GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-02 > 1170601840


From: "Steven Bird" <>
Subject: [DNA] Analysis of Albanian data from YHRD/Pericic Kosovo sample andrequest
Date: Sun, 04 Feb 2007 10:10:40 -0500


The source of the YHRD data is:

Roewer, Lutz et al, "Signature of recent historical events in the European
Y-chromosomal STR haplotype distribution" publ. in Human Genetics (2005 Jan
20).

Is there anyone on the list that can shoot me a PDF of this article? :-)
It has a lot of information about Albania... and the rest of Europe too.
The allele data from this study is published by YHRD in their download
section, although not the original article.

I passed the 101 allele values taken in Albania from the above study through
Whit's Bayesian Haplogroup Predictor and came up with some VERY interesting
numbers. To summarize:

E3b - 40% (Cruciani, Pericic, and Semino found their E3b Albanian samples
to be 100% E3b1a2, without exception, so we can be safe in assuming that the
overwhelming majority of these are the same subclade, if not all of them.)

R1b - 25%
J2 - 16% (again, based on Pericic, probably J2e)
I1b1 - 7%
R1a - 5%
I1a - 2%
I1b2 - 2%
J1 - 2%
H - 2% (N.B. Athey's predictor says a 90+% possibility on both samples)

(Note: percentages add up to 101% due to rounding.)


Without having seen the Roewer paper, I can say that the variances on the
E3b and J2 haplotypes appear to be high, signifying an older population.
The 40% number for E3b1a2 is very close to the Pericic et al 2005 percentage
for Kosovo of 45.6%, based on a similar sized sample (N=114).

Kosovo was estimated to have been 90% Albanian at the time of the Pericic
study (2005). Recent news reports now place that figure at 98%, due to
Serbs leaving the region under pressure from the Albanian population. A
2005 "Ethnic composition" map of Kosovo (found at Wikipedia under the main
article on Kosovo) shows that the area around Pristina is dominated by the
Albanian populace. We can therefore take the Pericic sampling of Kosovo as
being representative of the (inflowing) Albanian population:

To compare the two sets of data:

Pericic (Kosovo)
R1b-M173R1a-M17I1b*-P37 E3b1-M78 J2e-M102 H
I1a
21.2%13.5%11.5% 42.3% 1.9% 3.8%
5.8%

Roewer (Albania)
R1b R1a I1b1-P37 and I1b2 E3b J2
H I1a
25% 5% 9% 40% 16%
2% 2%

I assume that some of these differences are due to the proximity of Kosovo
to certain parts of northern Albania and that refugees would tend to walk
only as far as they had to, i.e., near the border. Therefore, both the
Serbian and Albanian populations that have fled this region from time to
time would have moved just over the borders, out of harm's way. So perhaps,
for example, the drop-off of J2 may be a reflection of Kosovo's greater
distance from Greece. This trend continues towards Bosnia-Herzegovina (see
below).

The I1b* population peaks sharply just northwest of Albania and Montenegro,
along the sea coast in Herzegovina (according to Pericic) at 63.8% of the
population of that region. (No samples were reported from Montenegro). E3b
is still high there at 20.4% and J2e is represented at less than 1% (.7%)
This high presence of I1b* is taken by Pericic and Rootsi to perhaps
represent a very old population in the region, because of the high genetic
variance. Interestingly, the Albanians tend to see themselves as
descendants of the Illyrians, but the genetic evidence would seem to support
a much greater Thracian presence, in line with the rest of the Balkan
peninsula, but at a higher percentage. In a spirit of compromise, they are
sometimes also described as "Thraco-Illyrians," a blending of the two
"barbarian" tribal groups north of the Greeks.

Maybe the Herzegovinians are descended from the "Pelasgians?" ;-)

Anyway, to answer the original question, a possible Roma presence does not
seem to account for all of that E3b1a2 found in Albania or Kosovo,
especially since H is low in both samples.

Steve Bird



>From: "Lawrence Mayka" <>
>
>I suspect that the alleged Hg H is merely an artifact of the skimpy
>9-marker
>haplotypes. The two haplotypes that appear to be H (the ones with
>22-10-11-12-15,17) immediately flip over to J2 simply by adding DYS388=15
>or
>16. (Unfortunately, DYS388 is not one of YHRD's 9 STRs.) A J2 percentage
>of 5.7% makes much more sense.
>
> > [mailto:] On Behalf Of Steven Bird
> > R1b-M173R1a-M17I1b*-P37E3b1-M78
> > J2e-M102HI1a
> > 21.2%13.5%11.5%42.3%1.9%3.8%5.8%
>..
> > The H haplotype is typical of gypsy populations (usually
> > close to half the
> > overall male population) so it doesn't seem due to that
> > influence in this case.
>..

_________________________________________________________________
FREE online classifieds from Windows Live Expo buy and sell with people
you know
http://clk.atdmt.com/MSN/go/msnnkwex0010000001msn/direct/01/?href=http://expo.live.com?s_cid=Hotmail_tagline_12/06

This thread: