DNA-R1B1C7-L Archives

Archiver > DNA-R1B1C7 > 2010-09 > 1285149539


From: "Sandy Paterson" <>
Subject: [R-M222] Ewings Incognito
Date: Wed, 22 Sep 2010 10:58:59 +0100


There is an interesting group of 5 Ewings (groups 3a and 3b) who do not
present with the characteristic Ewing markers at DYS19,DYS439,DYS449 and so
on.

Who are they?

Well, 4 out of 5 of them have DO have the typical 37,38 at CDYa,b. "But
CDYa,b are fast mutators" comes the roar. OK, let's have a look at the M222
frequency of 37,38 amongst non-Ewings. In my databse, 10.23% of non-Ewings
are 37,38.

So what are the chances of 4 or more out of 5 non-Ewings drawn at random
having 37,38 at CDYa,b?

Using a binomial distribution with a mean of .1023 we get :

NProbCumulative

0.58298.58298
1.33218.91516
2.07571.99687
3.00863.99950
4.00049.99999
5.00001 1.00000

What this tells us is that we can reject the hypothesis that the 3a and 3b
Ewings are non-Ewings on the grounds that the chances of 4 or more out of 5
(if the underlying frequency is .1023) is only .00001

We therefore accept the alternative hypothesis, namely that they are indeed
Ewings, probably from an earlier branch than the bulk of the Ewings.

I've argued before that it is ill-advised to ignore fast-mutating markers
when considering a haplogroup as 'young' as M222. I think this example
underlines my point. And what's more, one of the 3a group matches Lamont
108794 at a gd of 3 over 37 markers. This Lamont matches the 37,38 at CDYa,b


Sandy








This thread: