Archiver > GENEALOGY-DNA > 2006-01 > 1136707134

From: ellen Levy <>
Subject: Re: [DNA] Scientific American and N1a
Date: Sat, 7 Jan 2006 23:58:54 -0800 (PST)
In-Reply-To: <006101c6140e$ad780720$b6559045@Ken1>


Nothing in my response was, nor was intended to be,

First of all, this isn't "my literature." These are
DNA studies. The one I was citing to by Dupanloup can
be found in Molecular Biology and Evolution, 2004.

Many people on this list don't have access to the
actual studies due to unavailability and purchase
costs. This is a DNA list and I was citing (and
quoting) directly from a DNA study. I'm sorry if you
find this objectionable.

I have no idea what you mean by "half of Europe," and
frankly, at this point, I'm not sure I really want to

You can address the studies or attack me. Given the
list's dislike for personal attacks, I'd recommend
addressing the studies.

I'm not here to summarize your DNA data or even
necessarily address it. This was a discussion (and
remained fairly pleasant up until now) about Haak's
recent DNA study, Demic diffusion models vs. cultural
Diffusion, Dupanloup's study, and Pinhasis's study. I
read your criticism of these studies being "weakly
substantiated," and asked for clarification regarding
which study you were perceiving as such and on what
grounds. Given your objection to my quoting from
Dupanloup's study, I have to assume you have read it
and are familiar with its analysis and conclusions.

I think the Dupanloup quote sums up why researchers
can end up with erroneous conclusions concerning
Neolithic contributions to the European gene pool by
relying only on Y-chromosome results. Apparently, you
need to look at beyond the Y-chromosome data,
examining autosomal and mtDNA results as well.
Dupanloup examined the following: 34 populations for
mtDNA HVRI polymorphisms, 42 populations for 11 NRY
binary markers, 27 populations for 22 NRY binary
markers, 59 populations for DQa and four
tetranucleotide microsatellites polymorphisms.
Admixture effects the entire genome, according to
Dupanloup, and thus the larger the loci studied, the
more accurate picture will be obtained.

More importantly, the question of demic diffusion
needs to be addressed on a regional basis. Even if
one restricts oneself to considering only Y-chromosome
as you seem to wish to do, there are strong regional
geographic patterns to the diffusion model. In
Greece, Albania and Italy, your have approximately 50%
of the population with either E3b or J results (sorry,
don't know the frequency of G). I believe Albania is
considered part of the Balkans.

Furthermore, a point that Dienekes made in a post
recently is that Neolithic peoples moving out of the
Middle East were not just restricted to E3b and J2.
For example, there is an ongoing thread about the
possibiity of haplogroup H being part of the Neolithic
migrations, of which there is a low frequency in
Greece, Italy and (apparently) Portugal.

Considering only Y-chromsome results as you suggest, a
good study is L. Chikhi's "Y Genetic Data Support the
Neolithic Demic Diffusion Model," Proc. Natl. Acad.
Sci., 2002. I won't quote extensively from it, given
that you have probably already read it. Chikhi found
the proportion of Neolithic genes to decrease from
85-100% in Albania, Macedonia, and Greece to 15-30% in
France, Germany or Catalunia. However, using linear
regression, Chikhi found an average of Neolithic
contribution of 50% across all their European samples,
with a high of 56% in Mediterranean samples and 44% in
non-Mediterranean samples. This finding was
confirmed, of course, by similar findings in
Dupanloup's study, which utilized far more loci for
their analysis.

Ellen Coffman

This thread: