Y-DNA 37 Match and No Autosomal Match?

[shandy4473]

Good Day Everyone,

I have another case which I would like to request some input on please. Has anyone seen where to two males, A and B, both match at the Y-DNA 37 marker level and yet share 0cM at the autosomal DNA test level?

The Y-DNA 37 marker test between the two males has a genetic distance of 1.

My assumption is that the two males, A and B, are simply not related in a genealogical time frame. The autosomal DNA test was Relative Finder. The Y-DNA 37 marker test was performed at FTDNA.

Any thoughts???

Thanks
Steve

[travers]

Read the FAQ for Family Finder. There is a 90% chance of 3rd cousins matching on an autosomal DNA test and the chances of matching drop considerably beyond the 3rd cousin level. Same would apply to Relative Finder.

[zarlor]

Depending on the Haplogroup, especially, a GD of 1 on a 37 marker test could be quite far back. I have one 37 marker match myself with a GD of 0 that drops to a GD of 5 on a 67 marker test. And as mentioned above there is still a 10% chance even at the 3rd cousin level that someone you are related to won't show up on the FF test. If they are farther back then that your chances of not finding an FF match with them increase significantly.

One other item I will note, though. You say these two tests match 0cM on an autosomal test, but are you using FF as your gauge or are you actually comparing the two data files directly (like on gedmatch.com, for example)? If the latter then it might be worthwhile to upload your respective data files to gedmatch just to see if you can find a smaller segment match that FTDNA wouldn't show you. It's still quite possible you won't find any significant match there, of course, but it might be interesting to see.

[shandy4473]

Relative Finder

One other item I will note, though. You say these two tests match 0cM on an autosomal test, but are you using FF as your gauge or are you actually comparing the two data files directly (like on gedmatch.com, for example)?

Hi Zarlor. How are you? The 0cm was shown at 23andME via Relative Finder. Both parties took the Relative Finder test.

The Y-DNA 37 tests were done at FTDNA. One of parties has a Family Finder test done in addition to Relative Finder. There was a 23andME transfer purchased at FTDNA where the data file at 23andME will be transferred to FTDNA for the 2nd party.

Then we can see what Family Finder says on comparing both.

Thanks
Steve

[Geneadict]

My assumption is that the two males, A and B, are simply not related in a genealogical time frame.

Not a good assumption due to random nature of autsomal recombination making it more likely to NOT have autsomal segment share with anyone related beyond 3rd cousin.

One step off at 37 markers may still have common ancestor before genealogical time frame, but can not ascertain such from lack of a segment match in autosomal test. Enhancing to Y67 or Y111 may provide some additional insight to likely MRCA, but even these may not be conclusive.

[shandy4473]

Genealogical Time Frame

Not a good assumption due to random nature of autsomal recombination making it more likely to NOT have autsomal segment share with anyone related beyond 3rd cousin.

The assumption was that they aren't related in genealogical time frame. That means they still could be related, possibly right directly outside the window of what's defined as genealogical years. In my opinion, it's a fairly safe assumption. Especially since the probability of these tests missing an actual 3rd cousin to distant relationship is low.

In addition, they clearly are related (possibly non-genealogical time frame) via a male common ancestor since they both match at 36 markers on their Y-DNA, even though Y-DNA isn't autosomal DNA.

[dwight]

But that's not correct. The odds of an actual 4th cousin not matching are certainly not "low", and for 5th and 6th cousins, the odds are very high that you will not share minimal DNA for a match.

Especially since the probability of these tests missing an actual 3rd cousin to distant relationship is low.

[shandy4473]

Okay

But that's not correct. The odds of an actual 4th cousin not matching are certainly not "low", and for 5th and 6th cousins, the odds are very high that you will not share minimal DNA for a match.

Okay. Thanks for the correction. But my point is that they still may be related directly outside of a genealogical time frame. Just because the test doesn't show it, doesn't mean they aren't related outside the window of a genealogical time frame. My initial assumption, which could turn out to be wrong, is still a safe assumption.

[Geneadict]

Okay but my point is that they still may be related directly outside of a genealogical time frame. Just because the test doesn't show it, doesn't mean they aren't related outside the window of a genealogical time frame. My initial assumption, which could turn out to be wrong, is still a safe assumption.

I suppose anyone is entitled to define for themselves what they consider to be a "safe assumption", but I would think most people with experience using both Y STR DNA and autosomal DNA for genealogy and the facts as presented (i.e. no other info to support the assumption) would probably disagree. Happy for others to comment on their perspectives.

I personally would only consider something a "safe assumption" if it were "highly likely (or unlikely)" to be the case (depending on the assumption), or at minimum "more likely than not". You simply don't have enough info to determine this, so what's safe about assuming?

The person could be a 4th, 5th, or 6th cousin which for most people would be within a genealogical time frame (abt 150-200 years). If you can add geography or non-intertwining "known" ancestry for both into the evaluation, perhaps you can make the case for your most likely not having common ancestor within genealogical time frame, but might still be hesitant with possibility of NPEs. In this case additional YDNA marker mutations at Y67 or Y111 might provide you with more information to make a "safe assumption".

In this case, the lack of the autosomal match only tells you that you most likely are not related as 3rd cousin or closer. 90% chance not 3rd cousin, 99% not 2nd cousin or closer. Only around 50% chance of having segment share if you were 4th cousins, about 15% if you were 5th cousins, and less than 5% chance if you were 6th cousins or more distant. Less than 50% does not imply "safe" to assume anything.

Many people can give you examples of documented relatives related in the 4th, 5th, 6th cousin range or beyond who do not have an autosomal segment share and would never assume they are not related within a genealogical time frame with that person.

[shandy4473]

Maybe you aren't hearing me

I suppose anyone is entitled to define for themselves what they consider to be a "safe assumption", but I would think most people with experience using both Y STR DNA and autosomal DNA for genealogy and the facts as presented (i.e. no other info to support the assumption) would probably disagree. Happy for others to comment on their perspectives.

I personally would only consider something a "safe assumption" if it were "highly likely (or unlikely)" to be the case (depending on the assumption), or at minimum "more likely than not". You simply don't have enough info to determine this, so what's safe about assuming?

The person could be a 4th, 5th, or 6th cousin which for most people would be within a genealogical time frame (abt 150-200 years). If you can add geography or non-intertwining "known" ancestry for both into the evaluation, perhaps you can make the case for your most likely not having common ancestor within genealogical time frame, but might still be hesitant with possibility of NPEs. In this case additional YDNA marker mutations at Y67 or Y111 might provide you with more information to make a "safe assumption".

In this case, the lack of the autosomal match only tells you that you most likely are not related as 3rd cousin or closer. 90% chance not 3rd cousin, 99% not 2nd cousin or closer. Only around 50% chance of having segment share if you were 4th cousins, about 15% if you were 5th cousins, and less than 5% chance if you were 6th cousins or more distant. Less than 50% does not imply "safe" to assume anything.

Many people can give you examples of documented relatives related in the 4th, 5th, 6th cousin range or beyond who do not have an autosomal segment share and would never assume they are not related within a genealogical time frame with that person.

Maybe you aren't hearing me. So I will repeat for you one last time. Just because the test doesn't show a match, doesn't mean they aren't related outside the window of a genealogical time frame. They could or couldn't be!!!!

I never said they couldn't be a 4th, 5th, or 6th cousin. The test for whatever reason, may not have detected segments. It's an assumption I took on which means it could turn out to be wrong.

And where in the textbooks or documentation is "90% chance not 3rd cousin, 99% not 2nd cousin or closer. Only around 50% chance of having segment share if you were 4th cousins, about 15% if you were 5th cousins, and less than 5% chance if you were 6th cousins or more distant. Less than 50% does not imply "safe" to assume anything."

Sources please!!!!

[shandy4473]

Closing Thread

I am going to close this thread and move on. This is disruptive to the group. And I have been asked to take the high road by several of my peers in cases such as this.


Thanks Dwight, Zarlor, and travers for your valuable input!!!


Steve