World Families Forums

General Forums - Note: You must Be Logged In to post. Anyone can browse. => R1b General => Topic started by: Heber on August 24, 2011, 12:42:29 PM



Title: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 24, 2011, 12:42:29 PM
Recently, the debate on the origins of the major European Y chromosome haplogroup R1b1b2-M269 has reignited, and opinion has moved away from Palaeolithic origins to the notion of a younger Neolithic spread of these chromosomes from the Near East. Here, we address this debate by investigating frequency patterns and diversity in the largest collection of R1b1b2-M269 chromosomes yet assembled. Our analysis reveals no geographical trends in diversity, in contradiction to expectation under the Neolithic hypothesis, and suggests an alternative explanation for the apparent cline in diversity recently described. We further investigate the young, STR-based time to the most recent common ancestor estimates proposed so far for R-M269-related lineages and find evidence for an appreciable effect of microsatellite choice on age estimates.As a consequence, the existing data and tools are insufficient to make credible estimates for the age of this haplogroup, and conclusions about the timing of its origin and dispersal should be viewed with a large degree of caution.

http://rspb.royalsocietypublishing.org/content/early/2011/08/18/rspb.2011.1044.full.pdf+html

Dienekers posted the following on his blog:
"The paper could just as easily have been titled "An epitaph for Y-STRs". Of course, Y-STRs do carry information related to antiquity; and there are so many datasets collected from both academics and genealogist enthusiasts. Thus, they will continue to be used and analyzed for at least a few years more.
Nonetheless, the conclusion is inescepable that a very specific use of Y-STRs on modern populations, with the goal of discovering tight links with archaeological/historical events is all but dead.
The reason is simple: as clocks, they suck. A bad clock is not useless: it gives you some information about time. Moreover, you can often use several to iron out the inaccuracy of any single one of them".

The study make interesting reading and reviews the data from Myers, Balaresque and others.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Jdean on August 24, 2011, 03:34:27 PM
.
The reason is simple: as clocks, they suck. A bad clock is not useless: it gives you some information about time. Moreover, you can often use several to iron out the inaccuracy of any single one of them".


I haven't read the article myself but got the impression from other posts about this that problem was the data didn't support the authors hypothesis, namely that M269's spread appeared to resent, and hence there conclusion that STR data was unreliable.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 24, 2011, 07:06:04 PM
For those interested in L21 the supplimentary bit includes sample positions.  It shows how essentially useless Myres was in France and also even in this study they simply didnt sample between sites in the western half of France and Germany.  So basically the east of France is missing and the contouring simply assumes its the same as Germany.  The eastern limit of L21 shaded contours is simply a dot joining exercise between the easternmost French samples.  That is the real reason why it seems there is a stranded area of higher L21 in SE France.  Its because its the only part of eastern France sampled (its the Myres sample).  So, again, I think this gives a slightly misleading impression.  I think L21 probably decreases in a contoured kind of way from NW France towards south Germany. 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Jdean on August 24, 2011, 07:21:12 PM
For those interested in L21 the supplimentary bit includes sample positions.  It shows how essentially useless Myres was in France and also even in this study they simply didnt sample between sites in the western half of France and Germany.  So basically the east of France is missing and the contouring simply assumes its the same as Germany.  The eastern limit of L21 shaded contours is simply a dot joining exercise between the easternmost French samples.  That is the real reason why it seems there is a stranded area of higher L21 in SE France.  Its because its the only part of eastern France sampled (its the Myres sample).  So, again, I think this gives a slightly misleading impression.  I think L21 probably decreases in a contoured kind of way from NW France towards south Germany. 

Yes unfortunately Myres didn't exactly go to town on France, however there does seem to be quite a hot spot in the SE corner and I seem to remember the R-L21 project has had trouble picking up people in the middle of France as well.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 24, 2011, 07:25:20 PM
I haven't read the article myself but got the impression from other posts about this that problem was the data didn't support the authors hypothesis, namely that M269's spread appeared to resent, and hence there conclusion that STR data was unreliable.

That is a large part of what the paper is about. From the text:

"While recent evidence has increased support for the Neolithic spread of R-M269, [Balaresque, Myres] we conclude that at the present time it is not possible to make any credible estimate of divergence time based on the sets of Y-STRs used in recent studies."

Bushby concludes:

"For now, we can offer no date as to the age of R-M269 or R-S127, but believe that our STR analyses suggest the recent age estimates of R-M269 [Balaresque] and R-S116 [Myres] are likely to be younger than the true values"

However, he doesn't dismiss the use of STRs. He has used them himself to come to the conclusion that the age estimates by Balresque and Myres are too young. Rather he highlights the choice of STRs as being the critical factor.

"While researchers take into account STR mutation rates when estimating divergence time with ASD, commonly used STRs do not have the specific attributes that allow linearity to be assumed further into the past. The majority of haplogroup dates based on such sets of STRs may therefore have been systematically underestimated."

And he offers a mild rebuke about their practices,

"Furthermore, we show that it is the properties of Y-STRs, not the number used per se, that appear to control the accuracy of divergence time estimates, attributes which are rarely, if ever, considered in practise."

Other than the technical aspects of his research, there is discussion about what this means for the arrival and dispersal of M269 in europe. In addition to considering a paleolithic date for this, he does acknowledge that it may be even younger than either Balaresque or Myres suggest.

"If the R-M269 lineage is more recent in origin than the Neolithic expansion, then its current distribution would have to be the result of major population movements occurring since that origin."

Bushby however doesn't think this is likely for, in his own words:

"For this haplogroup to be so ubiquitous, the population carrying R-S127 would have displaced most of the populations present in western Europe after the Neolithic agricultural transition."

This last part however is nothing to do with genetics. The genetics allow for this possibility. He thinks it is unlikely because europe is already populated.

However, the arrival of the Indo European languages, which is beyond this study, is a possible scenario which may account for a later arrival of M269. In addition, the population studies by Shennan and Edinborough, 'Prehistoric population history: from the Late Glacial to the Late Neolithic in Central and Northern Europe' (2007) and Shennan, 'Evolutionary Demography and the Population History of the European Early Neolithic' (2009) suggest that after the initial steep rise in population levels at the beginning of the neolithic, the population crashed early in the 5th millenium BC, in some places to extremely low levels, and remained low for nearly a millenium, before a second steep rise in population occured. So we do have a possible gap between the early and later neolithic populations. The population increase may be a result of the arrival of M269.

In addition, correct me if I am wrong, from the few samples of ancient yDNA that we have, none from the LBK have been found to be R1b or R1a. Those that are found appear to be later.

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 24, 2011, 07:28:34 PM
For those interested in L21 the supplimentary bit includes sample positions.  It shows how essentially useless Myres was in France and also even in this study they simply didnt sample between sites in the western half of France and Germany.  So basically the east of France is missing and the contouring simply assumes its the same as Germany.  The eastern limit of L21 shaded contours is simply a dot joining exercise between the easternmost French samples.  That is the real reason why it seems there is a stranded area of higher L21 in SE France.  Its because its the only part of eastern France sampled (its the Myres sample).  So, again, I think this gives a slightly misleading impression.  I think L21 probably decreases in a contoured kind of way from NW France towards south Germany. 

Yes unfortunately Myres didn't exactly go to town on France, however there does seem to be quite a hot spot in the SE corner and I seem to remember the R-L21 project has had trouble picking up people in the middle of France as well.

The real problem if I recall correctly was getting anyone to test in the middle of France.  I recall any sample was hard in large chunks of France and strangely the area where it was less difficult also happened to be the high L21 area.  However, I would consider that both Myres, the new study and the project have all not really sampled the landlocked part of France.  


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 24, 2011, 07:39:15 PM
I only just glanced at that paper, so forgive me if I am mistaken, but didn't they exclude the Anatolian haplotypes because most of them were L11-? That seems to me a mistake if that is in fact what they did.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Jdean on August 24, 2011, 08:11:37 PM
The real problem if I recall correctly was getting anyone to test in the middle of France.  I recall any sample was hard in large chunks of France and strangely the area where it was less difficult also happened to be the high L21 area.  However, I would consider that both Myres, the new study and the project have all not really sampled the landlocked part of France.  

Depopulation of rural France is a bit of a headache in that country, could this also be our problem?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: GoldenHind on August 25, 2011, 04:47:58 PM
Quote from: authun link=topic=10031.msg124118#msg124118 [i
"While recent
In addition, correct me if I am wrong, from the few samples of ancient yDNA that we have, none from the LBK have been found to be R1b or R1a. Those that are found appear to be later.

best
authun


That is correct.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 25, 2011, 06:14:36 PM
I only just glanced at that paper, so forgive me if I am mistaken, but didn't they exclude the Anatolian haplotypes because most of them were L11-? That seems to me a mistake if that is in fact what they did.

They explain the rationale behind excluding Balaresque's Anatolian samples in the supplementary information.

http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp1.pdf

A


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 25, 2011, 06:40:26 PM
I only just glanced at that paper, so forgive me if I am mistaken, but didn't they exclude the Anatolian haplotypes because most of them were L11-? That seems to me a mistake if that is in fact what they did.

They explain the rationale behind excluding Balaresque's Anatolian samples in the supplementary information.

http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp1.pdf

A

That seems like a mistake to me. The title of the paper indicates that the subject was R-M269. The Anatolian R-M269 samples were all M269+. They excluded them because most of them were S127- (L11-), but if the paper were addressing just R-L11/S127, then it should have indicated that was the case.

Obviously R-M269xS127 would be ancestral to R-S127. By excluding the Anatolian samples, Busby, et al, were rigging the game, at least in part, since every study I have ever seen has found Anatolian R-M269 diversity to be greater than European R-M269 diversity.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 25, 2011, 06:52:57 PM
I would like to comment on the frequency maps that appear in supplementary information section of the Busby, et al, paper. Notice the different shading scale applied to L21 (S145)? Why is that? It creates confusion and the illusion that, compared with S21 (U106) and S28 (U152), L21 is isolated in extreme NW Europe.

I noticed that, according to Busby, the northern tip of Jutland has about 9-16% L21.

Or am I seeing things?

Time to resurrect Faux's Cimbri theory!


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 25, 2011, 07:13:28 PM
Here's another comment on the frequency maps in the supplementary info section.

Notice that S145 (L21) and S28 (U152), both mere subclades of P312, are compared with ALL of S21 (U106)? Why aren't a couple of S21's subclades shown instead? Or why isn't ALL of P312 compared with S21?

Why do I get the feeling that L21 has gotten the shaft in these last couple of reports?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 12:13:16 AM
Here's another comment on the frequency maps in the supplementary info section.

Notice that S145 (L21) and S28 (U152), both mere subclades of P312, are compared with ALL of S21 (U106)? Why aren't a couple of S21's subclades shown instead? Or why isn't ALL of P312 compared with S21?

Why do I get the feeling that L21 has gotten the shaft in these last couple of reports?
Hey, I'm just happy that L21 specific data is provided... no longer just R-P312(S116*). This is a couple of papers in a row now.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 26, 2011, 03:19:58 AM
(Sent to Dienekes, but I don’t know if he will publish it)

The great mathematician and a chemist as well Anatole Klyosov insists, but his calculations are completely wrong because he doesn’t take in consideration that there are mutations around the modal and a convergence to the modal as time passes, except for those clades that fix mutations far from the modal itself.
His theory doesn’t explain that R-L51 arose in Italy (a few minutes ago Janzen on Rootsweb: “This distribution would favour an origin of R-L51 in Eastern or Southern Europe, possibly in Italy”, what I am saying from many years) and why the 3 most important subclades were born: R-U152 in Central-North Italy, R-L21 in the British Isles and R-U106 in North Europe.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 26, 2011, 04:36:45 AM
The real problem if I recall correctly was getting anyone to test in the middle of France.  I recall any sample was hard in large chunks of France and strangely the area where it was less difficult also happened to be the high L21 area.  However, I would consider that both Myres, the new study and the project have all not really sampled the landlocked part of France.  

Depopulation of rural France is a bit of a headache in that country, could this also be our problem?

I remember visiting beautiful deserted villages in the Massif Central in the 80s. I understand some are now populated by visitors from the Netherlands and UK. Many of the great cafes of Paris are owned and run by exiles from Auvergne.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 26, 2011, 04:50:28 AM
Maju has an interesting analysis of the paper.

http://forwhattheywereweare.blogspot.com/2011/08/r1b-m269-debate-new-paper-vindicates-my.html

"Another serious criticism they make about Balaresque is the use of an Y-search dataset representing Ireland (surprisingly amateurish!) When compared with actual samples (Y-search relies on the good will of online reporters) the low diversity that Balaresque found for Ireland vanished".

Fro Busby Notes:

"Testing the variance calculations from the Irish population Balaresque et al used haplotypes downloaded from the online Ysearch database (http://www.ysearch.org) which is a repository for genetic genealogists to upload and compare haplotypes (P. Balaresque pers. comm.). We note, however, that 17-STR haplotypes, including the 9 STRs used in Balaresque et al’s analysis, are available for 681 Irish R-M269 derived individuals in Moore et al (3), which is, in fact, the study which Balaresque et al use to estimate R-M269 frequency in Ireland. A subset of the Moore et al samples were re-analysed in the current study for SNPs downstream of R-M269, and the original haplotype data are used here to calculate variance. To test if the Ysearch haplotypes were representative of the Irish R-M269 in Moore et al, we independently re-sampled the Moore et al dataset 10,000 times, selecting sub-samples of 75 haplotypes from which we estimated the variance using the same 9 STRs used in the Balaresque et al paper. The median
variance of these 10,000 repetitions was 0.354 with a 95% CI of (0.285-0.432). The lowest variance value out of the 10,000 samples was 0.242, which is still higher than the figure observed in the Balaresque et al Ysearch sample (0.208). We therefore believe our estimate of Irish R-M269 variance to be a more robust representation of the true variance than that estimated by Balaresque et al."

The L21/S145 map is quite interesting and clearly shows an Atlantic Facade and Isles cluster for L21.

The BBC has weighed in on the debate with a very clear analysis of the paper.

http://www.bbc.co.uk/news/mobile/science-environment-14630012

"A new study deals a blow to the idea that most European men are descended from farmers who migrated from the Near East 5,000-10,000 years ago.
The findings challenge previous research showing that the genetic signature of the farmers displaced that of Europe's indigenous hunters.
The latest research leans towards the idea that most of Europe's males trace a line of descent to stone-age hunters.
But the authors say more work is needed to answer this question."


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 08:15:03 AM
Here's another comment on the frequency maps in the supplementary info section.

Notice that S145 (L21) and S28 (U152), both mere subclades of P312, are compared with ALL of S21 (U106)? Why aren't a couple of S21's subclades shown instead? Or why isn't ALL of P312 compared with S21?

Why do I get the feeling that L21 has gotten the shaft in these last couple of reports?
Hey, I'm just happy that L21 specific data is provided... no longer just R-P312(S116*). This is a couple of papers in a row now.

Not me. Sloppy, badly done, inaccurate attention is worse than being ignored, in my opinion.

Myres did a crappy job (on L21, anyway), and this paper simply extends and perpetuates that legacy.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 09:00:20 AM
. . .

The BBC has weighed in on the debate with a very clear analysis of the paper.

http://www.bbc.co.uk/news/mobile/science-environment-14630012

"A new study deals a blow to the idea that most European men are descended from farmers who migrated from the Near East 5,000-10,000 years ago.
The findings challenge previous research showing that the genetic signature of the farmers displaced that of Europe's indigenous hunters.
The latest research leans towards the idea that most of Europe's males trace a line of descent to stone-age hunters.
But the authors say more work is needed to answer this question."


Oh, brother! Not the "stone age hunters" bit again.

Busby, et al, conveniently discard the Anatolian M269 haplotypes, and nobody notices?



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 09:06:28 AM
Here's another comment on the frequency maps in the supplementary info section.

Notice that S145 (L21) and S28 (U152), both mere subclades of P312, are compared with ALL of S21 (U106)? Why aren't a couple of S21's subclades shown instead? Or why isn't ALL of P312 compared with S21?

Why do I get the feeling that L21 has gotten the shaft in these last couple of reports?
Hey, I'm just happy that L21 specific data is provided... no longer just R-P312(S116*). This is a couple of papers in a row now.

Not me. Sloppy, badly done, inaccurate attention is worse than being ignored, in my opinion.

Myres did a crappy job (on L21, anyway), and this paper simply extends and perpetuates that legacy.

One thing I will say about the "S145" map, if it is based on accurate information, I think it tends to show that L21 is by far the single most frequent y haplogroup in the British Isles, including England. Just compare it to the maps of "S21" and "S28".

Just the same, I am aggravated at the shading scale used for the L21 map. It helps us to see the L21 clines in the British Isles but practically obliterates the continental ones. I think all three maps should have had the same shading scales for ease of comparison.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 09:57:26 AM
Sorry to post so many unanswered messages in a row, but, unless I missed something in reading back through Busby, it looks like they really just worked on R-S127 and not on R-M269. Am I wrong?

That seems a major shortcoming of the paper to me. S127 is pretty plainly descended from R-M269xS127. The maps in Busby of the distributions of R-M269 as a whole, then of R-M269xS127 and R-S127 show the progression from east to west. Leaving out R-M269xS127 from the calculations merely shows that S127 expanded rapidly once it hit central and western Europe. Had R-M269xS127 haplotypes been included, the increased STR diversity everybody sees in the East, declining as one moves west and north, would have been evident once again.

Everyone is missing this point and acting as if Busby is a major blow to the idea that M269 is fairly recent and arrived in Europe from the East.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 26, 2011, 10:05:21 AM
Just the same, I am aggravated at the shading scale used for the L21 map. It helps us to see the L21 clines in the British Isles but practically obliterates the continental ones. I think all three maps should have had the same shading scales for ease of comparison.

The maps cannot have the same shade, because the highest percentage of L21 is 79% and U152 has 43% and U106 a little lower. Anyway these maps show clearly where we have now the highest diffusion, which doesn't mean automatically the origin, but probably it isn't so far. Like I have said here and in other forums:
Central-North Italy for R-U152
The British Isles for R-L21
Central-North Europe for R-U106.

If we add that it is clear for everybody now that R-L51 was born in Italy (like many upstream haplogroups), probably it shouldn't be difficult to reconstruct the whole history.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 10:14:57 AM
Sorry to post so many unanswered messages in a row, but, unless I missed something in reading back through Busby, it looks like they really just worked on R-S127 and not on R-M269. Am I wrong?

That seems a major shortcoming of the paper to me. S127 is pretty plainly descended from R-M269xS127. The maps in Busby of the distributions of R-M269 as a whole, then of R-M269xS127 and R-S127 show the progression from east to west. Leaving out R-M269xS127 from the calculations merely shows that S127 expanded rapidly once it hit central and western Europe. Had R-M269xS127 haplotypes been included, the increased STR diversity everybody sees in the East, declining as one moves west and north, would have been evident once again.

Everyone is missing this point and acting as if Busby is a major blow to the idea that M269 is fairly recent and arrived in Europe from the East.


Here's another thing . . . ahem . . .

If Busby resurrects the old "stone age hunter M269" malarkey, as some seem to think, why didn't they find evidence of STR differentiation out of the old Cantabrian Ice Age Refuge?

I mean, wow, if M269 was in the Cantabrian Refuge (or in Italy . . . cough) back during the last Ice Age, one would think that was plenty of time for STR mutations to tell the story.

But the Busby paper doesn't announce any such startling revelation. Instead, it leaves out the R-M269xS127 haplotypes, effectively rigging the game, and works on just the central and western European S127 (L11) stuff. Not only that, they failed to look at the widespread distribution of P312 (S116), instead choosing two of its subclades (L21 and U152) and concluding that their distributions are "markedly localized".


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 10:20:30 AM
. . .

The BBC has weighed in on the debate with a very clear analysis of the paper.

http://www.bbc.co.uk/news/mobile/science-environment-14630012

"A new study deals a blow to the idea that most European men are descended from farmers who migrated from the Near East 5,000-10,000 years ago.
The findings challenge previous research showing that the genetic signature of the farmers displaced that of Europe's indigenous hunters.
The latest research leans towards the idea that most of Europe's males trace a line of descent to stone-age hunters.
But the authors say more work is needed to answer this question."


Oh, brother! Not the "stone age hunters" bit again.

Busby, et al, conveniently discard the Anatolian M269 haplotypes, and nobody notices?

I don't see where Busby indicates an Europen Paleothic or Mesolithic origin for R-M269. I think this is just the BBC media spinning things a bit to liven up an old debate.

I noticed both the adjustments that Busby did to the Irish data, the removal of Anatolian data, and the lack of Near East data.

Even though they broke out subclades in some of their reporting, in the real analysis of diversity they lumped some things together.

If you read Supplement 2 you can see they go through various "re-analysis" stages. They start out showing and east to west client where diversity is highest in the east but after a couple of turns to the crank that got it slightly tilted the other way.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 10:21:59 AM
Just the same, I am aggravated at the shading scale used for the L21 map. It helps us to see the L21 clines in the British Isles but practically obliterates the continental ones. I think all three maps should have had the same shading scales for ease of comparison.

The maps cannot have the same shade, because the highest percentage of L21 is 79% and U152 has 43% and U106 a little lower. Anyway these maps show clearly where we have now the highest diffusion, which doesn't mean automatically the origin, but probably it isn't so far. Like I have said here and in other forums:
Central-North Italy for R-U152
The British Isles for R-L21
Central-North Europe for R-U106.

If we add that it is clear for everybody now that R-L51 was born in Italy (like many upstream haplogroups), probably it shouldn't be difficult to reconstruct the whole history.

They could have the same shading scale, but they didn't.

L21 did not originate in the British Isles, just as R-M269 did not originate in Spain, even though Spain has a much higher frequency of M269 than western Asia.

I am surprised at you. It is very much a beginner's error to think that a y haplogroup must have originated in the place where it is most frequent today.

I know of no one but you, Gioiello, who thinks M269 spent the Last Ice Age in Italy. You keep on insisting on it and producing what amounts to individual anecdotes to support it, but I haven't seen any real evidence for it.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 10:24:11 AM
. . .

The BBC has weighed in on the debate with a very clear analysis of the paper.

http://www.bbc.co.uk/news/mobile/science-environment-14630012

"A new study deals a blow to the idea that most European men are descended from farmers who migrated from the Near East 5,000-10,000 years ago.
The findings challenge previous research showing that the genetic signature of the farmers displaced that of Europe's indigenous hunters.
The latest research leans towards the idea that most of Europe's males trace a line of descent to stone-age hunters.
But the authors say more work is needed to answer this question."


Oh, brother! Not the "stone age hunters" bit again.

Busby, et al, conveniently discard the Anatolian M269 haplotypes, and nobody notices?

I don't see where Busby indicates an Europen Paleothic or Mesolithic origin for R-M269 . . .


If you look at what I quoted, you will see I was responding to Heber's posting of an article from the BBC which mentioned the "stone-age hunters" thing.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: IALEM on August 26, 2011, 10:46:55 AM
As I read it, the logic for Busby et alii to let out the Turkish M269 (xS127) is because M269 in Europe is overwhelmingly S-127, so it can´t be directly derived from the Turkish xS127 population. the logic is.
S-127 is dominant in Western Europe
S-127 alone (excluding xS127) shows no higher diversity in Turkey than in Western Europe
Then, S-127 did not originate in Turkey, so it wasn´t direclty related to a Neolithic expansion.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 10:47:12 AM
.... Everyone is missing this point and acting as if Busby is a major blow to the idea that M269 is fairly recent and arrived in Europe from the East.
This is kind of like the "Revenge of the Nerds" . (No offense, nerds are smart!) They've been building up frustration and now have a counter-argument to wield. I think I have this right, but one blogger said "STRs suck". LOL.  Oh, well.

I think this is an opportunity for growth. I have seen what Tim Janzen and Dienekes have seen and Busby tried to demonstrate. STR variance is not a pure straight line correlation with time/generations and some STR's may misbehave. This is no reason to throw the baby out with the bath water, though.  

On another forum, a sharp fellow who goes by Myth69 has tried to calculate the length of time that individual STR's show linear variance growth. This is a step in the right direction. Ironically, Myth69's analysis shows that three 3 of the 10 STRs that Busby used across the board in his diversity calculations are not useful for more than about 3000 years and a 4th about 5000 years. The Neolith advances were about 7000 years ago so 4 of Busby's 10 markers should have been thrown out for his diversity calculations, by his own argument. Apparently, Busby's analysis of these 4 STRs was different than Myth69's.

Here is where the growth comes in. Folks like Ken Nordtvedt have maintained that these STR non-linearity issues are insignificant and report that as a result of their simulations. I think this is true in some circumstances, probably most, but I have to agree with Janzen and Dienekes that sometimes the issues make a difference. It would take someone like Ken, but hopefully, we'll see a new round of TMRCA estimators that try to account for STR mutation / time linearity issues beyond just throwing the multi-copy markers out.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 10:50:22 AM
As I read it, the logic for Busby et alii to let out the Turkish M269 (xS127) is because M269 in Europe is overwhelmingly S-127, so it can´t be directly derived from the Turkish xS127 population. the logic is.
S-127 is dominant in Western Europe
S-127 alone (excluding xS127) shows no higher diversity in Turkey than in Western Europe
Then, S-127 did not originate in Turkey, so it wasn´t direclty related to a Neolithic expansion.
A bit of circular reasoning! eh?

This is what "re-analysis" can do for you.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 10:56:04 AM
...  That seems like a mistake to me. The title of the paper indicates that the subject was R-M269. The Anatolian R-M269 samples were all M269+. They excluded them because most of them were S127- (L11-), but if the paper were addressing just R-L11/S127, then it should have indicated that was the case.

Obviously R-M269xS127 would be ancestral to R-S127. By excluding the Anatolian samples, Busby, et al, were rigging the game, at least in part, since every study I have ever seen has found Anatolian R-M269 diversity to be greater than European R-M269 diversity.
I have to say I have the same disappointment. This reminds me a bit of political argument. Throw around enough stats about this or that: 1) 15 STRs when you only use 10 across the board, or 2) or subclade maps but the key diversity cacluations lump and divides a bit haphazardly ... and through the convolution claim victory.

I also see the classic strawman logical fallacy of extension. Pick out an opponent's argument where there are issues that are not fully resolved and band on that argument (STR variance in this case) and ignore the other pro-arguments that may not fit your counter. Also don't provide an alternative (oops, I almost said "budget") at all. LOL.

I guess, I shouldn't be disappointed. Why would academics be immune from politics?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: IALEM on August 26, 2011, 11:17:49 AM
As I read it, the logic for Busby et alii to let out the Turkish M269 (xS127) is because M269 in Europe is overwhelmingly S-127, so it can´t be directly derived from the Turkish xS127 population. the logic is.
S-127 is dominant in Western Europe
S-127 alone (excluding xS127) shows no higher diversity in Turkey than in Western Europe
Then, S-127 did not originate in Turkey, so it wasn´t direclty related to a Neolithic expansion.
A bit of circular reasoning! eh?

This is what "re-analysis" can do for you.
Yes, however they go out of the circle in their conclusions by stating that
1) S-127 originated in Europe before the expansion of Neolithic
2) Then, their subhaplogroups are linked to the Neolithic expansion within Europe At the subhaplogroup
level, then, R-M269 is split into geographically
localized pockets with individual R-M269 subhaplogroups
dominating, suggesting that the frequency
of R-M269 across Europe could be related to the
growth of multiple, geographically specific sub-lineages
that differ in different parts of Europe.
A recent analysis of radiocarbon dates of Neolithic sites
across Europe [46] reveals that the spread of the Neolithic
was by no means constant, and that several ‘centres of
renewed expansion’ are visible across Europe, representing
areas of colonization

So, Busby et alii scenario would be
1) M-269 arrives into Europe before Neolithic
2) S_127 is born from this pre-Neolithic population in Europe
3) THe expansion of S-127 subhaplogroups is linked to the Neolithic second wave.
4) So, by default, the first wave of Neolithic population in Europe was no R-M269 related, and din´t make a significant contribution to the present population of Western Europe


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 11:58:48 AM
If S127 (L11) was born in Europe, he was the offspring of an M269xS127 father, so leaving out M269xS127 haplotypes is clearly a mistake, since the point, it seems to me, is to track the origin of European S127 (although the paper's deceptive title mentions the "cautionary tale of . . . M269", not S127). It's a little like knowing who someone's father is and where he was born and then attempting to trace the family tree without including him.

Personally, I think it quite likely that the expansion of M269 and its descendant S127 (L11 - Man, I hate having to switch to these "S" equivalents!) is connected to the spread of the Indo-Europeans rather than to the Neolithic farmers, if the two are separate.

I suspect no modern scientist is going to grab the M269-is-Indo-European ball and run with it. That would be too big a political hot potato, given Europe's recent history and the current dominance of political correctness.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 26, 2011, 12:07:44 PM
I know of no one but you, Gioiello, who thinks M269 spent the Last Ice Age in Italy. You keep on insisting on it and producing what amounts to individual anecdotes to support it, but I haven't seen any real evidence for it.
I thank IALEM for having said by words that everybody can understand what is the real meaning of this paper.
To Rich I would say that what the paper says now is what I have said in other words, simple like those of IALEM (if my bad English aids me), in these last 5 years. He knows that I don’t consider him like those who banned me from Rootsweb and “forums-dna” and if I am here that is due to him. But now I say to him that, if he hasn’t understood the paper, he can go an read my thousands of letters of these last years.
1)   I haven’t said that R1b1b2 is Palaeolithic. I have linked it to the Italian Younger Dryas, which is a little before the Mesolithic.
2)   R1b1b2 is predominant in West Asia but probably was also in Europe and in Italy in the same times.
3)   It isn’t said that they are the same. They could be different, and now we are finding some SNPs of the Eastern R1b1b2 and it isn’t said that Eastern R1b1b2 hasn’t come from Europe.
4)   What I have said is that the most ancient R1b1b2 was born in Italy and is the Italian YCAII=17-23, that R-L51 was born in Italy (now also Janzen thinks that) and that the SNP L150, present ancestral only in the Italian Romitti, was a sign of this. Then Italy has the pathway from R-M207 (2 Tuscans out 50 in 1000 Genomes Project, the other 3 Puertoricans but no one amongst Spaniards) to at least R-L51. After Italians went out but also 1 of the 3 most diffused subclades of R1b1b2, R-U152, was born here.
5)   The expansion happened with agriculture, demic to South Balkans and perhaps South Italy, but cultural from previous hunter-gatherers and foragers. The Grotta delle Arene Candide in Liguria was one of the point from where the seafarers diffused the agriculture to West, South France and Iberia.

I have said much more but for now it is enough. Lastly I would say that the only Italian scholars who replied to my letters are here: Cristian Capelli and Gianmarco Ferri. But it is from Fulvio Cruciani that I am waiting for the definite proofs of all that.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: IALEM on August 26, 2011, 12:19:15 PM
The point of the article is the "Peopling of Europe", so they are leaving out those M-269(xS127) haplotypes because they considere them irrelevant and misleading to that aim.
They consider the possibility of a post Neolithic arrival of M269 and the subsequent explosion of S-127 derived haplogroups in Europe, but they find it unlikely because they would have found the descendants of the Neolthic population expansion.
In that regard it is odd they cite Bocquet and de Miguel study on Neolithic demography. Mari Paz de Miguel (I know her personally) has stressed the 2 cicles in Neolithic population expansion, because it could fit equally a postNeolithic arrival with second wave of Neolithic population expansion


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 12:19:52 PM
I don't think this current paper says much of anything that is of much use. Its authors' decision to exclude the M269xS127 haplotypes renders it impotent, in my opinion. I realize I am no scientist, but it is not hard to see that M269xS127 is ancestral to S127.

L150 and L51 are so new that I don't think any major studies have included them yet. Gioiello's mention of them is precisely what I meant about individual anecdotes rather than real evidence. This individual Italian or that one who is R-L51* is not proof of anything, any more than finding a y-haplogroup A Englishman proves the Garden of Eden was somewhere near Manchester.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 26, 2011, 12:35:17 PM
You speak of descent, and I think having demonstrated that from an Italian R1b1b2 was born an R1b1b2a (like Romitti), an R-L150+ (like me), an R-L51, overwhelming in italy, and so on. This till R-S127 which generated R-U106 and R-P312. And this is an European history, above all Italian I think at least in its beginning but also in one the most diffused R-subclades like R-U152. What more?

I have written in this blog also about R-M335, R1b1b1 and others. Also on R1a (M420), overwhelming European and Italian, disappointing a Nationalist Indian who kept silent after.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 26, 2011, 01:52:52 PM
Note how shy are these Italian scholars who speak to the Lords of the World: “R-S21 has a frequency of 44 per cent in Friesland, and R-S28 reaches 25 per cent in the Alps” (page 7).

Italy has the 5 highest lines of frequency (from 23 to 43 per cent) that no other country has and the top, from 40 to 43 per cent, is reached in Piedmont and Central Italy.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: MHammers on August 26, 2011, 03:03:28 PM
I haven't read the whole study, but from what I can gather from the discussion is..

1) The subclade selection is spotty and not as inclusive as what is in the amateur projects.
2) STR variance does not accumulate in a linear fashion and is therefore unreliable.

Number 1 would seem to make number 2 a questionable conclusion, imo.  Also, does this change the fact that numerous observances have the
R1b snp phylogeny running east to west.  The str variance is more or less showing the same pattern by subclade as well.  Number 2 may very well be accurate, but I think the most recent observable pattern is east to west thus reflecting something like the Neolithic or Bronze age.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 26, 2011, 03:36:56 PM
So, Busby et alii scenario would be
1) M-269 arrives into Europe before Neolithic
2) S_127 is born from this pre-Neolithic population in Europe
3) THe expansion of S-127 subhaplogroups is linked to the Neolithic second wave.
4) So, by default, the first wave of Neolithic population in Europe was no R-M269 related, and din´t make a significant contribution to the present population of Western Europe

Not quite, Busby acknowledges that it may be more recent than the start of the neolithic:

"If the R-M269 lineage is more recent in origin than the Neolithic expansion, then its current distribution would have to be the result of major population movements occurring since that origin."

He doesn't think it is likely as, if this was the case, its dominance in europe could only be explained by a replacement of the early neolithic farmers. However, he doesn't appear to be aware of the population crash at the end of the LBK reported by Shennan and Edinborough. The crash which affected several parts of northern europe reduced the populations to their low mesolithic levels and remained at these low levels for nearly 1000 years.

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: GoldenHind on August 26, 2011, 03:54:45 PM

I noticed that, according to Busby, the northern tip of Jutland has about 9-16% L21.

Or am I seeing things?

Time to resurrect Faux's Cimbri theory!

I think one of the most interesting things in the Myres' data for Denmark is that all R1b subclades for which they tested: U106(XU198), U198, L21, U152 and P312* have their highest frequencies in northern Denmark- presumably northern Jutland. I can't believe this is coincidental, but I don't know what conclusion to draw from it. I doubt very much the resurrection of Faux's Cimbri theory is the solution. However there might be some connection with the map recently mentioned by Authun which shows the heaviest concentration of Bronze Age sword (Griffzungen) finds in northwest Jutland.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: GoldenHind on August 26, 2011, 04:01:26 PM
Here's another comment on the frequency maps in the supplementary info section.

Notice that S145 (L21) and S28 (U152), both mere subclades of P312, are compared with ALL of S21 (U106)? Why aren't a couple of S21's subclades shown instead? Or why isn't ALL of P312 compared with S21?

Why do I get the feeling that L21 has gotten the shaft in these last couple of reports?

It is a constant irritation to me that every study of R1b in Europe compares all of U106 (occasionally with the exception of U198) with individual P312 subclades such as L21, U152, etc., which is obviously comparing apples with oranges. How informative would it be to compare all of P312 with U198, L48, L271 etc.?

Although I know I will never live long enough to see peace in the Middle East, I am now beginning to doubt that I will ever see in my lifetime a reliable academic study on R1b in Europe.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Nolan Admin - Glenn Allen Nolen on August 26, 2011, 04:56:26 PM
rms2

"This individual Italian or that one who is R-L51* is not proof of anything, any more than finding a y-haplogroup A Englishman proves the Garden of Eden was somewhere near Manchester."


http://freepages.genealogy.rootsweb.ancestry.com/~nolenancestry/nolan_DNA_contact_info.html

* Garden of Eden setting equals (Central Asia) Almaty, Kazakhstan: The Fatherland of Apples

"An earlier traveler through the region, one Victor Vitkovich, proclaimed these naturally occurring groves to be “a marvelous garden where apples and pears look down on you from the trees and beg to be eaten.”


http://s1230.photobucket.com/albums/ee494/glennnolen/Tien%20Shan%20with%20Lake%20Issyk-kul/

Gates or Mountain Passes to the Garden of Eden: Tien Shan with Lake Issyk-Kul

Glenn Allen Nolen


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Bren123 on August 26, 2011, 05:31:30 PM
. . .

The BBC has weighed in on the debate with a very clear analysis of the paper.

http://www.bbc.co.uk/news/mobile/science-environment-14630012

"A new study deals a blow to the idea that most European men are descended from farmers who migrated from the Near East 5,000-10,000 years ago.
The findings challenge previous research showing that the genetic signature of the farmers displaced that of Europe's indigenous hunters.
The latest research leans towards the idea that most of Europe's males trace a line of descent to stone-age hunters.
But the authors say more work is needed to answer this question."


Oh, brother! Not the "stone age hunters" bit again.

Busby, et al, conveniently discard the Anatolian M269 haplotypes, and nobody notices?



Here's a quote from the BBC article:

Furthermore, they suggest that some of the markers on the Y chromosome are less reliable than others for estimating the ages of genetic lineages. On these grounds, they argue that current analytical tools are unsuitable for dating the expansion of R-M269.

If they're saying that the markers( or some) are unreliaable,how can they then decide that most European men have ancestry dating back to hunter gatherers? Another question, have  they found any  aR1b in European hunter gatherer populations?

Another quote from the BBC article:

Co-author Dr Jim Wilson from the University of Edinburgh explained: "Estimating a date at which an ancestral lineage originated is an interesting application of genetics, but unfortunately it is beset with difficulties."

The more I look at this, the more I'm convinced it's pseudo-science!


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 26, 2011, 06:04:45 PM
I think one of the most interesting things in the Myres' data for Denmark is that all R1b subclades for which they tested: U106(XU198), U198, L21, U152 and P312* have their highest frequencies in northern Denmark- presumably northern Jutland.

The sampling locations are given in the map Fig S1 in the supplementary information,
http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp1.pdf

The supplementary data is given in Excel format at:
http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp2.xls

The Hg frequency for northern Denmark is 9.5%, as opposed to 4.1% and 5.3% for Southeast Denmark and West Denmark resp. Also, the Norwegian sampling location is 8%, nearly twice that of southern Jutland.

cheers
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 26, 2011, 06:38:26 PM

Here's a quote from the BBC article:

Furthermore, they suggest that some of the markers on the Y chromosome are less reliable than others for estimating the ages of genetic lineages. On these grounds, they argue that current analytical tools are unsuitable for dating the expansion of R-M269.
I've been saying let's not throw the baby out with the bath water. First, I'm not convinced that Busby et al's perspective on STR variance correlation with time is as dire as they say. Almost everybody from Myres, Balaresque, Wells/National Genographic to FTDNA uses STR variance. Second, I disagree with Busby's implication that "more" data is not better. That defies the law of large numbers in statistics.

I agree using junk data is not useful. I agree STR variance correlation with time is a legitimate issue and more work needs to be done. Busby seemed to be taking a step by evaluating the period of time individual STRs are correlated linearly with time. Supposedly, they used their approach to pick 10 STRs to evaluate. I think they, like most, are constrained by the available data.

Other people are also addressing this. There is a fellow who goes by Mth69 on another forum who seems to be pursuing this in a reasonable way. He posted his analysis on how long individual STRs are useful.

Quote from: Mth69
Based on the above mentioned, one can roughly estimate the timeframe for each locus where saturation effects are relatively insignificant. The idea is that up/down ratio will not change very much due to repeat number changes that can occur. Variance formulas do not care about specific ratio of up and down mutations, but require that the ratio is approximately constant.
Here is the link where he posted his results:  http://beforepresent.dyndns.info/misc/loci.xls

I added this comment
Quote from: Mikewww
As an aside, I find it a bit ironic that the recent Busby study on R-M269 cites the problems with STR variance and a linear association with time, but then they use the following on their STR list that they used across the board.

389i, 391, 439 which would be applicable no more than 3300 years. 389b which is applicable less than 5000 years. Four of Busby's ten may not be applicable for studying a period as long ago as the initial Neolithic advances.

Mth69's analysis is not published so not worth for formal argument purposes. Anectdotally, I definitely think 439 is a poor choice for Busby's inclusion.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 06:49:52 PM
Interesting that the sample from SE France was nearly 20% L21+ (S145+), while U152 (S28) was about 13%. P312xL21,U152 was 29% there, but that would be divided up among the subclades not tested for this paper and would probably not all be P312*.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 26, 2011, 07:24:58 PM
Interesting that the sample from SE France was nearly 20% L21+ (S145+), while U152 (S28) was about 13%. P312xL21,U152 was 29% there, but that would be divided up among the subclades not tested for this paper and would probably not all be P312*.

Here are some interesting facts from the Busby paper relative to L21, facts that are really hardly if at all discernible from the L21 (S145) map in the supplementary info section.

Rennes in Bretagne had a frequency of 40% L21xM222 (not surprising).

Poitiers had 14%.

Marseille was 11% L21xM222.

Lille, up in the northeast near the Belgian border, had 10% L21xM222 (that one did surprise me - it is higher than I thought it would be).

Paris had 10% L21xM222.

Chalon-sur-Saône, in the east, near Dijon, had 8% L21xM222.

These frequencies are percentages of the total y-dna. That means L21 is actually huge all over France. I mean, 8% of the total y-dna in an area is pretty significant, let alone 10% and up.

If the shading scale on Busby's L21 map had been like those of its U106 and U152 maps, the true L21 picture in France would have been more readily apparent.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanL on August 26, 2011, 09:54:13 PM
Sorry to post so many unanswered messages in a row, but, unless I missed something in reading back through Busby, it looks like they really just worked on R-S127 and not on R-M269. Am I wrong?

I’m sorry to break it to you, but I think that you are in fact wrong. Busby et al(2011) did do an analysis of the STR variance of R1b-M269xS127, it is shown in Figure-2b)

(http://rspb.royalsocietypublishing.org/content/early/2011/08/18/rspb.2011.1044/F2.large.jpg)

http://rspb.royalsocietypublishing.org/content/early/2011/08/18/rspb.2011.1044/F2.large.jpg

They did not exclude the Anatolian R1b-M269 from that initial analysis (Figure-2 BTW doesn’t show the Anatolian samples taken by Balaresque et al(2009), but those collected by Myres et al(2010) and new ones from this study), in fact they did not exclude the Anatolian R1b-M269 at all*(I’ll explain what exactly happened). This is what Busby et al(2011) says about it exactly.
 
Quote from: Busby et al(2011)
We next calculated STR diversity for each population for the whole R-M269 lineage, and for the R-S127 and R-M269(xS127) sub-haplogroups, and investigated the relationship between average STR variance and longitude and latitude in exactly the same fashion as Balaresque. We provide estimates of uncertainty for these values by bootstrapping over individuals, and report the median of the observed variance values and its 95 per cent CI (figure 2). We normalized latitude and longitude, and performed a linear regression between these values and the median microsatellite variance for the three R-M269 sub-haplogroups. We found no correlation with latitude (data not shown) and, contrary to Balaresque, we did not find any significant correlation between longitude and variance for any haplogroup.

So in fact your confusion comes from an additional analysis done on Balaresque et al(2009) that was performed in the supplementary section, where they did test for variance excluding the Turkish haplogroups, and including them. Figure-2b shows that the highest variance for R1b-M269xS127 is not in Anatolia, and there is no East-West gradient, as the highest variance appears in Central Europe, rather than Anatolia. What differentiates Busby et al(2011) from Balaresque et al(2009) is that Busby et al(2011) tested for downstream SNPs(i.e. S127/L11, S-116, U-106, etc) and took the variance of R1b-M269xS127, and R1b-S127 separately; unlike Balaresque et al(2009) that did so without discriminating between the downstream clades and the basal clades. In any case, Figure-2a which is very similar to Balaresque et al(2009) where the total variance of R-M269 is taken regardless of subclades reveals that there is no such thing as an East-West pattern, and the Eastern-most variance appears in par with the Western-most, with the peak happening somewhere in the center(Coinciding with  the peak of variance in Figure-2b which also happens in Central Europe). 

Quote from: Busby et al(2011)
The Balaresque dataset presents genotype data only to the resolution of SNP R-M269. Our results show that the vast majority of R-M269 samples in Anatolia, approximately 90 per cent, belong to the R-M269(xS127) sub-haplogroup. Removing these Turkish populations from the Balaresque data and repeating the regression removes the significant correlation (R2 = 0.23, p = 0.09; details in the electronic supplementary material and figure S2). These populations are therefore intrinsic to the significant correlation.

So what they did was to test the Balaresque et al(2009) dataset by removing the Turkish populations, which is shown in Figure 2-B of the supplementary info pdf file. This removal showed that once the Anatolian samples are removed there is no longer an East-West gradient in terms of variance. Then recalculated they changed the Irish sample provided in Balaresque et al(2009) for one they gathered, and displayed both the Irish population and the Anatolian(Turkish) populations on the same figure-2D of the supplementary info pdf, and it turns out the Irish have a higher diversity than all of the Turkish population.  This is their procedure:

Quote from: Busby et al(2011)

Testing the variance calculations from the Irish population
Balaresque et al used haplotypes downloaded from the online Ysearch database (http://www.ysearch.org) which is a repository for genetic genealogists to upload and compare haplotypes (P. Balaresque pers. comm.). We note, however, that 17-STR haplotypes, including the 9 STRs used in Balaresque et al’s analysis, are available for 681 Irish R-M269 derived individuals in Moore et al (3), which is, in fact, the study which Balaresque et al use to estimate R-M269 frequency in Ireland. A subset of the Moore et al samples were re-analysed in the current study for SNPs downstream of R-M269, and the original haplotype data are used here to calculate variance. To test if the Ysearch  haplotypes were representative of the Irish R-M269 in Moore et al, we independently re-sampled the Moore et al dataset 10,000 times, selecting sub-samples of 75 haplotypes from which we estimated the variance using the same 9 STRs used in the Balaresque et al paper. The median variance of these 10,000 repetitions was 0.354 with a 95% CI of (0.285-0.432). The lowest variance value out of the 10,000 samples was 0.242, which is still higher than the figure observed in the Balaresque et al Ysearch sample (0.208). We therefore believe our estimate of Irish R-M269 variance to be a more robust representation of the true variance than that estimated by Balaresque et al. However, we note that the positive correlation between longitude and variance is still present after removing only the Irish and retaining the Balaresque et al Turkish populations. If we replace the variance calculated by Balaresque et al with that calculated from our repetitions, then the correlation is no longer significant, independent of whether or not we remove the Turkish samples (Figure S2).

That seems a major shortcoming of the paper to me. S127 is pretty plainly descended from R-M269xS127. The maps in Busby of the distributions of R-M269 as a whole, then of R-M269xS127 and R-S127 show the progression from east to west. Leaving out R-M269xS127 from the calculations merely shows that S127 expanded rapidly once it hit central and western Europe. Had R-M269xS127 haplotypes been included, the increased STR diversity everybody sees in the East, declining as one moves west and north, would have been evident once again.
Everyone is missing this point and acting as if Busby is a major blow to the idea that M269 is fairly recent and arrived in Europe from the East.[/size]

There is no major shortcoming, as I already showed above they did include R1b-M269xS127 variance calculation, and there is no East-West gradient, instead the highest diversity appears to happen in Europe(Possibly Central Europe, or Western Balkans). Again your whole argument here is based on the fact that you misunderstood the paper, not that the paper has any shortcomings.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: GoldenHind on August 27, 2011, 12:28:39 AM
I think one of the most interesting things in the Myres' data for Denmark is that all R1b subclades for which they tested: U106(XU198), U198, L21, U152 and P312* have their highest frequencies in northern Denmark- presumably northern Jutland.

The sampling locations are given in the map Fig S1 in the supplementary information,
http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp1.pdf

The supplementary data is given in Excel format at:
http://rspb.royalsocietypublishing.org/content/suppl/2011/08/18/rspb.2011.1044.DC1/rspb20111044supp2.xls

The Hg frequency for northern Denmark is 9.5%, as opposed to 4.1% and 5.3% for Southeast Denmark and West Denmark resp. Also, the Norwegian sampling location is 8%, nearly twice that of southern Jutland.

cheers
authun

The data from Denmark is just a repeat of the Myres' data, but that from Norway is new to this paper. Thanks very much for pointing that out. I find it so interesting that I'll start a new thread on it. Is there anyway of determining the size of their Norway sample?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Maliclavelli on August 27, 2011, 01:21:26 AM
There is no major shortcoming, as I already showed above they did include R1b-M269xS127 variance calculation, and there is no East-West gradient, instead the highest diversity appears to happen in Europe(Possibly Central Europe, or Western Balkans). Again your whole argument here is based on the fact that you misunderstood the paper, not that the paper has any shortcomings.

I appreciated your posting, whoever you are, but I'd make you note that Central Europe and Western Balkans are nearby Italy. If there is smoke, there is fire, and they are simply the regions of superimposition of different migrations of R1b1b2 from elsewhere, but the fire, I think, is in Italy. My studies about R-L51, published on many forums, demonstrated that its frequency crashes down to East, out of the Italian border.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 27, 2011, 04:12:36 AM
The data from Denmark is just a repeat of the Myres' data, but that from Norway is new to this paper. Thanks very much for pointing that out. I find it so interesting that I'll start a new thread on it. Is there anyway of determining the size of their Norway sample?

It's 138 and as you say Busby's own sample to supplement Myres. They sampled about 60 additional sites. If you can get to open the Excel file, they have listed all the STRs, around 2,200.

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 27, 2011, 05:42:12 AM
@Goldenhind

Just so I don't mislead you, the figures I posted are for R-S145* whereas I now realise that you emboldened "all R1b subclades". If you can't open the Excel, the figures for Norway are:

Norway (n=138)
R-M269         0.442
R-S167         0.428
R-M269(xS127)   0.014
R-S127*         0.428
R-S127(xS21,S116)   -----
R-S21*         0.203
R-S21(xS29)      0.203
S29         -----         
R-S116*         0.225
R-S116(xS145,S28)   0.109
R-S145*         0.080
R-S145(xM222)   0.065
R-M222         0.014
R-S28         0.036


To save you digging out Myres:

Denmark North (n=42)
R-M269         0.476
R-S167         0.452
R-M269(xS127)   0.024
R-S127*         0.452
R-S127(xS21,S116)   -----
R-S21*         0.238
R-S21(xS29)      0.214
S29         0.024
R-S116*         0.214
R-S116(xS145,S28)   0.071
R-S145*         0.095
R-S145(xM222)   0.095
R-M222         ------
R-S28         0.048

Denmark West (n=19)
R-M269         0.368
R-S167         0.316
R-M269(xS127)   0.053
R-S127*         0.316
R-S127(xS21,S116)   -----
R-S21*         0.211
R-S21(xS29)      0.211
S29         -----
R-S116*         0.105
R-S116(xS145,S28)   0.053
R-S145*         0.053
R-S145(xM222)   0.053
R-M222         -----
R-S28         -----

Denmark Southeast (n=49)
R-M269         0.265
R-S167         0.265
R-M269(xS127)   -----
R-S127*         0.265
R-S127(xS21,S116)   0.041
R-S21*         0.122
R-S21(xS29)      0.122
S29         -----
R-S116*         0.102
R-S116(xS145,S28)   0.041
R-S145*         0.041
R-S145(xM222)   0.041
R-M222         -----
R-S28         0.020

I'm not used to this S notation which I presume is down to Jim Wilson. I know it's not your usual but thought it better to leave the translation up to you. Hope I haven't misled you.

cheers
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 27, 2011, 08:23:42 AM


I’m sorry to break it to you, but I think that you are in fact wrong. Bubsy et al(2011) did do an analysis of the STR variance of R1b-M269xS127, it is shown in Figure-2b)
. . .



So, Jean, if there was no correlation between variance and longitude for M269 as a whole, why do Busby et al say this (on page 5; I have underlined and italicized part of it for emphasis)?

"The Balaresque dataset presents genotype data only to
the resolution of SNP R-M269. Our results show that the
vast majority of R-M269 samples in Anatolia, approximately
90 per cent, belong to the R-M269(xS127)
sub-haplogroup. Removing these Turkish populations
from the Balaresque data and repeating the regression
removes the significant correlation
(R2 ¼ 0.23, p ¼ 0.09;
details in the electronic supplementary material and
figure S2). These populations are therefore intrinsic to
the significant correlation.
"


And this (on page 6; again the underlining and italics are mine)?

"Here, we have confirmed with the broadest analysis to
date that the spatial distribution of Y chromosome haplogroup
M269 can be split by R-S127 into European
and western Eurasian lineages. Contrary to the results
of Balaresque, we see no relationship between diversity
and longitude (figure 2) for R-M269. The presence of
two sets of populations in the Balaresque paper appears
to be causal to the observed relationship
: the underestimated
diversity of the Irish population and the inclusion
of the Turkish chromosomes
, the majority of which
potentially belong to the non-European clade
R-M269(xS127). When these elements are properly
taken into account, jointly or independently, the correlation
no longer exists. This correlation is the central tenet to
the hypothesis that R-M269 was spread with expanding
Neolithic farmers."


If Busby et al indeed found that there was no east-to-west variance cline in the whole M269 set, why say that removing the Turkish sample "removes the significant correlation" and that "[t]hese populations [i.e., the Turkish] are therefore intrinsic to the significant correlation"?

Why say that if they indeed found no correlation with the Turks included?

And why say, "The presence of two sets of populations [i.e., both the S127 and M269xS127 sets] in the Balaresque paper appears to be causal to the observed relationship" if in fact there is no observed relationship?

If I misunderstood this paper, and it is as straightforward and devastating as you seem to be saying, then these passages are the reason I misunderstood. But it sure sounds like they had to tinker with Balaresque's datasets a bit, mainly by excluding the M269xS127 haplotypes, in order to "remove the significant correlation" and "the observed relationship".

Why bother with that if an analysis of the whole set of M269 haplotypes had already demonstrated that there is no east-to-west decrease in variance?

Honestly, maybe I'm wrong, but it sounds to me like Busby et al did, in fact, find an increase in variance in M269 as a whole as one heads east and had to remove the M269xS127 haplotypes in order to eliminate it. (I must confess, however, that those "bootstrap variance" graphs in Figure 2 are Greek to me, as I am sure they are to most of us here.)

I guess I did misunderstand the paper, if in fact they measured the variance of M269 as a whole, did not exclude R-M269xS127, and found no increase in variance as one heads east. Apparently I am not the only one here who misunderstood it. The passages I quoted above, however, tend to convince me that I did not misunderstand the paper at all.

Where does the paper say the Irish haplotypes had greater variance than the Turkish haplotypes? I missed that. Or did they mean just the Turkish S127 haplotypes?

What baffles me about this paper, as a layman, is that everything I have seen to date, not just Balaresque, has concluded that M269 variance increases as one heads east across Europe into western Asia. That includes Vince Vizachero's fairly recent calculations. Not only that, but it is pretty plain that the SNP trail leads from east to west, as well.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 27, 2011, 09:51:20 AM
....
What baffles me about this paper, as a layman, is that everything I have seen to date, not just Balaresque, has concluded that M269 variance increases as one heads east across Europe into western Asia. That includes Vince Vizachero's fairly recent calculations. Not only that, but it is pretty plain that the SNP trail leads from east to west, as well.
I'm okay with whatever the answer is regarding R-M269's direction and timeframe of expansion and think it is worthwhile to consider all viewpoints and accept whatever is most probable. In fact, it felt good to be a tall, dark, artistic Cro-Magnon man.

I think there are a couple of key points to evaluate and some challenges. The challenges include 1) identifying the best STR set and appropriate mutation rates to use, 2) breaking up the asterisk paragroups further, particularly R-L11*, R-L23*, etc. and 3) low sample sizes of deep clade tested long haplotypes from Central Europe to West Asia.

The key points to evaluate are STR variance and SNP diversity against the SNP based phylogeny. If R-L11*, and probably more importantly R-L23xL11, appear to be dominated by a few "eastern" subclades and those subclades have low STR variance then I think the argument for a central European origin for R-L23 is good. This would follow with a concept along the ht35 idea where ht15 types expanded west while ht35 expanded east.

However, what I can see so far of L11*, anyway, I don't think L11* is just a few subclades. I tried to assign the 47 long haplotypes I've found to clusters. I could assign 12 of the 47 to 4 different clusters that had potential off-modal STR signatures. These probably should be called deep ancestral varieties, though, as they are not generally of the genealogical timeframe. I checked the GD's of the rest of the R-L11* (the non-clustered ones) and found the closest GD's over 67 to others in the group were in the upper teens (17, 18 seemed to about the norm but sometimes up to 21,22 or down to 15,16.) That would be for the closest individual, the furthest individual from each of the non-clustered folks would usually be 30 or just over 30. It looks like a lot of small subclades to me are in R-L11* so finding a place with a lot of R-L11* folks would be helpful in determining origin.

Probably, R-L23* is more important as the to be determined  R-L11* hot spot appears to sit at the foot of the great expansion in Europe with its sons P312 and U106 near by. BTW, the variance numbers I calculated for R-L11* are about 5 to 20% greater than P312's. For all of the trashing of STR variance, it still seems to produce usable information.

The reason I say R-L23* is probably most important is that R-L11 could really be based out of the core of Europe. However, by the time we travel the bridge on R-L23xL11 back to the R-L23 MRCA we are very likely to be in some place like SW Asia or the Caucasus. So it would be really important to determine if R-L23xL11 was just a couple of subclades that were "eastern" or if they were a large number of unclusterable folks. If there are many diverse subclades in R-L23xL11 then you have high haplogroup diversity and a likely expansion point.

Perhaps Busby has a strong point in that R-L11 is about the same diversity all across Europe and therefore may be truly of European origin. I don't think anyone is saying that couldn't be true.  However, as you move up the R1b phylogenetic tree from L11 you are going to run into the inevitable - the origins of R-M269 and finally R-M343.

Does anyone think R1b-M73 got to Anatolia, the Caucasus, the Urals and Pakistan from Europe?
Does anyone think R1b-V88 got to Sub-Sahara Africa from Europe?





Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: IALEM on August 27, 2011, 10:32:15 AM


He doesn't think it is likely as, if this was the case, its dominance in europe could only be explained by a replacement of the early neolithic farmers. However, he doesn't appear to be aware of the population crash at the end of the LBK reported by Shennan and Edinborough. The crash which affected several parts of northern europe reduced the populations to their low mesolithic levels and remained at these low levels for nearly 1000 years.

best
authun
Well, they in fact cite Bucquet and de Miguel demographic model for to the second wave of Neolithic regional centers to explain the regional pattern of S-127 subclades, however they fail to see that it fits equally a model of postNeolithic replacement of population.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanLo on August 27, 2011, 11:34:44 AM

So, Jean, if there was no correlation between variance and longitude for M269 as a whole, why do Busby et al say this (on page 5; I have underlined and italicized part of it for emphasis)?

"The Balaresque dataset presents genotype data only to
the resolution of SNP R-M269. Our results show that the
vast majority of R-M269 samples in Anatolia, approximately
90 per cent, belong to the R-M269(xS127)
sub-haplogroup. Removing these Turkish populations
from the Balaresque data and repeating the regression
removes the significant correlation
(R2 ¼ 0.23, p ¼ 0.09;
details in the electronic supplementary material and
figure S2). These populations are therefore intrinsic to
the significant correlation.
"


And this (on page 6; again the underlining and italics are mine)?

"Here, we have confirmed with the broadest analysis to
date that the spatial distribution of Y chromosome haplogroup
M269 can be split by R-S127 into European
and western Eurasian lineages. Contrary to the results
of Balaresque, we see no relationship between diversity
and longitude (figure 2) for R-M269. The presence of
two sets of populations in the Balaresque paper appears
to be causal to the observed relationship
: the underestimated
diversity of the Irish population and the inclusion
of the Turkish chromosomes
, the majority of which
potentially belong to the non-European clade
R-M269(xS127). When these elements are properly
taken into account, jointly or independently, the correlation
no longer exists. This correlation is the central tenet to
the hypothesis that R-M269 was spread with expanding
Neolithic farmers."


Because they are talking about two completely different sets, they showed that in their set (A combination of new samples+Myres et al(2010) which does have Anatolian populations in it and can be seen in Figure S1 of the supplementary table) there is no correlation from East to West for R1b-M269, which is what we observed in Figure 2a). This figure is the most similar, in terms of analysis, to what Balaresque et al(2009) did, yet in their samples there isn’t any observable East-West correlation. Then on figure-2b) he shows the variance of R1b-M269(xS127), and this map showed that the highest variance occurs in Central Europe, if you look at the frequency map, Western Europe has barely nothing, so it wasn’t even included in the variance calculation. Now the next thing they(Busby et al(2011)) did, was to do a re-analysis of Balaresque et al(2009), and that is what you quoted above. But the analysis was done on the supplementary info, not on the main page of the study.

Quote from: Busby et al(2011)
"The Balaresque dataset presents genotype data only to the resolution of SNP R-M269. Our results show that the vast majority of R-M269 samples in Anatolia, approximately 90 per cent, belong to the R-M269(xS127) sub-haplogroup. Removing these Turkish populations from the Balaresque data and repeating the regression removes the significant correlation (R2 ¼ 0.23, p ¼ 0.09; details in the electronic supplementary material and figure S2). These populations are therefore intrinsic to the significant correlation.

So basically what they did, was a re-analysis of Balaresque et al(2009) data with and without the Turkish population, those analysis are shown in Figure-S2, they also included a different Irish sample, which turned out to have a median variance of 0.35, which in turn, completely destroyed  the high variance of the Turkish sample.(See Figure-S2 D).


If Busby et al indeed found that there was no east-to-west variance cline in the whole M269 set, why say that removing the Turkish sample "removes the significant correlation" and that "[t]hese populations [i.e., the Turkish] are therefore intrinsic to the significant correlation"?

Why say that if they indeed found no correlation with the Turks included?

They are saying that once the Anatolian population sets are removed from the Balaresque et al(2009) set there isn’t any longer an East-West, so this argument tackles more the whole “Mutation arising in a surf-wave theory”. Because once the Anatolian populations are removed from Balaresque et al(2009), the European populations do not show any East-West gradient. Nonetheless, when they change the Balaresque et al(2009) Irish sample for a new one, it turned out that the new Irish sample with variance 0.35 changed the whole dynamics of the gradient, causing an East-West dichotomy(With both IR and TK3 now having  the highest variance).  Now like I said before, this is an analysis done on the Balaresque et al(2009) dataset, it has nothing to do with what is shown in Figure-2b), which was an analysis done on the variance of Busby et al(2011) samples and Myres et al(2010)  samples.

And why say, "The presence of two sets of populations [i.e., both the S127 and M269xS127 sets] in the Balaresque paper appears to be causal to the observed relationship" if in fact there is no observed relationship?

If I misunderstood this paper, and it is as straightforward and devastating as you seem to be saying, then these passages are the reason I misunderstood. But it sure sounds like they had to tinker with Balaresque's datasets a bit, mainly by excluding the M269xS127 haplotypes, in order to "remove the significant correlation" and "the observed relationship".

Again removing the Anatolian samples from Balaresque et al(2009) was just one analysis(To test intra-European variance) done on the supplementary info. Another analysis was the inclusion of an Irish sample that had a variance of 0.35, which turned out to be higher than that of the three Anatolian populations from Balaresque et al(2009). So there isn’t just one thing that seems to conflict with Balaresque et al(2009) but two:

1-   A new data set(Busby+Myres) showing no East-West gradient in terms of variance for R1b-M269(xS127), or R1b-M269 in general.
2-   The fact that re-analyzing the Balaresque et al(2009) with both removal of Anatolian samples, and change of Irish sample removed the East-West gradient. 

Why bother with that if an analysis of the whole set of M269 haplotypes had already demonstrated that there is no east-to-west decrease in variance?

Because had they not re-analyzed Balaresque et al(2009) then you would have their dataset vs. Balaresque et al(2009) dataset, and there is no way they could say that Balaresque et al(2009) data set is invalidated. Instead they did that re-analysis of the other set.



Where does the paper say the Irish haplotypes had greater variance than the Turkish haplotypes? I missed that. Or did they mean just the Turkish S127 haplotypes?

Look in the supplementary info pdf, under the section titled: “Re-analysis of the Balaresque dataset”, then look at the sub-section titled: “Testing the variance calculations from the Irish population”. They used all the haplogroups present in R1b-M269, because this was a re-analysis of Balaresque et al(2009) which did not type in any downstream SNPs.


What baffles me about this paper, as a layman, is that everything I have seen to date, not just Balaresque, has concluded that M269 variance increases as one heads east across Europe into western Asia. That includes Vince Vizachero's fairly recent calculations. Not only that, but it is pretty plain that the SNP trail leads from east to west, as well.

I think the situation is far more complicated than that(i.e. There are more possibilities than Magdalenian expansion from FC Refuge, or Neolithic expansion from Anatolia, or Bronze Age expansion from the Caucasus, and hopefully they would be taken into account soon). Certainly ancient DNA isn’t helping out much either, it is better to look at variance calculations in terms of basal vs. downstream haplogroups (i.e. Variance for R1b-M269(xS127), variance for R1b-S116, variance for R1b-U106). Let’s hope in the future aDNA would help us out with this puzzle, in any case, I look forward to Otzi the Iceman Y-DNA Haplogroup.

PS: I don’t know what’s going on, but this is the third  account I have to create to be able to log in, everytime I have created an account(JeanLohizun, JeanL and now JeanLo) I keep getting this message:
“Unable to send a message, please contact the administrator”
Then whenever I log out and I try to log back in, it doesn’t let me, because I was never asked to set up a password in the first place. Then when I ask it to send me a password change link to my email, it never does it.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 27, 2011, 12:16:10 PM
Well, they in fact cite Bucquet and de Miguel demographic model for to the second wave of Neolithic regional centers to explain the regional pattern of S-127 subclades, however they fail to see that it fits equally a model of postNeolithic replacement of population.

Yes, I saw that but they appear to have difficulty, understandably, reconciling it with an existing population in situ. Shennan's curves, if correct, would provide an explanation:

http://www.cpt.co.uk/maps/shennan_pop_curves.gif

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 27, 2011, 12:19:24 PM

PS: I don’t know what’s going on, but this is the third  account I have to create to be able to log in, everytime I have created an account(JeanLohizun, JeanL and now JeanLo) I keep getting this message:
“Unable to send a message, please contact the administrator”
Then whenever I log out and I try to log back in, it doesn’t let me, because I was never asked to set up a password in the first place. Then when I ask it to send me a password change link to my email, it never does it.

I sent Terry (the site owner) an email about your login problems. Hopefully, he'll have the problem fixed soon.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 27, 2011, 12:41:26 PM
@Goldenhind

Just so I don't mislead you, the figures I posted are for R-S145* whereas I now realise that you emboldened "all R1b subclades". If you can't open the Excel, the figures for Norway are:

Norway (n=138)
R-M269         0.442
R-S167         0.428
R-M269(xS127)   0.014
R-S127*         0.428
R-S127(xS21,S116)   -----
R-S21*         0.203
R-S21(xS29)      0.203
S29         -----         
R-S116*         0.225
R-S116(xS145,S28)   0.109
R-S145*         0.080
R-S145(xM222)   0.065
R-M222         0.014
R-S28         0.036

. . .

I notice that, while the total S145 (L21) figure is fairly high (8%), a sizable chunk of that is M222 (1.4% of the y-dna total, about 18% of the L21 total). That makes me think that perhaps the notion that much of the L21 in Norway is the product of Viking Period slave trade might be correct. I am really reluctant to admit that, and admitting it won't make me any friends, that's for sure.

But the presence of British Isles-specific L21 subclades in Norway, like M222, certainly weakens the argument that L21 predates the Viking Period there. Besides that, we have one or two (I can't remember whether it's one or two) Norwegians in our project who are pretty close matches to the Scots Modal, and I believe there is at least one Norwegian who is L159.2+, which defines a subclade that is also otherwise pretty Isles-localized.

If Busby's sample is representative, then U152 is pretty anemic in Norway, and could also very likely be attributed to the Viking slave trade, since U152 is much more common in the British Isles than it is in Norway.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 27, 2011, 02:36:14 PM
The situation with L21 in Scandinavia has always been different in that the variance is much lower than say France.  However, I think the emphasis on slaving the Celtic peoples seems a bit odd when you consider that their biggest impact was annexing half of England for some time.  You would think that would have given them a far bigger prey than raiding down the low populated rocky islands off western Scotland and setting up town enclaves in Ireland.   


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 27, 2011, 02:49:26 PM
The situation with L21 in Scandinavia has always been different in that the variance is much lower than say France.  However, I think the emphasis on slaving the Celtic peoples seems a bit odd when you consider that their biggest impact was annexing half of England for some time.  You would think that would have given them a far bigger prey than raiding down the low populated rocky islands off western Scotland and setting up town enclaves in Ireland.   

Undoubtedly, which is why probably most if not all of the U152 in Scandinavia and much of the U106 must also be products of the Viking Era slave trade, if the L21 is.

The presence of M222 at about 18% of the L21 in Norway is a puzzler, however, if the L21 there predates the Viking Period.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Terry Barton on August 27, 2011, 03:05:16 PM
Heber - I just fixed your 3 accounts.  Sorry about that.  I need my programmer t ofix the underlying problem.  Hoping he is back in action this coming week.  Terry


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 27, 2011, 03:46:22 PM
Heber - I just fixed your 3 accounts.  Sorry about that.  I need my programmer t ofix the underlying problem.  Hoping he is back in action this coming week.  Terry

Terry. I believe you are referring to Jean's account. I have one account and it works fine.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: GoldenHind on August 27, 2011, 05:06:31 PM
Regarding the Norway data in Busby study: it really deserves a topic on its own, which I intend to start when I have time (I am rather pressed for time at the moment). Thanks again to Authun for pointing it out and determining the sample size. The next question is the area of sampling and the source. Is it country wide or limited to a specific area? Since Jim Wilson is one of the co-authors and has been said to have conducted  sampling in Norway, I wonder if he might be the source.

Regarding M222: I don't see how one can put all M222 in Norway down to Viking slavery. It ignores the possibility of emigration from Britain to Norway since the Viking period. It also assumes M222 is specifically a British marker. I think the jury is still out on that issue. Jean M has suggested that it may have originated in the La Tene culture. If so, it may have arrived in Norway in the Bronze Age or other time before the Vikings.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 27, 2011, 07:19:20 PM
Regarding the Norway data in Busby study: it really deserves a topic on its own, which I intend to start when I have time (I am rather pressed for time at the moment). Thanks again to Authun for pointing it out and determining the sample size. The next question is the area of sampling and the source. Is it country wide or limited to a specific area? Since Jim Wilson is one of the co-authors and has been said to have conducted  sampling in Norway, I wonder if he might be the source.

Regarding M222: I don't see how one can put all M222 in Norway down to Viking slavery. It ignores the possibility of emigration from Britain to Norway since the Viking period. It also assumes M222 is specifically a British marker. I think the jury is still out on that issue. Jean M has suggested that it may have originated in the La Tene culture. If so, it may have arrived in Norway in the Bronze Age or other time before the Vikings.

The biggest question mark for me is over whether the variance dating method is correct.  I just cant make sense of Europe having been repopulated in the Bronze Age.  It just doesnt tally with the archaeological evidence.  If the dates are systematically coming out too young then you wonder if this is also true for the dating of M222 etc.  If they are a lot older than we think then the whole ball game changes.  If M222 is a BC marker then a whole range of possibilities opens up compared to a 400AD date as currently suggested. 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 28, 2011, 07:46:26 AM
Regarding the Norway data in Busby study: it really deserves a topic on its own, which I intend to start when I have time (I am rather pressed for time at the moment). Thanks again to Authun for pointing it out and determining the sample size. The next question is the area of sampling and the source. Is it country wide or limited to a specific area? Since Jim Wilson is one of the co-authors and has been said to have conducted  sampling in Norway, I wonder if he might be the source.

Regarding M222: I don't see how one can put all M222 in Norway down to Viking slavery. It ignores the possibility of emigration from Britain to Norway since the Viking period. It also assumes M222 is specifically a British marker. I think the jury is still out on that issue. Jean M has suggested that it may have originated in the La Tene culture. If so, it may have arrived in Norway in the Bronze Age or other time before the Vikings.

The biggest question mark for me is over whether the variance dating method is correct.  I just cant make sense of Europe having been repopulated in the Bronze Age.  It just doesnt tally with the archaeological evidence.  If the dates are systematically coming out too young then you wonder if this is also true for the dating of M222 etc.  If they are a lot older than we think then the whole ball game changes.  If M222 is a BC marker then a whole range of possibilities opens up compared to a 400AD date as currently suggested. 

