GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-05 > 1117355744


From: "Andrew and Inge" <>
Subject: RE: [DNA] Middle Eastern ancestral markers on new Euro 1.0 test
Date: Sun, 29 May 2005 10:35:44 +0200
In-Reply-To: <REME20050528143146@alum.mit.edu>


Hi John

> In principle, the test could look at the autocorrelation
function of the markers (if there were enough markers) to determine how
old the admixture was, but my guess is that 300 markers is not enough.


I guessed that this was your guess.

>It's not at all clear that the frequency contrast is declining over the
years. If that were necessarily true, then there would never be any
contrast in the first place. It all depends on things like natural
selective pressures (where relevant) and genetic drift. Most markers
do not have really strong selective value, and so the contrasts are
not BIG.

Not sure what point you are making. But in any case in an isolated region,
surnames and Y DNA haplotype will tend to die out over many generations "by
chance". That's my understanding anyway.

> If this is done as intended, then having a lot of markers which are more
> common in a region is very unlikely to be a chance remnant of something
that
> happened 1000 years ago let along 30,000 years ago.

> You're not taking in the big picture. We're not talking about a lone
ancestor who wandered away from home and made a tiny genetic
contribution elsewhere -- we're talking about a host of wanderers
either together or successive over millennia, as well as the slow
accumulation of small statistical flukes, all operating independently.

Actually I believe I mentioned this possibility - that there are pockets in
Europe which contain "undiluted" DNA from somewhere that might lead people
to believe that the people there all have immigrant grand parents so to
speak. Is this what you mean?

Best Regards
Andrew


This thread: