Surprising Variation in Levels of DNA Match

[Wexflyer]

My father and an aunt both did DNA testing with Anc*stry.
There is no doubt that they are full siblings, with many thousands of shared matches.

But the variabillity in the level of matching from them to other family members continues to surprise me.  From nieces and nephews, to distant relatives.

To take just one example that I noticed today:

My aunt has a 111 cM match to someone I will call Z.
My father has no match at all to Z.

Seems to me that this should be outside the error bars, so to speak. And not the first example I have noticed.

[Wexflyer]

My father and an aunt both did DNA testing with Anc*stry.
There is no doubt that they are full siblings, with many thousands of shared matches.

But the variabillity in the level of matching from them to other family members continues to surprise me.  From nieces and nephews, to distant relatives.

To take just one example that I noticed today:

My aunt has a 111 cM match to someone I will call Z.
My father has no match at all to Z.

Seems to me that this should be outside the error bars, so to speak. And not the first example I have noticed.

I just spent several hours working out my connection to Z, which involved adding two generations to Z's family tree.  I now know our connection:
The MRCA from my father and aunt is g-g-grandparents
The MRCA from Z is g-g-g-g-grandparents

So the relationship to my father and aunt is is 3rd cousin, twice removed.

[TreeDigger]

The further back the MRCA is, the more distribution slash inheritance of DNA is a matter of the luck of the draw. I liken it to walking through a room in which your ancestors throw up hands of confetti while you walk through it; you never know which colored bit will stick 

For example..

The children from my 2x great-grandfather's official marriage inherited more of his DNA than the ones from his relationship to another woman (my 2x great-grandmother), which resulted i.e. in a 95cM match with a half third cousin v.s. two matches of 51cM & 42cM with two full third cousins.

An interesting fact which I personally think may be related is that my 2x great-grandfather was in his 30's and early 40's when fathering the kids from his marriage, and late 50's when fathering those with my 2x great-grandmother.

So even though the matches with the descendants of the half-siblings have less segments, the cM is higher than that with matches of the full siblings. Seeing how science has already determined that the quality of a man's sperm degrades the older he gets, it makes sense that this translates to the quality slash amount of DNA he passes on to his children as well.

To me, at least, this has become a factor when trying to determine which relationship a match descends from if there's no tree or one with not enough generations showing. And it's usually correct.

[Wexflyer]

So the relationship to my father and aunt is is 3rd cousin, twice removed.

The relationship predictor at this site
https://dna-sci.com/tools/segcm/

Says that a 111 cM match for a 3C2R relationship is only 3.8%. Interesting.

[Galium]

The children from my 2x great-grandfather's official marriage inherited more of his DNA than the ones from his relationship to another woman (my 2x great-grandmother), which resulted i.e. in a 95cM match with a half third cousin v.s. two matches of 51cM & 42cM with two full third cousins...


. Seeing how science has already determined that the quality of a man's sperm degrades the older he gets, it makes sense that this translates to the quality slash amount of DNA he passes on to his children as well.

That can't really be the case.  Everyone inherits half their DNA from thier mother, and half from their father.  Quality doesn't affect that.

So all children of your GGgrandfather will equally have had half their genetic inheritance from him.  The difference you are seeing is in the descendents of the children from each relationship, who will have had a more random selection of GGgrandfather's genes.

[LizzieL]

The relationship predictor at this site
https://dna-sci.com/tools/segcm/

Interesting that this website also breaks down the percentage probability for a particular relationship into which side of the family the match comes from. tried it out with a recent match that I'm still investigating and got quite a different percentage depending on whether it was mother's or father's side

[LizzieL]

So the relationship to my father and aunt is is 3rd cousin, twice removed.

The relationship predictor at this site
https://dna-sci.com/tools/segcm/

Says that a 111 cM match for a 3C2R relationship is only 3.8%. Interesting.

I've got a 3C1R  with a 159cM match over 8 segments, which according that website should only be 1.0%. Can't see any endogamy, each person in  that line married someone born a long way from where they were.

[Romilly]

DNA seems to be very randomly assigned…

Because of my longstanding Brickwall with my Paternal Grandfather, many of my family members have tested, including my late Mother, sister, daughter, 1st Cousins from both sides and many others. Shared DNA varies widely, as do levels with matches. It really doesn’t seem to conform to standardised tables!

Romilly

[Glen in Tinsel Kni]

Many of the sites represent shared dna as a percentage and display pie chart type diagrams, it's misleading as that 0.5%, 1%,5% or however much of shared dna  isn't just one single slice of pie, it's made up of multiple tiny slice, each so small that for two individuals to have the exact same slice is far more unlikely than it is likely.