Archiver > GENEALOGY-DNA > 2010-02 > 1266232513

From: Alan R <>
Subject: Re: [DNA] : variance of S116*
Date: Mon, 15 Feb 2010 11:15:13 +0000 (GMT)


Thanks for doing those calculations.  I kind of hoped you would when I posted!  I am surprised by the results to some degree and broadly speaking it would best support an east-west movement from eastern to west-central Europe before spreading elsewhere.  I had thought a route along the Med. from the east to Italy to France/Spain then the north through France or perhaps a dual Med. and Danubian route more likely. 

However, the 67 marker samples seem extremely small and the results depending on which method is chosen seem to vary in an odd way.  So, I have to say that my levels of confidence that the picture is not potentially being badly distorted is not high. .  Do the advantages of using only 67 markers outweigh the problem of the tiny sample representing most areas?

One other thing that springs to mind is having a similar look at the regional differences between the variance of U152 south of the Alps in Italy with that from north of the Alps in Switzerland/south Germany etc.   I suspect sample number may be a big issue again though. 



Tim said

Dear Alan,
    It is a good suggestion to look at R-S116/P312* variance data.  I
thus decided to do intraclade TMRCA estimates of the currently available
37-marker and 67-marker R-S116/P312* haplotypes from the FTDNA R-P312
project at 
to see if the current data supports any particular theory in regards to the
area of origin of R-S116/P312* and its spread through Europe.  You summed up
the various hypotheses nicely.  The current data shows that R-S116/P312* has
its highest variance in Eastern Europe.  The hypothesis best supported by
the data is that R-S116/P312* moved along a route through eastern Europe to
the west (possibly along the Danube as you suggest).  We also see that the
variance of R-S116/P312* decreases from east to west, suggesting that the
R-S116/P312 SNP occurred somewhere in Eastern Europe, Russia, or Ukraine.
It also seems clear that R-S116/P312* moved in Spain and Italy (as well as
all of Southern Europe) relatively recently.  The precise timing of the
R-S116/P312 SNP in my opinion is still uncertain, but my best guess is that
it occurred somewhere between 4000 and 7000 years ago.  Also see
73 for background.  Below are intraclade TMRCA estimates using various
marker sets for groups of haplotypes from various geographical regions as
well as subsets for selected countries or areas.  It should be kept in mind
that intraclade TMRCA estimates are generally lower than the true TMRCA by
some margin, possibly as much as 30% in some cases.
Tim Janzen

11 67-marker samples from Eastern Europe:
50 markers:    3552
10 YHRD markers using YHRD mutation rates: 5194
24 slow markers: 6553

14 37-marker samples from Eastern Europe:
25 markers:    3380
10 YHRD markers using YHRD mutation rates: 4995

23 67-marker samples from Western Europe:
50 markers:    3271
10 YHRD markers using YHRD mutation rates: 5356
24 slow markers: 3635

38 37-marker samples from Western Europe:
25 markers:    3420
10 YHRD markers using YHRD mutation rates: 4939

16 67-marker samples from Southern Europe:
50 markers:    2574
10 YHRD markers using YHRD mutation rates: 2980
24 slow markers: 4000

26 37-marker samples from Southern Europe:
25 markers:    3717
10 YHRD markers using YHRD mutation rates: 2749

13 67-marker samples from Scandinavia:
50 markers:    3185
10 YHRD markers using YHRD mutation rates: 2794
24 slow markers: 6317

15 37-marker samples from Scandinavia:
25 markers:    2801
10 YHRD markers using YHRD mutation rates: 3019

61 67-marker samples from the Great Britain and Ireland:
50 markers:    3373
10 YHRD markers using YHRD mutation rates: 3998
24 slow markers: 3408

76 37-marker samples from the Great Britain and Ireland:
25 markers:  3524
10 YHRD markers using YHRD mutation rates: 4115

Averages of the TMRCA for each of the regions above:
Eastern Europe: 4734
Western Europe: 4124
Southern Europe: 3204
Scandinavia: 3623
Great Britain and Ireland: 3683

Various subcategories:

5 67-marker samples from the Poland, Lithuania, Bohemia, and Prussia
(N17254, 73467, 97920, 104128, 109663):
50 markers:    2738
10 YHRD markers using YHRD mutation rates: 4068
24 slow markers: 7742

3 67-marker samples from Ukraine (150177, 1401, 153495):
50 markers:    3145
10 YHRD markers using YHRD mutation rates: 3229
24 slow markers: 4856

6 67-marker samples from France (132950, 127980, 46496, 114326, 49328,
50 markers:    3581
10 YHRD markers using YHRD mutation rates: 4793
24 slow markers: 3729

4 67-marker samples from Germany (N3432, 144510, 86995, E6422):
50 markers:    2654
10 YHRD markers using YHRD mutation rates: 4995
24 slow markers: 4097

4 67-marker samples from Spain (115590, 114718, 16848, 135663):
50 markers:    2171
10 YHRD markers using YHRD mutation rates: 2951
24 slow markers: 2927

4 67-marker samples from Italy (N62889, 30530, 141935, N47555):
50 markers:    1673
10 YHRD markers using YHRD mutation rates: 3292
24 slow markers: 1171

7 67-marker samples from Sweden (29135, 44772, 76755, 97998, 123539, N11543,
50 markers:    2647
10 YHRD markers using YHRD mutation rates: 2224
24 slow markers: 4587

-----Original Message-----
[mailto:] On Behalf Of Alan R
Sent: Saturday, February 13, 2010 3:37 AM
Subject: [DNA] Re (DNA): variance of S116*

In theory, most population movements have entered western Europe by either a
Danubian or a Mediterranean route or much more rarely by some route from the
east to the Baltic.  That being so, the original route (and of course the
origin point) should have higher variance than the areas it only spread to
in later secondary movements.  As S116* must by phylogenetic definition
include some lineages that are older than L21, U152 etc, it would be
interesting to know how the variance of this group varies on an area by area
basis in much the same way as L21 has been looked at.  If it travelled along
the Med then we would expect a decrease in variance along the Med. from east
to west (say from Italy to Iberia) and for all Med. S116* to have
significantly more variance than all central/north European S116*.  However,
if S116* moved along a Danubian route then we would expect central European
S116* to have significantly more variance than all of the Med.
and also for central European S116* variance to decrease from east to west.

So, I wonder if our more statsitically gifted list members would consider
looking into this? 


This thread: