GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2008-09 > 1221386547


From: "Dienekes Pontikos" <>
Subject: Re: [DNA] deCODEme results: Creating your own BGA Ancestry Test
Date: Sun, 14 Sep 2008 13:02:27 +0300
References: <ea3bd9560809132134w266b6bc6o59b3cfa2ad123af0@mail.gmail.com><455801.79726.qm@web28101.mail.ukl.yahoo.com>
In-Reply-To: <455801.79726.qm@web28101.mail.ukl.yahoo.com>


When the ancestral populations do not cover the observed variation,
then erratic results such as this will be observed.

In particular, decodeme uses HapMap samples to represent ancestral
populations. While the Yoruba and San are both "Africans", they aren't
the same kind of Africans. In fact there is great substructure within
Africa.

http://dienekes.blogspot.com/2008/07/campbell-tishkoff-review-paper-on.html

Thus, a test which determines an individual's "Africanness" by
comparing them with Yoruba is likely to produce erroneous results.

On Sun, Sep 14, 2008 at 11:34 AM, Anders Pålsen <> wrote:
> David, list
>
> I checked the San people reference individual, if there is any people beeing isolated it must be the San people of Africa, here is two examples of several in particular locations of the chromosomes:
>
> Chr 1 - Asian admixture 32%
> Chr 6 - European admixture 42%
>
> Obviously this cant be correct, but why? As I have shown with the X-haploblock the variability among the different part of the world is great even in such a small haploblock as 0.68% cM, maybe this is a clue to the problem. A practical example illsustrate the story.
>

--

Dienekes' Anthropology Blog
http://dienekes.blogspot.com


This thread: