DNA matches only appear to be from 3 grandparents

Question

I have a bit of a quandary and I hope some of the WikiTree DNA experts can advise.

My DNA matches appear to relate to only 3 of my grandparents: Both of my paternal grandparents and my maternal grandfather. This is puzzling me.

Now my maternal grandfather was 100% Jewish, so my matches are definitely skewed with an overwhelming amount of folks with Jewish ancestry.   

But my very top most matches are distant cousins from both of my paternal grandparents' lines.  

My maternal grandmother's maternal grandparents were first cousins who descend from a rather endogamous group of Bavarian villagers.  And most of her paternal ancestry dates back to 17th century colonial America. So, I thought for sure I'd see at least some matches to her line. 

In fact, I feared that the endogamy on both sides of my Mom's family tree would make it nearly impossible to find matches on my paternal line.  But this has not been the case.

My Mom so far refuses to test, which would make identifying her maternal line matches so much easier - at least on FamilyTreeDNA - as I could use FTDNA's Shared Origins feature and sort by ethnic group (all of her non-Jewish matches would clearly be from her maternal line).

So how else can I identify matches that relate to my maternal grandmother's line?

Comments

Any siblings or cousins of your mother that would be willing to test?

---

Unfortunately, my Mom only has one sibling - my uncle - who is the one that talked her out of taking the test I had already purchased... after she had initially agreed to indulge me.

My maternal grandmother had only one sibling, who did not have children. So my Mom has no maternal first cousins to test.

One of my Mom's paternal first cousins tested and she and her son show up among my top matches. That at least proves that Grandpa was indeed her Dad!

---

It would be interesting (for you, not the rest of us!) to know why your uncle talked your mother out of the test, and if that could shed any light on the mystery.

---

I was going to say that "siblings won't help", because they're on BOTH grandparents' sides, but perhaps that's not entirely true. For relatives THAT close, you might be able to get a good "read" for whether they are half-relative or not. It would at least tell you if you all have the SAME two grandparents on that side, and you could see if they're seeing the same thing. They might happen to have gotten a better match to a 3C on grandma's side - one that would stand out better in their list. It's always good to compare notes.

As to 1Cs, her mom doesn't have any, which may be at the heart of the problem.

---

My 73 year old uncle is convinced that the government will use genealogy DNA tests to frame people for crimes. Seriously.

I laughed when my Mom told me. "Do you really believe the government is looking to frame some random dude in his 70s?", I asked. She kind of agreed it was ridiculous, but still says no.

---

I can sympathize.  My brother refuses to take a Y-DNA test (even though neither my father no my grandfather had any brothers, so no one else I know can do it!) and all he says is that maybe someone in the future will invent some malevolent use for DNA that we haven't thought of yet.

---

You might laugh when you hear that people in "primitive cultures" refused to have their picture taken because they thought it would "steal their souls". I see this as a modern day equivalent of the exact same thing. It's a fairly new thing, that captures the essence of who they are, in a way - the paranoia we see isn't so surprising.

---

Yes, I knew that.  But never thought it rational.  The ridiculous thing is that my brother's autosomal DNA results are already on Ancestry, GEDmatch, and WikiTree.  I can't imagine how Y-DNA could be more threatening than autosomal.

Also, we've had a public-school education where science was taught, unlike the "primitive" people who didn't have that advantage.

---

Point is, your uncle's DNA could be used to finger somebody else.

---

I once talked with a woman who thought she might be wrongly accused of crimes based on DNA.  She didn't know anything of the details about how it worked, or the meaning of matches of various strength, etc.  The uncle may be in the same category.

Sometimes I wonder if these people who distrust DNA know how notoriously inaccurate eye-witness identifications can be.

---

RJ, I don't think my uncle would have any problem with DNA being used to identify a blood relative who committed rape or murder.

I do wonder if he worries that the daughter he put up for adoption might one day find him. Of course, she would show up as one of my top matches if she ever did test (at least on FTDNA, MyHeritage and GEDmatch anyway).

---

Yes, I have corresponded with several people who've identified parents through DNA without their parents having tested.  A half-sibling match is unmistakable.  For the most part, the reunions have worked out well in the long run, although the news can be shocking at first.

---

Suddenly it all makes more sense! The uncle's government conspiracy nonsense is really just a place holder for the real reason, which he doesn't want to talk about. The mom has picked up on this at some level, and is trying to protect him.

Well, they're probably just telling themselves that they can keep trying to hide this thing by doing that, but they're fooling themselves. Such a test would show up on Jana's test like a searchlight in the night - 830cM, plus or minus.

I've had several such tests show up on my results. The first was an adoption case, and I figured it out for her, from her matches, and from the research I had done on my relatives long ago. The father turned out to be a 2nd cousin of mine - her biological grandmother was somebody I would visit (along with my mother) from time to time.

Another 2C1R turned up, and once the test manager started talking to me I was eventually able to figure out that her grandmother, who was adopted, had to be an illegitimate daughter of my grandmother's uncle - she was born in 1893!

A distant relative I helped out, simply didn't know who her father was. Her mom remembered a story about him, but it was a one-time thing and she didn't recall his name. I realized that one of her matches - also related to me had to be her grandmother, and the two were united for the first time!

There are other, more distant cases, in various stages.

Then there's the people on my wife's test... It just goes on and on!

Answer 1

I'm editing this answer based on the response below:

The first thing that comes to mind is that you might have identified the wrong person as your grandmother.

Comments

Julie,  I think your eyes played tricks on you. My Mom was born in 1951, so it would've said she was born in the 1950s.

---

You are quite right.  So sorry.  It's old age, I guess.  The little type is hard to read!

Answer 2

Looking at your tree, the first thing that pops out is that the maternal grandmother in question had only one sibling, who in turn had no descendants.

So basically - right off the bat - you simply have NO second cousins (or 2nd cousins once removed, in the generation after you) on that side. So THAT alone might give the impression that "something's missing". These are the folks that would otherwise be showing up near the top of your list amongst the obvious relations from your other grandparents.

If anything, this is as you would expect - as far as I went.

But really, you should go further than I did. Was the previous generation - your grandmother's grandparents - more prolific? Out at that level, relatives should show up on your list starting somewhere past around 140cM (unless they're in your mom's generation).

I have a similar dearth of matches for one of my gt-grandparents. "Long story short", none of his siblings had any children (well, a couple of them simply disappeared, so there are no KNOWN descendants). So I have no 4rd cousins on that side. Gt-grandpa didn't have much in the way of 1st cousins either. So in a way, it's a sort of confirmation that he's really the biological ancestor (if something untoward had secretly gone on, the bio-dad probably would have had more relatives!). But on the other hand, it's just a "hole" in your DNA research.

Maybe you just need to track down as close a relative as you can find on that side, and see if they'll test - and you might need to do more than one, since it's not 100% that you'll get a match, beyond 2C1R..

Comments

Thanks, Frank!

My maternal grandmother had at least 2 paternal first cousins (not sure if her youngest uncle had kids - can't find him in the census). And she had at least 8 maternal first cousins.

One generation further back and things get very prolific on both sides.

---

You're welcome!

Yeah, I had just gone back and looked. It looked like grandma's dad was one of three brothers. I think I found the one you're talking about, and he doesn't appear to have had any kids. The other one was harder for me to find, but yes, I'm seeing him with two daughters. At least one married, but didn't get a read on whether there were descendants after that.

Even if those two of your mom's cousins HAD descendants, there might be too few of them for one to pop up who isn't worried about having their soul stolen.  

How's it looking on the maternal side?

---

On another seemingly related note, I have 10 living 1st cousins on my father's side, and not a single one has tested - at least I haven't seen them on AncestryDNA or GEDmatch. Further, not a single one of their children (there's about 20 of them, all grown) have shown up either. I don't know that they're necessarily worried about "their soul", or the govt, or whatever - they just never seemed to have any interest in genealogy, whatsoever.

None of my 2Cs on that side have tested, either (there's something like 20 of them). But 5 of their kids have popped up. I've messaged a few - they ignore me. My dad met their parents a few times when he (and they) were kids, but that's about it - they never really knew each other much at all.

On my mother's side, though, I've got 6 1Cs, and half of them have tested, as have several of their children.

---

I have that situation also on my paternal Grandmom's father's side. Her father's older siblings had a ton of kids, one of his uncles had 11 kids, one aunt had 8 kids, etc. and yet so very few descendants have tested. I found more folks who tested via AncestryDNA, but not as many as I'd hoped and they aren't a responsive bunch at all.  

I'm not thinking my maternal Grandma isn't my Mom's mother or anything. Though I wouldn't have been shocked if there was a NPE situation with her parents... but my uncle is the spitting image of his maternal grandfather, so I doubt that's the case.

I am just curious how I might better filter my matches, if that's even possible.

---

As far as filtering goes, you might be able to do fairly well with just eliminating those who either match your dad's side, or who show up as having significant Jewish ancestry. Normally, I would discount admixture as not being useful for "real" genealogy,  but in cases like this, where there's a distinctly different ethnicity vs the others, and in a significant amount, it could be useful.

I hadn't really looked at your grandmother's mom before, but I see how that's a problem, too - they haven't been in America all that long. Also, while they did have at least SOME children, it doesn't look like they're super prolific. That's more like my own paternal grandfather. He certainly has matches, but its's something like two or three dozen - vs the hundreds on my paternal grandfather's side. And that's on AncestryDNA, where there's at least twice as many people as other places. PLUS, these particular relatives seem pretty inclined to actually test. PLUS, it seems like most of them have actually answered me. So, now that I think about it, they're a pretty cool crowd!

On the bright side, the intermarrying aspect should bump up the centimorgans somewhat, and the likelihood of getting a match, if they actually DO test. I've calculated that before, and the factor is 17/16 - not a huge bump, but a bump nonetheless.

Answer 3

Women chop their DNA into more pieces than men when passing down to their children. About 1.6 times as many if memory serves This means that the your maternal grandmother's DNA is in shorter pieces than your paternal grandfather's. So finding a match will be harder and since the generation estimating algorithms use an average of male and female,  the matches will be estimated as being more distant than they actually are.

You have the same mitochondrial DNA as your maternal grandmother but is not recombined and doesn't mutate very fast making it challenging to use for family history research.

Tim

Comments

That's very interesting, Tim. I did not know that.

---

Here is a link to an example research paper. The rates vary by chromosome but note the female centimorgan total is higher than the male.  (You will have to subtract the X chromosone figure to make it a fair comparison as males cannot recombine that.)

https://www.biostat.wisc.edu/~kbroman/teaching/statgen/2004/refs/decodemaps.pdf

Answer 4

How many close matches (20 cM or higher) do you have?  It might be helpful to consider a cluster analysis to help you spot unidentified clusters of more distant relatives that may belong to your grandmother's line.  There are several sites out there now that have automated the clustering process.  (I prefer the one at DNAGedcom.com personally, but it's not because there are any problems with the other ones.)  I did a cluster analysis of my uncle's ~2400 Ancestry matches (4th cousin or closer) and it's been really helpful.

Comments

Thanks, Lisa!  I'll have to look into that more.

I really have an overwhelming amount (thousands) of 3rd to 5th cousin range matches that are folks with all or mostly Jewish ancestry.

---

Yeah, that's a lot to slog through!  The nice thing about the clusters is that it may lump all of those together and make it easier to spot groups that don't fit in with that endogamous group.  I'd say give it a shot.  You may have to tinker with the min/max thresholds to get something manageable.