Authun has mentioned the Shennan and Edinborough study, which I also remember reading a couple of years ago. It showed extreme depopulation at the end of the Neolithic, so it wouldn't have taken all that much for even small groups of incomers to replace most of the y-dna population (and that's the part we're talking about - not the whole population).

I don't think STR variance is off by that much. I think it's probably pretty accurate. Busby is a tricky piece of work and not even a straightforward piece of writing. It has to be sorted through its supplementary pages to be understood and used just 10 STR markers. I don't think it's the final word on the subject. It also strikes me that it may represent a volley from those who have a lot invested (emotionally and in terms of academic reputation) in the old longue duree idea: R1b-in-the-Ice-Age-Refuge, blah, blah, blah.

M222 has been known about for a long time (since at least 2005, if I recall correctly) and has only very rarely been found on the Continent. It's about 18% of Busby's Norwegian L21 sample. Believe me, I DO NOT want to believe that most of the L21 in Norway has to be chalked up to the Viking slave trade, but that M222 figure makes me think it has to be.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Bren123 on August 28, 2011, 10:24:40 AM


Authun has mentioned the Shennan and Edinborough study, which I also remember reading a couple of years ago. It showed extreme depopulation at the end of the Neolithic,


Wasn't there a collapse at the end of the bronze age,also I think there was some climatic changes at this time as well!?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 28, 2011, 02:24:14 PM
Authun has mentioned the Shennan and Edinborough study, which I also remember reading a couple of years ago. It showed extreme depopulation at the end of the Neolithic, so it wouldn't have taken all that much for even small groups of incomers to replace most of the y-dna population (and that's the part we're talking about - not the whole population).

End of the LBK in the neolithic, not the whole neolithic.

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 28, 2011, 06:32:45 PM
...
The presence of M222 at about 18% of the L21 in Norway is a puzzler, however, if the L21 there predates the Viking Period.
What's the % that M222 is of L21 in Ireland? I think it is more than 18%, isn't it?

If we use the slower markers, as I think Tim Janzen and pretty much Mth69 advocate, M222 is close to half the age of L21. Depending on how you age L21 there is plenty of time pre-Viking (900AD or so) for M222 to show up in Norway.

I think another consideration should be that the origin of M222 is not completely nailed down as Ireland. Jean M has proposed it came to Scotland and then to NW Ireland via the La Tene Celts. We know there are a few M222 in Germany so who knows?

If M222 came from Germany across the North Sea to northern Great Britain and then Ireland, some could have diverted to Norway as well but just not have been so prolific there. M222 could have made it to Norway pre-Viking, during the Viking period and post the Viking period as merchants.

For my money, the better diagnostic marker for Ireland is L226+ which has been associated with Dalcassian surnames. I haven't found any L226 outside of the Isles, period and it is very, very much Munster based with some spill-over into the other regions. Of course, maybe these guys were just plain Viking adverse. Isn't Brian Boru a Dalcassian?

Also, L371+ is pretty much Wales/West Britain oriented. L371+ in Norway would a strong indicator.

Both L371 and L226 have strong off-modal STR patterns that I haven't seen in Norway.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 29, 2011, 09:36:58 AM
...
The presence of M222 at about 18% of the L21 in Norway is a puzzler, however, if the L21 there predates the Viking Period.
What's the % that M222 is of L21 in Ireland? I think it is more than 18%, isn't it?

If we use the slower markers, as I think Tim Janzen and pretty much Mth69 advocate, M222 is close to half the age of L21. Depending on how you age L21 there is plenty of time pre-Viking (900AD or so) for M222 to show up in Norway.

I think another consideration should be that the origin of M222 is not completely nailed down as Ireland. Jean M has proposed it came to Scotland and then to NW Ireland via the La Tene Celts. We know there are a few M222 in Germany so who knows?

If M222 came from Germany across the North Sea to northern Great Britain and then Ireland, some could have diverted to Norway as well but just not have been so prolific there. M222 could have made it to Norway pre-Viking, during the Viking period and post the Viking period as merchants.

For my money, the better diagnostic marker for Ireland is L226+ which has been associated with Dalcassian surnames. I haven't found any L226 outside of the Isles, period and it is very, very much Munster based with some spill-over into the other regions. Of course, maybe these guys were just plain Viking adverse. Isn't Brian Boru a Dalcassian?

Also, L371+ is pretty much Wales/West Britain oriented. L371+ in Norway would a strong indicator.



Mike,
Moffat and Wilson clearly identify M222 as North West Ireland. They also link it to the Dalriadian signature in the Scottish Isles.
The Clan Donald who were a hybrid Viking Dalriadan clan and were Lords of the Isles for several centuries, had extensive exchanges with Norway. That and the Viking slave markets based out of Dublin could explain some of the M222 in Norway.

“As a result of a process coyly termed social selection, scientists have identified an old lineage in Ireland dating from around 400 to 500. Known as M222, it is astonishingly common. No less than 20 per cent of all Irish men carry it! Its distribution is heavily weighted to the north with 40 percent in Ulster, 30 per cent in Connaught and 10 – 15 percent in in Munster and Leinster. No less than a fifth of all Irish men are directly descended from one man who lived 1,500 years ago”.
“Given the distribution of the marker and its bias to Ulster and especially to men with the O’Neill and O’Donnell surnames, there exists a clear candidate. The O’Neill kindred dominated Irish history from the fifth to the tenth centuries and their founder was the High King known as Niall Noigiallach”.
“Lord  Turlough O’Donnell who died in 1423, carried on the family tradition with gusto. He had 14 sons and 59 male grandchildren.”
“More than 50,000 Scottish men, most of them with the surname MacDonald or its varients, are the direct decendents of Semerled. The first Lord of the Isles and founder of Clan Donald, he ruled the Hebrides and was King of the Isle of Man”.
“Did it (the M222 marker) cross the sea with the war bands of Fergus Mor mac Erc and his ancestors? There is uncompromising evidence that it did. More than 6 percent of all Scottish men carry M222, around 150,000 are direct decendents of Niall, the High King of the Irish. The frequency of the marker is very pronounced on the west with 9 per cent and less in the east with 3 per cent on the axis from Galloway to Shetland. It occurs very often amongst men with ancient Scottish surnames and whose family trees can, in some cases, be traced back over three centuries. Those in Scotland with the M222 marker are not recent immigrants and their high incidence and geographic spread indicate a large scale movement of people – probably mainly from Ulster and probably around AD 500”.

http://www.box.net/shared/gsbm92c2ri

The frequencies of the M222 Y chromosome group are shown across the British Isles using pie charts. Up to 3000 samples were used to create this map.
“Other Irish specific markers from the period around AD 500 can be found in Scotland and their presence reinforces a sense of colonization. S168 (M226) is relatively rare and strongly concentrated around the River Shannon where it is now found in Tipperary and Limerick. This was once the territory of the Dalcassian clans, the decendents of the great High King Brian Boru. S169 (L159.2) is most common in Leinster, the lands of the Lagin clans, and it too is found in Scotland,  especially amongst men with the surnames Beattie and Ferguson”.
“The genetic and political divide between incomers and natives also had cultural facets. The Dalriadans spoke Irish Gaelic, Q-Celtic, while the Picts spoke P-Celtic and it appears that the languages were not mutually intelligible. When St Columba attempted to bring the Word of God to the Picts, it had to be translated”.
“A marker has been identified that is essentially unique to Scotland and rarely found elsewhere. It is known as R1b-str47 or R1b-Pict and around 10% of Scottish men carry it. In our towns and villages 250,000 Picts are quietly going about their daily lives. The distribution of the marker broadly matches the Pictish territory and where later incursions such as Dalriada Gaels and the Vikings overlaid it, the numbers are diluted.It is well represented in the east of Scotland above the Forth but much less so in Northern and Western Isles. R1b-Pict is at least 3,000 years old and possibly even older and is a subgroup of S145 (L21)”.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on August 29, 2011, 09:40:54 AM
Dienekes has posted another update to his blog on the subject and Anatole has replied to it. It makes for quiet a heated debate especially as both are leaders in their respective fields.

http://dienekes.blogspot.com/


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 29, 2011, 10:08:45 AM
What would explain the percentages of R-S145 and M222 found in South West France which are similar to those found in Norway?

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 29, 2011, 10:33:21 AM
What would explain the percentages of R-S145 and M222 found in South West France which are similar to those found in Norway?
Wow, I'll have to catch up with the numbers. I didn't know we had found M222 in SW France.  What's the numbers? Are these inland or coastal?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 29, 2011, 10:44:34 AM
.. Moffat and Wilson clearly identify M222 as North West Ireland. They also link it to the Dalriadian signature in the Scottish Isles.
The Clan Donald who were a hybrid Viking Dalriadan clan and were Lords of the Isles for several centuries, had extensive exchanges with Norway. That and the Viking slave markets based out of Dublin could explain some of the M222 in Norway.
“As a result of a process coyly termed social selection, scientists have identified an old lineage in Ireland dating from around 400 to 500. Known as M222, it is astonishingly common. No less than 20 per cent of all Irish men carry it! Its distribution is heavily weighted to the north with 40 percent in Ulster, 30 per cent in Connaught and 10 – 15 percent in in Munster and Leinster. No less than a fifth of all Irish men are directly descended from one man who lived 1,500 years ago”.
“Given the distribution of the marker and its bias to Ulster and especially to men with the O’Neill and O’Donnell surnames, there exists a clear candidate. The O’Neill kindred dominated Irish history from the fifth to the tenth centuries and their founder was the High King known as Niall Noigiallach”.
“Lord  Turlough O’Donnell who died in 1423, carried on the family tradition with gusto. He had 14 sons and 59 male grandchildren.”
“More than 50,000 Scottish men, most of them with the surname MacDonald or its varients, are the direct decendents of Semerled. The first Lord of the Isles and founder of Clan Donald, he ruled the Hebrides and was King of the Isle of Man”.
“Did it (the M222 marker) cross the sea with the war bands of Fergus Mor mac Erc and his ancestors? There is uncompromising evidence that it did. "
“A marker has been identified that is essentially unique to Scotland and rarely found elsewhere. It is known as R1b-str47 or R1b-Pict and around 10% of Scottish men carry it. In our towns and villages 250,000 Picts are quietly going about their daily lives. The distribution of the marker broadly matches the Pictish territory...
I have that book, too. I've found there are a lot of R1b-str47 (Scots modal) people who think they are Dal Riada descendants, but if I had to choose I'd go with Moffat and Wilson on this one that Scots Modal may have an ancient Pictish linage. However, I think some Picts made it into Ireland very early on so it is possible for a Pictish lineage to be "Dál Riata-ized" and then to have lea the charge back into Scotland to take it from their old cousins.

In terms of M222, I don't think Moffat and Wilson are specifying it true origin is NW Ireland.  If you believe the Irish myth stuff, it should be more to west and central anyway as it is Conor (of a hundred battles) related lineage.

I don't really have an opinion on if the M222 guys were Dál Riata. However, Jean has the proposal that M222 actually came from the continent and I can say that variance for M222 is higher in Great Britain than in Ireland, based on MDKAs that are pre-Industrial era and English sounding surnames.

I know the Dál Riata were supposed to be powerful and all that. Were they good sailors to? We know they were island hoppers.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 29, 2011, 11:38:32 AM
Wow, I'll have to catch up with the numbers. I didn't know we had found M222 in SW France.  What's the numbers? Are these inland or coastal?

South West France (n=83)
from Ramos-Luis et al 2009 for
Midi-Pyrénées

R-M269         0.627
R-S167         0.590
R-M269(xS127)   0.084
R-S127*         0.542
R-S127(xS21,S116)   0.012
R-S21*         0.036
R-S21(xS29)      0.036
S29         0
R-S116*         0.494
R-S116(xS145,S28)   0.313
R-S145*         0.072
R-S145(xM222)   0.060
R-M222         0.012
R-S28         0.108

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 29, 2011, 01:02:08 PM
Interesting that the sample from SE France was nearly 20% L21+ (S145+), while U152 (S28) was about 13%. P312xL21,U152 was 29% there, but that would be divided up among the subclades not tested for this paper and would probably not all be P312*.

Here are some interesting facts from the Busby paper relative to L21, facts that are really hardly if at all discernible from the L21 (S145) map in the supplementary info section.

Rennes in Bretagne had a frequency of 40% L21xM222 (not surprising).

Poitiers had 14%.

Marseille was 11% L21xM222.

Lille, up in the northeast near the Belgian border, had 10% L21xM222 (that one did surprise me - it is higher than I thought it would be).

Paris had 10% L21xM222.

Chalon-sur-Saône, in the east, near Dijon, had 8% L21xM222.

These frequencies are percentages of the total y-dna. That means L21 is actually huge all over France. I mean, 8% of the total y-dna in an area is pretty significant, let alone 10% and up.

If the shading scale on Busby's L21 map had been like those of its U106 and U152 maps, the true L21 picture in France would have been more readily apparent.

Yes it weird the way things seem to conspire against continental L21.  I would imagine over 10% of the population translates into something like 15-20% of R1b in France. 

I havent looked at the detail for Germany.  Where is the peak in the Rhineland?  Did they sample the Rhineland? 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 29, 2011, 02:45:07 PM
Wow, I'll have to catch up with the numbers. I didn't know we had found M222 in SW France.  What's the numbers? Are these inland or coastal?

South West France (n=83)
from Ramos-Luis et al 2009 for
Midi-Pyrénées

R-M269         0.627
R-S167         0.590
R-M269(xS127)   0.084
R-S127*         0.542
R-S127(xS21,S116)   0.012
R-S21*         0.036
R-S21(xS29)      0.036
S29         0
R-S116*         0.494
R-S116(xS145,S28)   0.313
R-S145*         0.072
R-S145(xM222)   0.060
R-M222         0.012
R-S28         0.108

best
authun

One out of 83 is the number of M222+ they found in that sample. There were five L21xM222 guys, apparently.

I would say that singleton M222 is an outlier. You can't say much more than that when you only find one.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 29, 2011, 03:35:08 PM
One out of 83 is the number of M222+ they found in that sample. There were five L21xM222 guys, apparently.

I would say that singleton M222 is an outlier. You can't say much more than that when you only find one.

But these arguments apply equally to the Norwegian sample.

best
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 29, 2011, 04:44:38 PM
One out of 83 is the number of M222+ they found in that sample. There were five L21xM222 guys, apparently.

I would say that singleton M222 is an outlier. You can't say much more than that when you only find one.

But these arguments apply equally to the Norwegian sample.
One or two does not a trend make, but we definitely should not dismiss the idea that M222 was in Norway or the Midi-Pyrénées or Germany from prior to the M222 heyday times in NW Ireland and the lowlands of Scotland.

I know the STR's of the M222 in Norway but don't in Germany and southern France. It would interesting to see if they are different than standard M222.

The one thing about the Midi-Pyrénées is that is I don't think it is thought of as a place of high Viking activity. This area is fairly landlocked so the likelihood of Viking or even lost Dal Riata sailors bringing them seems remote.
http://qed.princeton.edu/index.php/User:Student/Viking,_Magyar_and_Saracen_Invasions_in_9th_and_10th_Century_Europe

It is an area that Z196 and SRY2627 might have passed through... both of those are found in Germany so I think an accompanying M222 or two is a reasonable possibility along the way towards the Pyrenees  (If Jean M's proposals are correct.) The old languages of the Midi-Pyrénées are Gaulish, I believe. At least Occitan is which is where some think Catalan came from (Catalonia is where SRY2627 is high.)


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 29, 2011, 05:48:13 PM
If at least approaching 10% of the population throughout France is L21 it does raise the question as to how it was able to infiltrate everywhere to a significant level.  It actually suggests a rather more widespread significant input in France than even the project data suggested.  The widespread nature does make origin and spread direction harder to infer.

As for the very high L21 in Rennes.  Rennes is in the part of Brittany which spoke a latinate dialect and so automatically to link this with P-Celtic speakers from Britain may be making an assumption.  As I have said before, the best evidence that NW France was already very high in L21 before the Bretons is the very high L21 in Wales, Ireland, western Scotland etc.  That high L21 would not have happened repeatedly in Atlantic British Isles if the origin populations were not also high L21. 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on August 29, 2011, 06:22:02 PM
One out of 83 is the number of M222+ they found in that sample. There were five L21xM222 guys, apparently.

I would say that singleton M222 is an outlier. You can't say much more than that when you only find one.

But these arguments apply equally to the Norwegian sample.

best
authun

Only in isolation. There are at least two M222 in the Busby Norwegian data (a tad weak, I'll admit, but as many as in the much larger French sample), and I know of another (Skaar, kit N55657) from the R-M222 Project. There are at least four L159.2 Norwegians that we know of from the R-L159.2 Project. There is another Norwegian in the R-L21 Plus Project who matches the Scots Modal.

That seems like a fair amount of British Isles-localized stuff for such a small, under-sampled country.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on August 30, 2011, 12:23:24 PM
I know not everyone appreciates Anatole Klyosov but I do. I'm not saying I agree with him and his approach all of the time, even most of the time, but he is smart I find arguments that he is involved as entertaining.

http://www.blogger.com/comment.g?blogID=7785493&postID=6454420491495013977

Quote
O.K., since the readers are silent, let me explain why the Fig. 4 from the paper ("Here is a figure from the paper, showing age estimates of sub-haplogroups R-S21 vs. R-S116") is not adequate, mildly speaking.

First, they took totally wrong mutation rates - from Ballantyne et al (2010) [from father-son transmission]. The data are based on just a few mutations, and awfully unreliable. Let me show it.

Everyone who worked with haplotypes and their mutations, knows that DYS393 is a very slow marker, and DYS390 is a fairly fast one. Indeed, Chandler's table, the most reliable one for the first 12 markers, shows the respective mutation rate constants as 0.00076 and 0.00311 (mutations per marker per generation), a 4-time difference.

Whar do we see in the Ballantyne's data? 0.00211 and 0.00152 (!) DYS393 is FASTER than DYS390 (!!). An utter nonsense. How did it happen? Very simple: Among 1758 father-son pairs Ballantyne et al observed just 3 mutations in DYS393, and 2 mutations in DYS390, and they took it (!) as a solid base for their absurd mutation rate constants.

This is applicable to all their "mutation rates". The reason is that among those almost 2000 father-son pairs, there were 3, 2, 7, 5, 3, 6, 0, 0, 6, 9, 1, 6 mutations in the first 12 matkers. It just cannot be used for mutation rate estimates.

Enough? Not quite. Just a minor thing in this context. When one works with "fast" markers, a correction for back mutations is MUCH higher. Otherwise one obtains a systematic deviation in TRCAs from slow to fast markers.

So, the conclusion number one: forget about Fig. 4 and the "principal conclusions" of the Busby et al paper. They are all wrong.

Now, I can present here data on "age estimates of sub-haplogroups R-S21 vs. R-S116", based on a much better approach. This is an important question, because it likely sets a good DNA-related time estimate for Bell Beaker movements from Iberia up North to the continental Europe.

I think, Dienekes, that since you presented here negative "side of the coin" you with your fairness would like to see its positive side. Aren't you?

Thank you.

Anatole Klyosov

Monday, August 29, 2011 3:21:00 PM

I do agree with Anatole on how Busby et al picked certain STR's that didn't match very well with what they said their intent was, and was the main thrust of their argument. I still agree with Busby that STR selection is important, not just the limited individual STR's and number of STR's they chose.

I think some have gone a bit overboard in declaring the death of STR variance and TRMCA estimations. As Mark Twain would say,
Quote
reports of my death have been greatly exaggerated.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 30, 2011, 06:06:56 PM
I know not everyone appreciates Anatole Klyosov but I do. I'm not saying I agree with him and his approach all of the time, even most of the time, but he is smart I find arguments that he is involved as entertaining.

http://www.blogger.com/comment.g?blogID=7785493&postID=6454420491495013977

Quote
O.K., since the readers are silent, let me explain why the Fig. 4 from the paper ("Here is a figure from the paper, showing age estimates of sub-haplogroups R-S21 vs. R-S116") is not adequate, mildly speaking.

First, they took totally wrong mutation rates - from Ballantyne et al (2010) [from father-son transmission]. The data are based on just a few mutations, and awfully unreliable. Let me show it.

Everyone who worked with haplotypes and their mutations, knows that DYS393 is a very slow marker, and DYS390 is a fairly fast one. Indeed, Chandler's table, the most reliable one for the first 12 markers, shows the respective mutation rate constants as 0.00076 and 0.00311 (mutations per marker per generation), a 4-time difference.

Whar do we see in the Ballantyne's data? 0.00211 and 0.00152 (!) DYS393 is FASTER than DYS390 (!!). An utter nonsense. How did it happen? Very simple: Among 1758 father-son pairs Ballantyne et al observed just 3 mutations in DYS393, and 2 mutations in DYS390, and they took it (!) as a solid base for their absurd mutation rate constants.

This is applicable to all their "mutation rates". The reason is that among those almost 2000 father-son pairs, there were 3, 2, 7, 5, 3, 6, 0, 0, 6, 9, 1, 6 mutations in the first 12 matkers. It just cannot be used for mutation rate estimates.

Enough? Not quite. Just a minor thing in this context. When one works with "fast" markers, a correction for back mutations is MUCH higher. Otherwise one obtains a systematic deviation in TRCAs from slow to fast markers.

So, the conclusion number one: forget about Fig. 4 and the "principal conclusions" of the Busby et al paper. They are all wrong.

Now, I can present here data on "age estimates of sub-haplogroups R-S21 vs. R-S116", based on a much better approach. This is an important question, because it likely sets a good DNA-related time estimate for Bell Beaker movements from Iberia up North to the continental Europe.

I think, Dienekes, that since you presented here negative "side of the coin" you with your fairness would like to see its positive side. Aren't you?

Thank you.

Anatole Klyosov

Monday, August 29, 2011 3:21:00 PM

I do agree with Anatole on how Busby et al picked certain STR's that didn't match very well with what they said their intent was, and was the main thrust of their argument. I still agree with Busby that STR selection is important, not just the limited individual STR's and number of STR's they chose.

I think some have gone a bit overboard in declaring the death of STR variance and TRMCA estimations. As Mark Twain would say,
Quote
reports of my death have been greatly exaggerated.


The exchange of posts between Anatole and Dienekes before that was...ahem...   


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on August 30, 2011, 06:08:59 PM
I know not everyone appreciates Anatole Klyosov but I do. I'm not saying I agree with him and his approach all of the time, even most of the time, but he is smart I find arguments that he is involved as entertaining.

http://www.blogger.com/comment.g?blogID=7785493&postID=6454420491495013977

Quote
O.K., since the readers are silent, let me explain why the Fig. 4 from the paper ("Here is a figure from the paper, showing age estimates of sub-haplogroups R-S21 vs. R-S116") is not adequate, mildly speaking.

First, they took totally wrong mutation rates - from Ballantyne et al (2010) [from father-son transmission]. The data are based on just a few mutations, and awfully unreliable. Let me show it.

Everyone who worked with haplotypes and their mutations, knows that DYS393 is a very slow marker, and DYS390 is a fairly fast one. Indeed, Chandler's table, the most reliable one for the first 12 markers, shows the respective mutation rate constants as 0.00076 and 0.00311 (mutations per marker per generation), a 4-time difference.

Whar do we see in the Ballantyne's data? 0.00211 and 0.00152 (!) DYS393 is FASTER than DYS390 (!!). An utter nonsense. How did it happen? Very simple: Among 1758 father-son pairs Ballantyne et al observed just 3 mutations in DYS393, and 2 mutations in DYS390, and they took it (!) as a solid base for their absurd mutation rate constants.

This is applicable to all their "mutation rates". The reason is that among those almost 2000 father-son pairs, there were 3, 2, 7, 5, 3, 6, 0, 0, 6, 9, 1, 6 mutations in the first 12 matkers. It just cannot be used for mutation rate estimates.

Enough? Not quite. Just a minor thing in this context. When one works with "fast" markers, a correction for back mutations is MUCH higher. Otherwise one obtains a systematic deviation in TRCAs from slow to fast markers.

So, the conclusion number one: forget about Fig. 4 and the "principal conclusions" of the Busby et al paper. They are all wrong.

Now, I can present here data on "age estimates of sub-haplogroups R-S21 vs. R-S116", based on a much better approach. This is an important question, because it likely sets a good DNA-related time estimate for Bell Beaker movements from Iberia up North to the continental Europe.

I think, Dienekes, that since you presented here negative "side of the coin" you with your fairness would like to see its positive side. Aren't you?

Thank you.

Anatole Klyosov

Monday, August 29, 2011 3:21:00 PM

I do agree with Anatole on how Busby et al picked certain STR's that didn't match very well with what they said their intent was, and was the main thrust of their argument. I still agree with Busby that STR selection is important, not just the limited individual STR's and number of STR's they chose.

I think some have gone a bit overboard in declaring the death of STR variance and TRMCA estimations. As Mark Twain would say,
Quote
reports of my death have been greatly exaggerated.


Its like the old Oscar Wilde quip that 'the rumours of my death have been greatly exaggerated'


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on August 30, 2011, 07:22:24 PM
Quote
reports of my death have been greatly exaggerated.


Its like the old Oscar Wilde quip that 'the rumours of my death have been greatly exaggerated'

Or Spike Milligan's

"I demand a second opinion!"
From his book, Graveyard Wit.

cheers
authun


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: A.D. on September 06, 2011, 08:33:47 PM
I was always told the northern Irish and the Norwegian Vikings were BFF's and they both hated the Danes and Dalcassians and refused to fight at the Battle of Clontarf.
I mentioned some where else that (a BBC) program had some guy saying that iceland's mt-dna was something lie 70% Irish and not slaves? Of corse 100% y-dna was Nordic.
If anyone knows what that is please tell.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on September 09, 2011, 10:31:18 AM
I was always told the northern Irish and the Norwegian Vikings were BFF's and they both hated the Danes and Dalcassians and refused to fight at the Battle of Clontarf.
I mentioned some where else that (a BBC) program had some guy saying that iceland's mt-dna was something lie 70% Irish and not slaves? Of corse 100% y-dna was Nordic.
If anyone knows what that is please tell.

A.D.
Below are some relevant links:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2613751/
http://www.ncbi.nlm.nih.gov/pubmed/11179019
http://www.ncbi.nlm.nih.gov/pubmed/10931763

"The Icelanders are one of the most studied populations in human genetics [1]–[5]. According to historical and archaeological sources, Iceland was settled roughly 1100 years ago by a mixture of people that originated primarily from Scandinavia and the British Isles [6],[7]. Studies of mtDNA variation indicate that contemporary Icelanders trace about 37% of their matrilineal ancestry to Scandinavia, with the remainder coming from the populations of Scotland and Ireland [1],[8],[9]. In contrast, Y-chromosome analyses suggest that 75–80% of their patrilineal ancestry originated in Scandinavia"

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2613751/


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: A.D. on September 09, 2011, 11:01:05 AM
Would it be likley that the vikinng slave traders were not all Norse.Dublin and York seem to be the 2 main centers which seem to be fairly cosmopolitan. e.g. the battle of clontarf the opposition to Brian Boru was a Lienster/Viking mix.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: NealtheRed on September 09, 2011, 01:05:25 PM
Would it be likley that the vikinng slave traders were not all Norse

Yes. Slavery was inherent in Irish society as well. When the Irish were not fighting the Norse (and there was not a universal Irish coalition against the Norse, as there were alliances with them), they were busy fighting each other.

This also makes me think of a combined Irish/Norse exodus of some after the Battle of Clontarf. How can one explain the high numbers of R1b in Iceland? The Faroes may be similar too.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on September 10, 2011, 03:20:28 PM
In The Celtic Realms, Dillon and Chadwick say that slavery actually wasn't all that extensive in Irish society.

Quote from: The Celtic Realms, p. 96
There were slaves (mug, m., cumal, f.), perhaps mostly captives of war, but they do not appear to have been a large element in the population.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: rms2 on September 10, 2011, 03:22:02 PM
I noticed I can't get back into the Busby paper at the link posted in the original post without paying for it.

Is there another way?


Title: P312 and U106 localized centers and middle Neolithic expansions
Post by: Mike Walsh on October 05, 2011, 10:56:36 AM
Alan has brought this proposal up several times and I see that Busby referred to a similar alternative - A mid to late Neolithic expansion for R-L11 groups from continental Western and Central Europe.

When I originally read "The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269" by Busby et al.  I passed over this because I couldn't find the source document referenced and was more focused on STR diversity/clines rather than frequency charts.

Quote from: Busby
A recent analysis of radiocarbon dates of Neolithic sites across Europe [46] reveals that the spread of the Neolithic was by no means constant, and that several ‘centres of renewed expansion’ are visible across Europe, representing  areas of colonization, three of which map intriguingly closely to the centres of the sub-haplogroups foci (electronic supplementary material, figure S3).

Figure 3 is the one that shows R-S145(L21), R-S28(U152) and R-S21(U106) frequency maps and what Busby calls localized centers or sub-haplogroup foci.

The reference to #46 is "Detection of diffusion and contact zones of early farming in Europe from the space-time distribution of 14C dates" by Bocquet-Appel et al - 2009. I don't have the full document. Has anyone looked at this?
http://www.sciencedirect.com/science/article/pii/S0305440308002719

Maybe they were on to something here? I still have concerns about how L23* or L11* would have had to slip into Europe to initiate or ride population explosions from multiple localized centers pretty much simultaneously.

When you see the different outlays of U106 (really should think of it as L48, U198, Z18, L1) and P312 (U152, L2, L21,  Z196, SRY2627 - they are all about the same age) are these the remnants of a series of separate explosions or just a single explosion where the actual center has been obliterated or obscured?


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: alan trowel hands. on October 05, 2011, 06:00:52 PM
Mike
I think your last sentence is quite possible. There were distinct sudden expansions in the middle Neolithic after a long hiatus period after LBK.  I find in particular the similarity between U106 and TRB/Funnel Beaker very interesting.  U152 could represent expansion in the Alpine area and L21 the (again similar period) expansion into the wet Atlantic areas.  The only smoking gun that ties into that period of middle Neolithic expansion was the spread of developed dairy pastoralism.  However, as I have said before, if this happened, its spread through central Europe must have been low visibility and is not represented by a handy culture with typical pot trail.  All that can be said is the old LBK zone was transformed in that period, took up dairying and expanded into new zone that were previously shunned. 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on October 10, 2011, 03:54:59 AM
I was always told the northern Irish and the Norwegian Vikings were BFF's and they both hated the Danes and Dalcassians and refused to fight at the Battle of Clontarf.
I mentioned some where else that (a BBC) program had some guy saying that iceland's mt-dna was something lie 70% Irish and not slaves? Of corse 100% y-dna was Nordic.
If anyone knows what that is please tell.


A.D.
Below are some relevant links:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2613751/
http://www.ncbi.nlm.nih.gov/pubmed/11179019
http://www.ncbi.nlm.nih.gov/pubmed/10931763

"The Icelanders are one of the most studied populations in human genetics [1]–[5]. According to historical and archaeological sources, Iceland was settled roughly 1100 years ago by a mixture of people that originated primarily from Scandinavia and the British Isles [6],[7]. Studies of mtDNA variation indicate that contemporary Icelanders trace about 37% of their matrilineal ancestry to Scandinavia, with the remainder coming from the populations of Scotland and Ireland [1],[8],[9]. In contrast, Y-chromosome analyses suggest that 75–80% of their patrilineal ancestry originated in Scandinavia"

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2613751/

'Lost' early Viking site in Co Louth one of 'most important' in world

http://www.irishtimes.com/newspaper/ireland/2011/1010/1224305518882.html

by ELAINE KEOGH

"ANNAGASSAN IN Co Louth is home to one of the world’s most important Viking sites, a local curator has claimed.

The Vikings over-wintered in two places in Ireland: one would become Dublin, the other was believed to have been lost in time. Not any more.

A year after test trenches were dug on the “virgin” site, the results of radio-carbon testing on some of the artefacts recovered have confirmed that “Linn Duachaill” exists and is perfectly preserved underneath farmland."

If this site turns out to be as expected, it could further strengthen the historical links between the Vikings and Irish and move there sphere of influence further north to the borders of Ulster.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on October 10, 2011, 06:21:54 AM
The Vikings over-wintered in two places in Ireland: one would become Dublin, the other was believed to have been lost in time. Not any more.

I'm not too sure why the statement refers to only two places. Before this particular place was discovered, the viking base at Waterford was found, again whilst road building.

http://www.ncte.ie/viking/wford.pdf


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Heber on October 11, 2011, 03:37:29 AM
The Vikings over-wintered in two places in Ireland: one would become Dublin, the other was believed to have been lost in time. Not any more.

I'm not too sure why the statement refers to only two places. Before this particular place was discovered, the viking base at Waterford was found, again whilst road building.

http://www.ncte.ie/viking/wford.pdf

I agree. As the maps on your linked document show the Vikings were raiding along the east and south Coast and the large rivers Shannon and Erne. I remember reading somewhere that some made biannual trips to the Isles from Norway, after the Spring sowing and following the late Summer harvest.
I believe the overwintering sites such as Dublin, Wexford and Waterford led to the establishment of these towns. The Louth site never developed in this way and hopefully is in a relatively undisturbed condition.

edit:
"Raiding was often a part-time occupation. Chapter 105 of Orkneyinga saga describes the habits of Sveinn Ásleifarson. In the spring, he oversaw the planting of grain on his farm at Gáreksey. When the job was done, he went off raiding in the Hebrides and Ireland, but he was back to the farm in time to take in the hay and the grain in mid-summer. Then he went off raiding again until the arrival of winter."



Title: Re: P312 and U106 localized centers and middle Neolithic expansions
Post by: IALEM on October 11, 2011, 05:55:17 AM
Alan has brought this proposal up several times and I see that Busby referred to a similar alternative - A mid to late Neolithic expansion for R-L11 groups from continental Western and Central Europe.

When I originally read "The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269" by Busby et al.  I passed over this because I couldn't find the source document referenced and was more focused on STR diversity/clines rather than frequency charts.

Quote from: Busby
A recent analysis of radiocarbon dates of Neolithic sites across Europe [46] reveals that the spread of the Neolithic was by no means constant, and that several ‘centres of renewed expansion’ are visible across Europe, representing  areas of colonization, three of which map intriguingly closely to the centres of the sub-haplogroups foci (electronic supplementary material, figure S3).

Figure 3 is the one that shows R-S145(L21), R-S28(U152) and R-S21(U106) frequency maps and what Busby calls localized centers or sub-haplogroup foci.

The reference to #46 is "Detection of diffusion and contact zones of early farming in Europe from the space-time distribution of 14C dates" by Bocquet-Appel et al - 2009. I don't have the full document. Has anyone looked at this?
http://www.sciencedirect.com/science/article/pii/S0305440308002719

Maybe they were on to something here? I still have concerns about how L23* or L11* would have had to slip into Europe to initiate or ride population explosions from multiple localized centers pretty much simultaneously.

When you see the different outlays of U106 (really should think of it as L48, U198, Z18, L1) and P312 (U152, L2, L21,  Z196, SRY2627 - they are all about the same age) are these the remnants of a series of separate explosion or just a single explosions where the actual center has been obliterated or obscured?
The demographic process detected in the works of Bocquet-Appel is that after an initial demographic expansion in the early Neolithic in Europe there is a deep recession. As this process is found in both LBK and Cardial Neolithic and in sites spread all over Europe, the cause should be in the neolithic culture itself. Bocquet-Apel thinks the cause is the spread of disiases associated to domestic animals. 


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: authun on October 11, 2011, 07:39:16 AM
The population levels are also discussed in

Stephen Shennan & Kevan Edinborough, Prehistoric population history: from the Late Glacial to the Late Neolithic in Central and Northern Europe (2007)

Stephen Shennan, Evolutionary Demography and the Climate and Population History of the European Early Neolithic (2009) http://www.humbiol.com/pdf/pdfapriljune/SHENNAN%20HB%2081020312.pdf

Shennan & Edinborough mention the zoonoses of domesticated animals as being a possible cause, in the 2009 paper however, Shennan is still unclear as to the cause but discusses warefare. Changes in climate, rainfall, temperature etc affect harvests and Europe and there was much to be learned in terms of keeping sufficient seed stocks and in keeping them in good condition. I should imagine more than one group found the following year's seed stock had gone mouldy when it came to the growing season. Prevention of infectious zoonoses from domesticated animals is of course a major problem even today. In the period that we are talking about, they had it all still to learn.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mark Jost on October 11, 2011, 09:09:08 AM
Updates to the earlier Busby et all lively discussion:

Dr. Anatole A. Klyosov has posted several new english entries in the Russian DNA Proceedings With three articles. One is the reposting of an article that was earlier published in Proceedings of the Russian Academy of DNA Genealogy, vol. 3, No. 1, 2010 (in Russian), and it was later translated to english. The second was a Anatole A. Klyosov's discussion of the overview of Busby et al (2011) article in Proc. of the Royal Soc. (B) and Dienekes Pontikos “essay. And the third article was his response to letters.


http://aklyosov.home.comcast.net/~aklyosov/
- Proceedings of the Russian Academy of DNA Genealogy, vol. 4, No. 9, 2011 (in Russian and English)

http://aklyosov.home.comcast.net/Vestnik_4_09.pdf


Haplogroup R1b as a carrier of Proto-Türkic languages, aka Dene-Caucasian
languages, aka Erbin, that is a non-IndoEuropean language in its dynamics
during 16,000 to 3,000 years before present. A.A. Klyosov . . . . . . . . . . . . . . . . 1716

DISCUSSIONS
The recent infamous (and failed) attempt to discredit the mutation rate
constants. An overview of Busby et al (2011) article in Proc. of the Royal
Soc. (B) and Dienekes Pontikos “essay” in his Anthropology Blog.
Anatole A. Klyosov. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1831

Letters from the Readers: PERSONAL CASES
Part 31, letters 101-104. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1893

I know not everyone appreciates Anatole Klyosov but I do. I'm not saying I agree with him and his approach all of the time, even most of the time, but he is smart I find arguments that he is involved as entertaining.

http://www.blogger.com/comment.g?blogID=7785493&postID=6454420491495013977

Quote
O.K., since the readers are silent, let me explain why the Fig. 4 from the paper ("Here is a figure from the paper, showing age estimates of sub-haplogroups R-S21 vs. R-S116") is not adequate, mildly speaking.

First, they took totally wrong mutation rates - from Ballantyne et al (2010) [from father-son transmission]. The data are based on just a few mutations, and awfully unreliable. Let me show it.

Everyone who worked with haplotypes and their mutations, knows that DYS393 is a very slow marker, and DYS390 is a fairly fast one. Indeed, Chandler's table, the most reliable one for the first 12 markers, shows the respective mutation rate constants as 0.00076 and 0.00311 (mutations per marker per generation), a 4-time difference.

Whar do we see in the Ballantyne's data? 0.00211 and 0.00152 (!) DYS393 is FASTER than DYS390 (!!). An utter nonsense. How did it happen? Very simple: Among 1758 father-son pairs Ballantyne et al observed just 3 mutations in DYS393, and 2 mutations in DYS390, and they took it (!) as a solid base for their absurd mutation rate constants.

This is applicable to all their "mutation rates". The reason is that among those almost 2000 father-son pairs, there were 3, 2, 7, 5, 3, 6, 0, 0, 6, 9, 1, 6 mutations in the first 12 matkers. It just cannot be used for mutation rate estimates.

Enough? Not quite. Just a minor thing in this context. When one works with "fast" markers, a correction for back mutations is MUCH higher. Otherwise one obtains a systematic deviation in TRCAs from slow to fast markers.

So, the conclusion number one: forget about Fig. 4 and the "principal conclusions" of the Busby et al paper. They are all wrong.

Now, I can present here data on "age estimates of sub-haplogroups R-S21 vs. R-S116", based on a much better approach. This is an important question, because it likely sets a good DNA-related time estimate for Bell Beaker movements from Iberia up North to the continental Europe.

I think, Dienekes, that since you presented here negative "side of the coin" you with your fairness would like to see its positive side. Aren't you?

Thank you.

Anatole Klyosov

Monday, August 29, 2011 3:21:00 PM

I do agree with Anatole on how Busby et al picked certain STR's that didn't match very well with what they said their intent was, and was the main thrust of their argument. I still agree with Busby that STR selection is important, not just the limited individual STR's and number of STR's they chose.

I think some have gone a bit overboard in declaring the death of STR variance and TRMCA estimations. As Mark Twain would say,
Quote
reports of my death have been greatly exaggerated.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 23, 2012, 10:51:48 AM
Good, finally someone who understands the various gyrations that Busby went through and will defend them.  If Busby's right, that's fine with me, but I don't "get it" yet.

I don't like how Busby jacked around with removing data from the samples.  They didn't like Barlaresque's conclusions based on R-M269 which shows Anatole (Turkey) with significantly higher variance.  They re-sample and deselect from Irish and then just plain throw away Turkey to get the outcomes they are looking for.

Let's go over to the Busby thread that Razyn pointed out if you want to debate Busby.

Quote from: Mikewww
Are you talking about the Supplementary Information where they refer to Figure S2 panel B?

Look at a little closer and read the notes.
Quote from: Busby
Figure S2: Reanalysis of Balaresque et al R-M269 Samples. The first panel, A, shows the significant correlation between longitude and mean (observed) variance found in their original dataset.The second panel, B, shows the same data, including the Irish group but without the three Turkish (TK) populations. Removing the Turkish samples can be justified on the grounds that, in our dataset c.90% of Anatolian samples were un-derived at SNP S127, whereas the majority of individuals in European populations were derived at this SNP

No, I’m referring to figure-2b in the main study, the one that shows a map with the frequency of RxS127 on the left and a graph of bootstrap variance relative to a normalized longitude on the right.
....
Well it wasn’t quite like that with the Irish, here is their explanation as to what they did:
Quote from: Busby et al(2011)
.....  However, we note that the positive correlation between longitude and variance is still present after removing only the Irish and retaining the Balaresque et al Turkish populations. If we replace the variance calculated by Balaresque et al with that calculated from our repetitions, then the correlation is no longer significant, independent of whether or not we remove the Turkish samples (Figure S2).
 


My understanding is that "bootstrapped" means "resampled" so the Figure 2 charts a, b and c on the right from the main body of the paper are based on Busby's resampling, not on "observed" variance.

The Supplementary Information is intended to support the primary body of the paper and show the details. The Supplementary Information Figure S2 panels A through E go through the gyrations of the actual observed variance calculations.  

It is critical to Busby that he handles the Turkish and particularly the Irish samples differently to get what I guess are the appropriate outcomes.

Supplementrary S2 ( http://rspb.royalsocietypublishing.org/content/early/2011/08/18/rspb.2011.1044/suppl/DC1 ) Panel A is just a straight look at "observed variance" from what the data is. This is Barlaresque's view and is the straw-man proposal that Busby appears to be after. The Irish sample shows the lowest variance and the three Turkish samples show the highest (with most of the continent in between) - resulting in the clear east/west cline.

Panel B is where the Turkish samples are thrown out because "Removing the Turkish samples can be justified on the grounds that, in our dataset c.90% of Anatolian samples were un-derived at SNP S127."  The east/cline diminishes signficantly but is still there slightly because of the low diversity in the Irish sample.   I don't see how it is valid to throw out the Turkish sample because it contains a lot of L11(S127).  Do multiple analyses at multiple levels of the phylogeny if that is what you want, but why throw out a location altogether?

Panel C throws out the Irish sample but leaves the Turkish in. The east/west cline is present even without the low diversity Irish sample.

Panel D introduces something new.  This is where the "bootstrapping" comes in. The observed Irish variance is now replaced by the resampled variance.  The Irish diversity is now higher than all of the European samples and is now as high as the highest Turkish sample. There is still a slight east/west cline, but apparently this is deemed as "insignificant".  I'm okay with that determination, I just don't trust the resampling.

Quote from: Busby et al(2011)
Testing the variance calculations from the Irish population
... We note, however, that 17-STR haplotypes, including the 9 STRs used in Balaresque et al’s analysis, are available for 681 Irish R-M269 derived individuals in Moore et al (3), which is, in fact, the study which Balaresque et al use to estimate R-M269 frequency in Ireland. A subset of the Moore et al samples were re-analysed in the current study for SNPs downstream of R-M269, and the original haplotype data are used here to calculate variance.[/b] To test if the Ysearch haplotypes were representative of the Irish R-M269 in Moore et al, we independently re-sampled the Moore et al dataset 10,000 times, selecting sub-samples of 75 haplotypes from which we estimated the variance using the same 9 STRs used in the Balaresque et al paper. The median variance of these 10,000 repetitions was 0.354 with a 95% CI of (0.285-0.432). The lowest variance value out of the 10,000 samples was 0.242, which is still higher than the figure observed in the Balaresque et al Ysearch sample (0.208). We therefore believe our estimate of Irish R-M269 variance to be a more robust representation of the true variance than that estimated by Balaresque et al. However, we note that the positive correlation between longitude and variance is still present after removing only the Irish and retaining the Balaresque et al Turkish populations. If we replace the variance calculated by Balaresque et al with that calculated from our repetitions, then the correlation is no longer significant, independent of whether or not we remove the Turkish samples (Figure S2).

Even though there are thousands of high resolution SNP Irish samples available, Busby resamples 75 at a time for only 9 STRs.  He uses Moore's data.  I don't understand why they had to use the Moore "A Y-Chromosome Signature of Hegemony in Gaelic Ireland," 2006.  L11(S127) wasn't even tested for. Even if they don't like Ysearch, YHRD or whatever, there is plenty of deeply tested data from Ireland.

...anyway, after the "bootstrapping", Busby's "robust" representation of Irish R-M269 now zooms the Irish to the top of STR variance, jumping by Britain and Continental Europe and pulling even with the highest Turkish sample.  Does this sound right?  Should the hypothesis be that two different forms of R-M269 survived a bottleneck and exploded to take over Europe from both directions - one from Ireland and the other from Turkey?

I just don't like comparing data across regions with different treatments per region.  If "robust  bootstrapping" is good for Ireland then it should be done for continent and Turkey too.

As noted above, Busyby states this condition on his analysis.
Quote from: Busby
variance calculated by Balaresque et al with that calculated from our repetitions

However, this particular quote didn't complete the condition.  The condition is he replaced the variance calculations for Ireland only, and with a different data set to boot.  I'm not a brilliant statistician but I know that apples and oranges don't add up.

um.... I fear this is a problem with almost any of these Pan-European studies. They are piecing together data from multiple sources, but I hope that at least they will be consistent within their respective studies in their treatment of the data.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanL on April 23, 2012, 11:41:05 AM
Good, finally someone who understands the various gyrations that Busby went through and will defend them.  If Busby's right, that's fine with me, but I don't "get it" yet.
My understanding is that "bootstrapped" means "resampled" so the Figure 2 charts a, b and c on the right from the main body of the paper are based on Busby's resampling, not on "observed" variance.

It is bootstrapped because it includes the dataset from Myres et al(2010) plus new populations included in the new study. It is still random datasets and not chosen datasets to fit some purpose. So yes, this is observed variance in a set of populations from the Myres et al(2010) dataset, and the new paper. Refer to table-S1 in the supplementary excel file for a list of populations used and the sources of the study that first sampled the datasets.  Again, we are not talking about sampling at convenience but about amplifying the Myres et al(2010) dataset with new populations.


The Supplementary Information is intended to support the primary body of the paper and show the details. The Supplementary Information Figure S2 panels A through E go through the gyrations of the actual observed variance calculations. 

It is critical to Busby that he handles the Turkish and particularly the Irish samples differently to get what I guess are the appropriate outcomes.

Supplementrary S2 ( http://rspb.royalsocietypublishing.org/content/early/2011/08/18/rspb.2011.1044/suppl/DC1 ) Panel A is just a straight look at "observed variance" from what the data is. This is Barlaresque's view and is the straw-man proposal that Busby appears to be after. The Irish sample shows the lowest variance and the three Turkish samples show the highest (with most of the continent in between) - resulting in the clear east/west cline.

Panel B is where the Turkish samples are thrown out because "Removing the Turkish samples can be justified on the grounds that, in our dataset c.90% of Anatolian samples were un-derived at SNP S127."  The east/cline diminishes signficantly but is still there slightly because of the low diversity in the Irish sample.   I don't see how it is valid to throw out the Turkish sample because it contains a lot of L11(S127).  Do multiple analyses at multiple levels of the phylogeny if that is what you want, but why throw out a location altogether?

Panel C throws out the Irish sample but leaves the Turkish in. The east/west cline is present even without the low diversity Irish sample.

Panel D introduces something new.  This is where the "bootstrapping" comes in. The observed Irish variance is now replaced by the resampled variance.  The Irish diversity is now higher than all of the European samples and is now as high as the highest Turkish sample. There is still a slight east/west cline, but apparently this is deemed as "insignificant".  I'm okay with that determination, I just don't trust the resampling.

Quote from: Busby et al(2011)
Testing the variance calculations from the Irish population
... We note, however, that 17-STR haplotypes, including the 9 STRs used in Balaresque et al’s analysis, are available for 681 Irish R-M269 derived individuals in Moore et al (3), which is, in fact, the study which Balaresque et al use to estimate R-M269 frequency in Ireland. A subset of the Moore et al samples were re-analysed in the current study for SNPs downstream of R-M269, and the original haplotype data are used here to calculate variance.[/b] To test if the Ysearch haplotypes were representative of the Irish R-M269 in Moore et al, we independently re-sampled the Moore et al dataset 10,000 times, selecting sub-samples of 75 haplotypes from which we estimated the variance using the same 9 STRs used in the Balaresque et al paper. The median variance of these 10,000 repetitions was 0.354 with a 95% CI of (0.285-0.432). The lowest variance value out of the 10,000 samples was 0.242, which is still higher than the figure observed in the Balaresque et al Ysearch sample (0.208). We therefore believe our estimate of Irish R-M269 variance to be a more robust representation of the true variance than that estimated by Balaresque et al. However, we note that the positive correlation between longitude and variance is still present after removing only the Irish and retaining the Balaresque et al Turkish populations. If we replace the variance calculated by Balaresque et al with that calculated from our repetitions, then the correlation is no longer significant, independent of whether or not we remove the Turkish samples (Figure S2).

Even though there are thousands of high resolution SNP Irish samples available, Busby resamples 75 at a time for only 9 STRs.  He uses Moore's data.  I don't understand why they had to use the Moore "A Y-Chromosome Signature of Hegemony in Gaelic Ireland," 2006.  L11(S127) wasn't even tested for. Even if they don't like Ysearch, YHRD or whatever, there is plenty of deeply tested data from Ireland.

...anyway, after the "bootstrapping", Busby's "robust" representation of Irish R-M269 now zooms the Irish to the top of STR variance, jumping by Britain and Continental Europe and pulling even with the highest Turkish sample.  Does this sound right?  Should the hypothesis be that two different forms of R-M269 survived a bottleneck and exploded to take over Europe from both directions - one from Ireland and the other from Turkey?

I just don't like comparing data across regions with different treatments per region.  If "robust  bootstrapping" is good for Ireland then it should be done for continent and Turkey too.

As noted above, Busyby states this condition on his analysis.
Quote from: Busby
variance calculated by Balaresque et al with that calculated from our repetitions

However, this particular quote didn't complete the condition.  The condition is he replaced the variance calculations for Ireland only, and with a different data set to boot.  I'm not a brilliant statistician but I know that apples and oranges don't add up.

um.... I fear this is a problem with almost any of these Pan-European studies. They are piecing together data from multiple sources, but I hope that at least they will be consistent within their respective studies in their treatment of the data.

Let me give you a quote from the Balaresque et al(2010) study, it is located under Materials and Methods:

Quote from: Balaresque et al(2010)

A total of 2,574 DNA samples from European males, assigned to populations based on two generations of residence, were typed for the SNP M269 [17], defining hgR1b1b2. Following PCR amplification using the primers 59- TAAAGATCAGAGTATCTCCCTTTG-39 and 59-ATTTCTAGGAGTTCACTGTATTAC-39, the T to C transition was analysed by digestion with BstNI, which cleavesM269- C-allele chromosomes only. Samples fromthe Iberian peninsula were typed using the SNaPshot (ABI) procedure [31]. Haplotype data were obtained for up to 20 Y-specific microsatellites [32,33]. Data from the Ysearch database (http://www.ysearch.org) for Germany (GE) and Ireland (IR) were added, together with published data for Turkey, subdivided into East, West, and Central subpopulations based on published sampling information [14]. To avoid a bias from very large samples of hgR1b1b2 (GE and IR), these were randomly subsampled to give sample sizes of 75. This allowed a comparison of nine-locus haplotypes (DYS19, DYS388, DYS389I, DYS389II, DYS390, DYS391, DYS392, DYS393, and DYS439) for 849 hgR1b1b2 chromosomes, subdivided into 23 populations. Greek and Serbian samples were too small for population-based analyses, but were included in Network analysis.


So the only reason why Busby sampled sets of 75 random haplotypes at a time was because Balaresque et al(2010) team did the exact same thing, and the only reason why they did not break down the resolution of R1b1b2 on the Balaresque et al(2010) study, was because the Balaresque et al(2010) study did not do it. So what the Busby et al. team basically did on the Supplementary file was to show that even using the erroneous methodology of the Balaresque et al. study, even without increasing the resolution of the sub-clades of R1b1b2 there isn’t any East-West gradient as the Balaresque et al. team said. Now the only reason why they didn’t do it for Turkey, or for any other European country, is because Balaresque et al. didn’t choose 75 random samples for Turkey, or any other European country.


This in fact goes to show what I said before, you guys so gladly took in the Balaresque et al. study and conclusions, that you didn't even question their sampling methodology. While you harshly criticized  the Busby et al(2011) team for what you consider to be bias, you didn’t even realized that the Balaresque et al(2010) team did the exact same thing, with the only difference being that not once did the Busby et al(2011) team got such a low variance for the Irish sample when extracting 75 haplotypes at random 10000 times. This actually says a lot about the Balaresque et al(2010) study, it says that they only took 75 haplotypes at random once, and didn’t even bother to run a few hundred sampling scenarios to see if the observed variance fell within the average variance.

All I can say is that 10000 random samples are far better than a single random sample.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Arch Y. on April 24, 2012, 11:26:20 PM
Would it be likley that the vikinng slave traders were not all Norse

Yes. Slavery was inherent in Irish society as well. When the Irish were not fighting the Norse (and there was not a universal Irish coalition against the Norse, as there were alliances with them), they were busy fighting each other.

This also makes me think of a combined Irish/Norse exodus of some after the Battle of Clontarf. How can one explain the high numbers of R1b in Iceland? The Faroes may be similar too.

The high numbers of R1b in Iceland always perplexed me and of course when I mentioned it, some of the HG I guys stated that HG I is the most predominant (obviously). Even though there is a ton of data on Iceland for a "Celtic" genetic connection, however, the R1b numbers were discounted. Also, I always wondered about the sagas and how they originated. Perhaps it took a literate Celt to teach an intelligent barbarian how to write.

