Archiver > GENEALOGY-DNA > 2010-12 > 1291920390

Subject: Re: [DNA] NW European R1b from Iberia?
Date: Thu, 9 Dec 2010 18:46:30 +0000 (UTC)
In-Reply-To: <>

>From: "Tim Janzen" <>
>I decided to look at the data this evening as well
from the same project at  Depending
on the markers chosen for analysis, there is some variation for each region.
In general, it appears that the highest variance for R-P312* is in Eastern
Europe if you include all three estimates and average them.  However, if you
only include the results from the 50 markers estimate then you see that the
highest variance is in France and England, as Robert noted in his message
earlier today.  It still seems clear that R-P312* moved into Spain and Italy
(as well as all of Southern Europe) relatively recently.  The TMRCA
estimates are as follows using a 30 year generation interval:
15 67-marker samples from Eastern Europe:
50 markers:  3544
10 YHRD markers using YHRD mutation rates: 5718
24 slow markers: 7597

36 67-marker samples from Spain and Portugal:
50 markers: 2992
10 YHRD markers using YHRD mutation rates: 4168
24 slow markers: 4649


Averages of the TMRCA for each of the regions above:
Eastern Europe: 5620
Spain and Portugal: 3936

The Bell Beaker Culture may well have been
responsible for the initial dissemination of R-P312 and R-U106.

My response:

Dear Tim,

Thank you for the data.  Since we employ quite different approaches, it does not make much sense to discuss here technicalities, such as why 30 years per generation (I do not use "generations" in my approach at all, I use time spans in years) and why to take averages from such different figures, some of which are clearly incorrect since differ by 200% and more.  Instead of venturing into an endless and generally fruitless discussion, I just share with you and folks here results of my quick analysis of the same series of haplotypes, and compare data on U106 and P312* in the same terms and under the same angle .

Let me remind you that for U106 an average number of mutations per marker in 67 marker haplotypes were:

0.221 for South-West Europe

0.234 for North-West Europe

0.234 for Scandinavia

0.236 for Central Europe

0.239 for North-East Europe

0.260 for South-East Europe.

All these figures are rather close to each other, and are withing the margins of error. However, if to take the average values listed above and forget about error margins, then this range corresponds to 3525 ybp (SW Europe) to 4225 ybp (SE Europe). As I have indicated, haplotypes in Iberia very likely went through a bottleneck between 4500 and 4000 ybp, hence, their lower TMRCA, even lower than those for "younger" subclades.

Now, for P312* haplotypes the respective values are:

0.248 for Eastern Europe

0.227 for Spain and Portugal.

As you see, it is in the same range and within the same margin of error. For one sigma the figures are 0.248+/-0.016 and 0.227+/-0.013, for two sigma it is 0.248+/-0.032 and 0.227+/-0.026, respectively.

What does it mean? It means that the "age" of U106 and P312 are practically the same, and that Eastern Europe does not have "older" P312 or U106 than any other regions in Europe.

These "variance" in mutations per marker translates to 4050 ybp for P312* in Eastern Europe, and 3650 ybp in Iberia. Technically, the difference is within margin of error. Or/and it reflects that population bottleneck I was talking about.

If you believe that "fast" markers and recLOH in DYS464 and YCAII could distort the data, here are data with removed these markers, for 61 marker haplotypes, in terms of average mutations per markers with respect to their base haplotypes:      

For U106:

0.175 for South-West Europe

0.203 for North-West Europe

0.208 for Scandinavia

0.197 for Central Europe

0.211 for North-East Europe

0.227 for South-East Europe.

For P312*:

0.225 for Eastern Europe

0.206 for Spain and Portugal.

As you see, now Iberia is even closer to the rest of the pack.


Anatole Klyosov

This thread: