DNA differences between two half cousins

[vegetablearian]

I don't think it indicates anything.

470 cM is bang on target for a half first cousin according to the chart here I always refer to.

I haven't found a lot of research on typical shared segments. The only one I've found is this page with a graph plotting segments shared between niece/nephew-aunt/uncle, grandparent-grandchild and half-sibling relationships. Those relationships all share the same amount of dna and the only way to try to tell them apart purely with dna is by number of segments. But as you can see and as the title mentions, there's a lot of overlap. I can't work out logically why grandparent-grandchild has fewer segments than half-siblings because to my mind it's still 2 steps (unlike niece/nephew-aunt/uncle which is 3), but it does.

Sometimes, twice as many segments would indicate an additional recombining, ie another generation. But I can't think of any kind of extra information that would change how these two half cousins are related to you. There are other potential relationships for that amount of dna but I don't know how they'd square with your family tree. If anyone doesn't have their genetic parents, the whole thing falls apart, so they pretty much must. I guess it's possible that the one with 20 segments is a half 1st cousin once removed who happens to be at the upper end of the range of shared dna but I absolutely wouldn't assume it based on this alone.

So I would assume that the average number of segments for half cousins is about 15, or at least includes 10 and 20, and that both results are fully normal. Perhaps someone will be able to assemble some stats on this like the shared cM project but I can see it being a lot of work!

[Mart 'n' Al]

We know the details of our birth and families, so we know I am a half cousin of each, although they are not related to each other. I just observed this minor oddity and wondered if there was a scientific reason.  It might just be coincidence. Insufficient data!

Martin

[hdw]

I've been trying to identify which segment on which chromosome contains the DNA of a particular surname in my tree but that's obviously impossible. The guy who is clearly the grandson of one of my paternal Watson uncles matches me with segments on 12 chromosomes, and on chromosome 3 there are 3 segments, the biggest one 82.4cM.

I have another Watson match, a woman whose late father was my 2nd cousin (we exchanged emails most Saturday afternoons), and it might seem likely that she would match me on chromosome 3 too, but we have no matching segments there at all although we have a large shared segment on chromosome 4.

Harry

[Mart 'n' Al]

Harry, perhaps we should discus this in detail, in private messages so we can give specific details?

Martin Watson