Gerald de Windsor (Constable of Pembroke Castle) - Genetic Testing

That's amazing. I know nothing about genetic testing so does the Y chromosome test only go up through the male line? How do you know the genetic make up of Gerald? I'm quite fascinated, he's my 22nd great grandfather. Cheers, Kim Ellis

Kim:

In theory the Y chromosome is passed unchanged from father to son and so on. In fact, there are a predictable number of mutations every generation.

For example: If the average number of mutations is 6 per generation, My father's Y will have 6 differences from me and my brothers will have 12 differences from me. My grandfather's Y will have 12 differences from me and my first cousins will have 24 differences from me.

The testing service has a large data base of other men interested in genealogy. They found a man who had differences suggesting a common male ancestor about 25 to 30 generations back. He and I compared our paper trails and found a common ancestor 27 generations back, Gerald FitzWalter. We are descended from different sons of Gerald. This is definitive proof that I am descended from Gerald and pretty good proof that my paper trail from me to Gerald is correct. It is possible that one of the father's in the trail is actually a different brother.

There is a similar test for mitochondrial DNA which passes from mother to daughter.

Where do you fit into my descent from Gerald?

Jim

jim@jhglaw.net

There is a similar test for mitochondrial DNA which passes from mother to daughter.

I think the mutation rate is more like 1 per 5-6 generations, otherwise it wouldn't be a useful test at all.

Gerald is supposed to be my 25th great uncle, but it's via mixed descent (sons and daughters) - and it's another case of "Six Degrees of Dick Cheney".

Re: mutation rate.

The number of generations per mutation is only meaningful when put in a context of (a) number of markers tested and (b) assumed mutation rate. The current estimate is that there is probably one mutation in every generation on the entire y-chromosome. The question is also complicated by the fact that different markers seem to mutate at different rates.

If you're not afraid of a little math, here's a fairly simple formula to get a rough estimate:

(1 / (m*MU))
where mu is the assumed mutation rate and
M is the number of markers tested

For example, assume an overall mutation rate of .002. (Some people now think .003 or even .004 might be a better approximation.)

And assume 67 markers tested.

That gives (.002 x 67) = .134 probability of a mutation in each generation.

Then, (1 / .134) = 7.5 generations

So, assuming an overall mutation rate of .002 and 67 markers tested, there would be about one mutation every 7.5 generations. If you assume a higher mutation rate, say .003, it works out to one mutation every 4.9 generations. Using .004 works out to one mutation every 3.7 generations.

Mutations are random. They can happen at any time. This is just a rough guide to see if you're on the right track.

The Kerchner DNA project has good information on this topic, including a more complex (but still simplified) formula for calculating the probability of a mutation in each generation.
http://www.kerchner.com/dnamutationrates.htm