more European matches.
-
Biggles, thanks for posting the chart, I haven’t seen it before. Really helps to put it in perspective.
-
Thanks for the replies. Much appreciated and very interesting. It seems to not be an exact science. My wife does show 17% German DNA from her mother's side that comes from her GG mother. It is interesting that her German father's DNA could have not been inherited only one generation further back.
I have obviously been thinking about this and as my father-in-law is still alive, 92 years old, I might get him to take a DNA test.
-
Getting F in L to test would be good.
His DNA Matches should be about two times the cM as your Wife’s so that may elevate some of her lower cM matches into the 20cM plus results where Shared Matches become visible.
Do remember that Ancestry’s “gene pools” that they use are still going to be restrictive and for German ethnicities they may well be based upon a German enclave in the USA.
Watch this and I suggest you start to Group your DNA Matches
https://www.youtube.com/watch?v=btQwRIyhuns
-
The first thing to understand is that there is no such thing as "German" DNA, any more than there is English, Scottish or Irish DNA. All there is, is DNA which is more or less prevalent today in people with Scottish, Irish, German or English descent.
The second thing is to consider what Ancestry actually does - which may not be quite what the marketing blurb sometimes suggests.
What Ancestry does, is form reference panels of, in their words "people with deep family roots in a specific geographic area or cultural group". What that means is they use people whose families can be shown to have lived in a particular region for several generations. So their German reference panel (which at 2000 members is the same size as the Scottish and Irish reference panels) consists of people who have established roots living in Germany for several generations.
They then analyse the DNA of the people in the reference panel, measure the percentages of people who have each of 1001 separate segments, and use these percentages to decide on the "origin" of any particular segment.
Let us call one of these segments the "Abba" segment. It turns out the Abba segment is quite common in the Scottish reference group - about 40% say. It is also quite common in Germany (let's say 38%) and Norway (62%), but as I'm sure you've guessed already, I'm going to make this particular segment most common in Sweden (75%, say). (Just in case it's isn't obvious, I'm making these numbers up, I have no idea what typical ranges they get.) The point is, now that they have have measured the percentage abundance of the said segment, the "Abba segment is now categorised as "Swedish".
Now they analyse my DNA. It turns out I have the said Abba segment, so Ancestry now allocates 1 of my 1001 segments as Swedish.This does not mean I must have gained that segment from a long lost Swedish ancestor - it just means I have a particular segment of DNA which is more common NOW in people whose families lived in Sweden for several generations in the past than in any of their other reference groups.
Should Ancestry find 10 such "Swedish" segments, that is 10/1001, ie roughly 1% so it now classes me as 1% Swedish. BUT, because each of those MAY in fact be from Scottish ancestors given that, as with my earlier example of my entirely fictitious Abba segment, those segments may also be relatively common here in Scotland, they put an error range of between 0 and 1% on my Swedish ancestry.
Given that the UK has during it's history been invaded by Romans, Angles, Saxons, Jutes, Vikings from various parts of Scandavia as well as Norman French, has provided refuge to many exiles from religious or other forms of prosecution, and has in turn fought over or invaded large parts of the globe, it is perhaps not surprising that we share a lot of those segments with many other groups.
What all that means is that what used to be called ethnicity estimates are just that - glorified guesses based on statistics and probability - which only indicate that you share DNA with people whose ancestors are from a particular region in the past. If you want the fuller picture look at the ranges on the given identified regions.
Most importantly, always remember that DNA is only one more piece of evidence that might link you to a given ancestor.
-
Good explanation Albufera 32.
To add to the “unexactness” of the ethnicity estmates I remember a while ago reading a document which gave numbers of the people in some sample populations. For some of the rarer populations the bumbers were really very low - a hundred or so, or fewer. That has propably increased over the years.
Participants also needed to know that their lineage as far back as their great grandparents (someone will correct me if I am wrong) came from a specific area. Unsure if they needed a paper trail to prove it, or how it worked if they came from an area without a paper trail.
So it seems like they largely took people’s word for it, which I imagjne must be open to guesswork and error.
-
The first thing to understand is............
I am unable to thank you enough for making this such a simple and easy to understand explanation of what has been a difficult to understand subject for me personally.
-
The numbers in each reference panel are listed on Ancestry in the section on DNA.
In all there are 107 reference panels, with a total of 116 830 DNA samples making up those panels, so the average size of the panels is about 1000. Almost all the European panels are 2000, whilst the lowest I could see was Burusho with just 17, followed by Indigenous Arctic at just 24.
Perhaps partially explaining the sudden appearance of Iceland in many people's ancestral regions, Iceland, one of the new panels is just 295, whilst Cornwall, which I think is also new, is 1090.
I should also mention that my explanation is a little simplified, I think Ancestry's clever algorithm is a little bit more complex than I described, but in principle, it works rather like the way I described, I believe.