Arch

Arch


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 12:25:18 AM
Good, finally someone who understands the various gyrations that Busby went through and will defend them.  If Busby's right, that's fine with me, but I don't "get it" yet.
My understanding is that "bootstrapped" means "resampled" so the Figure 2 charts a, b and c on the right from the main body of the paper are based on Busby's resampling, not on "observed" variance.

It is bootstrapped because it includes the dataset from Myres et al(2010) plus new populations included in the new study. It is still random datasets and not chosen datasets to fit some purpose. So yes, this is observed variance in a set of populations from the Myres et al(2010) dataset, and the new paper. Refer to table-S1 in the supplementary excel file for a list of populations used and the sources of the study that first sampled the datasets.  Again, we are not talking about sampling at convenience but about amplifying the Myres et al(2010) dataset with new populations.
We don't know what the intent of the authors was in their "robust re-analysis" was. I'm not sure what subjective terms like "convenience" or "amplifying" mean related to a scientific study. They played with throwing the Irish out, throwing the Turkish out and then resampling the Irish.

I understand that bootstrapping relates to resampling, but does it seem right to you to only do resampling on the Irish and then compare them with other geographies that weren't resampled?

Does it seem right to you that the Irish "resampled" or bootstrapped variance was suddently higher than everything except the highest Turkish sample? I'm not asking you to make any assumptions about the Turkish sample but what about the continental samples in between?  Are you and Busby saying that R-L11 originated in Ireland?  That's okay, but I don't think they are saying that either.  All I gather is they are saying Balaresque is wrong. That's why I call their approach a "strawman" attack. They don't really have a strong counter-hypothesis, just a strong attack against the opponents strawman that they choose.

Of course, they also messed around with STR selections to find ones that were "fit" although they weren't really through the Neolithic period, according to their own standards.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 12:39:16 AM
...
This in fact goes to show what I said before, you guys so gladly took in the Balaresque et al. study and conclusions, that you didn't even question their sampling methodology. While you harshly criticized  the Busby et al(2011) team for what you consider to be bias, you didn’t even realized that the Balaresque et al(2010) team did the exact same thing, with the only difference being that not once did the Busby et al(2011) team got such a low variance for the Irish sample when extracting 75 haplotypes at random 10000 times. This actually says a lot about the Balaresque et al(2010) study, it says that they only took 75 haplotypes at random once, and didn’t even bother to run a few hundred sampling scenarios to see if the observed variance fell within the average variance.

All I can say is that 10000 random samples are far better than a single random sample.
To be clear, as I've said before, I think the Balaresque study has errors too and is not comprehensive. I know I've said in the past their sampling is not a true cross-sectional representative sample.  I know I've said their STR lengths were too limited.  Perhaps you forgot or never saw those kinds of comments.  I think that Balaresque's study includes serious errors.  Is that harsh enough?

I am not sure how we measure harshness or gladness but perhaps my attitude towards Busby is harsh because their approach does not include a strong hypothesis but just a serious of attacks (not necessarily coordinated) to "move the ball back down the field" on the momentum started by Myres and Balarseque. I think they even said as much in a press interview.

I think the Barlaresque, Myres and Busby studies should be viewed in toto, versus just individual elements. 

I am very glad of all three studies. The more spotlight on R1b, the better.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 12:41:46 AM
....
All I can say is that 10000 random samples are far better than a single random sample.

I think we both agree that studies done to date are inadequate.

... but what about STR diversity, do you see a usefulness for it?  We've talked about it and how much effort Balaresque and Busby have put it into it.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanL on April 25, 2012, 09:13:42 AM
We don't know what the intent of the authors was in their "robust re-analysis" was. I'm not sure what subjective terms like "convenience" or "amplifying" mean related to a scientific study. They played with throwing the Irish out, throwing the Turkish out and then resampling the Irish.

I understand that bootstrapping relates to resampling, but does it seem right to you to only do resampling on the Irish and then compare them with other geographies that weren't resampled?
Does it seem right to you that the Irish "resampled" or bootstrapped variance was suddently higher than everything except the highest Turkish sample? I'm not asking you to make any assumptions about the Turkish sample but what about the continental samples in between?  Are you and Busby saying that R-L11 originated in Ireland?  That's okay, but I don't think they are saying that either.  All I gather is they are saying Balaresque is wrong. That's why I call their approach a "strawman" attack. They don't really have a strong counter-hypothesis, just a strong attack against the opponents strawman that they choose.

Of course, they also messed around with STR selections to find ones that were "fit" although they weren't really through the Neolithic period, according to their own standards.

Ok it seems you have some misunderstanding about the Busby et al(2011) study. First of all, there are three different analyses carried out throughout the study, you are mixing them all together either by accident or purposely to fit some point. Analysis-1 was done on the main study, and it was using the Myres et al(2010) dataset, and a newly collected dataset by Busby et al(2011) to analyze the bootstrapped variance of the new set which was made up of the old sets. Now if you want to question the legitimacy of the sampling techniques feel free to expose your reasons. I on the other hand having looked at table-S1 see no reason to question that the new populations would somehow drive the variance  up or down, they are as random as the Myres et al(2010) dataset, and the combined set would still be  as random. During this Analysis-1 there was no throwing out the Turkish, or re-sampling the Irish, and it was during this analysis that it was shown that the  greatest variance of RxS127 occurs in Central Europe, and that there is no East-West variance gradient for R-S127+, refer to figure-2b) and figure-2c) in the main study, not the supplementary file.

The second analysis was to test the effect of microsatellite choice on age estimates, and this again has nothing to do with analysis-1 or analysis-3, and if one looks at figure-4, the results are obvious.

The third analysis which was done on the supplementary text-S1 has nothing to do with analyses1&2, and it consisted in a series of steps. First they removed the Turkish samples on the basis that 90% of Turkish samples from Myres et al(2010) were RxS127, which they extrapolated to the Balaresque et al(2010) dataset, and assumed the Turkish samples would be mostly RxS127 as well. When they removed the Turkish sample, the East-West gradient disappeared. Now the next thing they did was to using the same approach that the Balaresque et al(2010) team took, they sample randomdly 75 Irish and calculated the variance, but unlike Balaresque et al(2010) who did it once,  they did it 10000, and not once did they find such low variance like Balaresque et al. did. So of course, given that they randomly sampled the set 10000, it is obvious that their variance, which was the mean variance of the randomly sampled sets was way more robust than that of Balaresque et al. Irish sample.

Does it seem right to you that the Irish "resampled" or bootstrapped variance was suddently higher than everything except the highest Turkish sample? I'm not asking you to make any assumptions about the Turkish sample but what about the continental samples in between?  Are you and Busby saying that R-L11 originated in Ireland?  That's okay, but I don't think they are saying that either.  All I gather is they are saying Balaresque is wrong. That's why I call their approach a "strawman" attack. They don't really have a strong counter-hypothesis, just a strong attack against the opponents strawman that they choose.

Of course, they also messed around with STR selections to find ones that were "fit" although they weren't really through the Neolithic period, according to their own standards.

Of course it seems right that the Irish “resampled” variance is higher than it was on the Balaresque et al. sample, in fact, is not a question whether it should be higher or lower, it is a question whether it is more accurate  or not? My answer: yes it is by far more accurate as they actually took the time to resample 10000 random sets. Again, there is no reason to re-sample the continental samples, because none of them ,except for the Germans were sample using the methodology of choosing 75 random haplotypes and then calculating the variance. I’m getting the sense that by presuming that Busby et al(2011) would be implying that R-L11 originated in Ireland, you sir are engaging in reductio ad absurdum.  First and foremost, in his primary analyses nowhere does it show that the variance of R-S127 is higher in Ireland, so no reason to think that. Secondly if you presume that the analysis done on the supplementary text-S1 can be used as a clear cut indicative of origin based on variance, then it would mean that R1b1b2 in general not R-L11 has peaks of variance on both Turkey and Ireland. Since all you gather is that all they are saying is that Balaresque et al. must be wrong, and completely missed the first and second analysis done on the main study, where the Balaresque et al. team isn’t even touched, is not surprised that you have done the exact same thing that you claim the Busby et al. team did a “strawman” attack. You don’t really have a strong counter-argument to analysis-1 and analysis-2 so you decide to attack analysis-3 in the supplementary text S1, then based upon the misinterpretation of the analysis-3, you engaged in reduction ad absurdum, and try to generalize your conclusions about analysis-3 two the whole study. Because if you look at it, you will see that even if the analysis they did on the Balaresque et al. dataset on text S1 were wrong, that wouldn’t make their analysis on the Myres et al.+Busby et al. combined sets wrong, nor would it invalidate the findings of the appreciable effect of microsatellite choice on age estimates. So yes, there is a strawman attack here, but is not coming from the Busby et al. team towards the Balaresque et al. study, but from you to the new study.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 09:39:11 AM
Ok it seems you have some misunderstanding about the Busby et al(2011) study. First of all, there are three different analyses carried out throughout the study, you are mixing them all together either by accident or purposely to fit some point.
I admit I am confused by their analyses and I actually do appreciate you really reading through these studies. You are the first person to really discuss the details of their numbers so I appreciate that. Thank you.

As far any point I have on this, at least at this time, we just got here (conversation-wise) because on some other threads, STR diversity has been attacked so I just defend that which usually gets into counter-arguments based on Busby's work.  Let me repeat, I think the Busby study was useful.

Let me go back and reread all three studies to look at their data sampling, etc.

Quote from: JeanL link=topic=10031.msg129923#msg129923
...So yes, there is a strawman attack here, but is not coming from the Busby et al. team towards the Balaresque et al. study, but from you to the new study.
I agree that I'm attacking some of the holes in their study. I'm also not writing a scientific paper.

I agree there really are holes in the Myres and Barlaresque studies and Busby is picking on them.  

The reason why I think the Busby approach is a strawman approach is that they are doing various analyses that aren't consistent with each other to make their counter-points, but do not really to construct an alternative hypothesis (at least a comprehensive one.) It seems Busby's focus is on Barlaresque's proposals and not much on Myres', actual age estimates, or the whole concept by Wells of a Central Asia origin for P, Q and R.  To me that is not looking at the whole picture which is also why I consider their attack as a strawman approach.

I'm actually not as anti-Busby as you think.  I don't know if you noticed, but I've never disagreed with Busyb that R1b-S127(L11) has insignificant differences across STR variance across Europe.  I've agreed that is a fair point... just that my interpretation is that L11 spread across Europe quickly.

I've also agree that STR evaluation is useful.  I just think that using limited numbers like 10 or 15 is not enough.  That's what I see when I do my own comparisons on hundreds of long haplotypes anyway.  I also think Busby's application of STRs does not match their own linear duration standards. That is an attack, but perhaps I just don't understand. Can you explain?

I've attacked the points I don't understand to see how they are defended, and to see if I can be convinced.  That's really all that I'm doing.   I'm fine with R-L11 having a Central European origin if that is the ultimate conclusion. R-P312's observed variance is highest in SE France and U106's is highest along Poland and Baltic States. Central Europe definitely is a possibility for L11.  I'm not sure about R-L23* though.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 11:18:33 AM
Does it seem right to you that the Irish "resampled" or bootstrapped variance was suddently higher than everything except the highest Turkish sample? I'm not asking you to make any assumptions about the Turkish sample but what about the continental samples in between?  Are you and Busby saying that R-L11 originated in Ireland?  That's okay, but I don't think they are saying that either. ....
...  I’m getting the sense that by presuming that Busby et al(2011) would be implying that R-L11 originated in Ireland, you sir are engaging in reductio ad absurdum.

I didn't presume that Busby is saying R-L11 originated in Ireland. I asked that question. I'm just pointing out that the treatment of different data by different methods by Busby (and by Myres or whoever) leads to unusual results.  

BTW, a reductio ad absurdum argument is not necessarily a logical fallacy. It's only a logical fallacy "Where such an argument is premised on a false dichotomy, the ostensible proof is a logical fallacy." (Wikipedia)  ... so such an argument is only a bad thing if STR diversity does not have a relationship with time (that would be the false dichotomy if that is the case but we can see that both Barlaresque and Busby do a great deal of analysis based on the hypothesis that STR diversity is related to time.)


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanL on April 25, 2012, 11:25:57 AM
I've also agree that STR evaluation is useful.  I just think that using limited numbers like 10 or 15 is not enough.  That's what I see when I do my own comparisons on hundreds of long haplotypes anyway.  I also think Busby's application of STRs does not match their own linear duration standards. That is an attack, but perhaps I just don't understand. Can you explain?

You are right; they showed that there is a significant effect of microsatellite choice in age estimates that they should have used that finding when calculating TMRCA of R-S127 haplogroup which is on figure-4a. However, in figure-2 they did not calculate TMRCA in generations, but explored the bootstrapped variance, and in fact they do not seem to think that variance is affected by choice of STR, which is why they used 10 STRs on figure-2.  In a nutshell they showed that microsatellite choice can have an effect on age estimates, but still used a combined set of 10 STRs to explore variance.  Perhaps they think one should choose the STRs when calculating TMRCA based on similarity on mutations rates and the presumed time span for common ancestry, i.e. use the average mut/marker for the slowest or fastest STRs depending on the presumed TMRCA, but not the average mut/marker for the whole set, but if you want to calculate variance use the combined set of STRs.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: razyn on April 25, 2012, 03:20:30 PM
You are right; they showed that there is a significant effect of microsatellite choice in age estimates that they should have used that finding when calculating TMRCA of R-S127 haplogroup which is on figure-4a. However, in figure-2 they did not calculate TMRCA in generations, but explored the bootstrapped variance, and in fact they do not seem to think that variance is affected by choice of STR, which is why they used 10 STRs on figure-2.  In a nutshell they showed that microsatellite choice can have an effect on age estimates, but still used a combined set of 10 STRs to explore variance.  Perhaps they think one should choose the STRs when calculating TMRCA based on similarity on mutations rates and the presumed time span for common ancestry, i.e. use the average mut/marker for the slowest or fastest STRs depending on the presumed TMRCA, but not the average mut/marker for the whole set, but if you want to calculate variance use the combined set of STRs.

Perhaps Busby wrote part 1, et wrote part 2, and Al wrote the summary.  Or some other committee-based procedure.  Each person who took part in the study may have had a clear methodology -- and followed it, as far as he or she went.  It has multiple authors, not to mention multiple sources of the sampled DNA; if there are faults, they probably aren't all Busby's.

Anyway that's how it appeared to me last summer, when I read it... and agreed with Mike, at the time, that the paper as a whole seemed internally disjunct.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 25, 2012, 03:36:44 PM
You are right; they showed that there is a significant effect of microsatellite choice in age estimates that they should have used that finding when calculating TMRCA of R-S127 haplogroup which is on figure-4a. However, in figure-2 they did not calculate TMRCA in generations, but explored the bootstrapped variance, and in fact they do not seem to think that variance is affected by choice of STR, which is why they used 10 STRs on figure-2.  In a nutshell they showed that microsatellite choice can have an effect on age estimates, but still used a combined set of 10 STRs to explore variance.  Perhaps they think one should choose the STRs when calculating TMRCA based on similarity on mutations rates and the presumed time span for common ancestry, i.e. use the average mut/marker for the slowest or fastest STRs depending on the presumed TMRCA, but not the average mut/marker for the whole set, but if you want to calculate variance use the combined set of STRs.

Perhaps Busby wrote part 1, et wrote part 2, and Al wrote the summary.  Or some other committee-based procedure.  Each person who took part in the study may have had a clear methodology -- and followed it, as far as he or she went.  It has multiple authors, not to mention multiple sources of the sampled DNA; if there are faults, they probably aren't all Busby's.

Anyway that's how it appeared to me last summer, when I read it... and agreed with Mike, at the time, that the paper as a whole seemed internally disjunct.

Yes, but I'm not sure Myres and Balaresque are all that clearer. I'm trying to understand that better, but at least I'm able to repeat some of their calculations. In the Busby paper, I can't get the same results for some of these things.  Unless I'm missing something, I think Busby (and Myres) should include their "resampling" runs as the supplementary information. Maybe I'm just missing it though.  I sure as heck don't feel comfortable with data collected with different methods. I have some renewed admiration for the National Genographic Project.... but I guess the difference is they have money.


Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: JeanL on April 25, 2012, 04:23:40 PM

In the Busby paper, I can't get the same results for some of these things.  Unless I'm missing something, I think Busby (and Myres) should include their "resampling" runs as the supplementary information. Maybe I'm just missing it though. 

The populations that Busby et al. used from Myres have their 10 STRs listed on Table-S2, it has the population codes and haplogroups. The new populations that Busby sampled are on Table-S3, it has the repeat value for 15 STRs, the population codes, and the haplogroups. Though they did a very poor job when it comes to haplogroups, they only identify R-M269, S116 and S21, for example there are 2-French Basques who are R-M269(xS127) but we don't know if the other people that are listed there as R-M269 are all R-M269(xS127) or if there are some R-S127(xS116,S21). Feel free to analyze it.

I know I can duplicate some of the values provided by Myres et al. in the supplementary tables, I don't get the exact same values of variance, but something close to it. For example Myres et al.(2010) got that the variance for R-L23(xM412) on Turkey was 0.277, I repeated the same procedure using my modal calculator, and I got the variance for R-L23(xM412) for Turkey to be 0.2828. Their variance for the Caucasus R-L23(xM412) was 0.292, mine was 0.3063.



Title: Re: The peopling of Europe and the cautionary tale of Y chromosome lineage R-M269 (B
Post by: Mike Walsh on April 26, 2012, 01:51:44 PM

In the Busby paper, I can't get the same results for some of these things.  Unless I'm missing something, I think Busby (and Myres) should include their "resampling" runs as the supplementary information. Maybe I'm just missing it though. 

The populations that Busby et al. used from Myres have their 10 STRs listed on Table-S2, it has the population codes and haplogroups. The new populations that Busby sampled are on Table-S3, it has the repeat value for 15 STRs, the population codes, and the haplogroups. Though they did a very poor job when it comes to haplogroups, they only identify R-M269, S116 and S21, for example there are 2-French Basques who are R-M269(xS127) but we don't know if the other people that are listed there as R-M269 are all R-M269(xS127) or if there are some R-S127(xS116,S21). Feel free to analyze it.

I know I can duplicate some of the values provided by Myres et al. in the supplementary tables, I don't get the exact same values of variance, but something close to it. For example Myres et al.(2010) got that the variance for R-L23(xM412) on Turkey was 0.277, I repeated the same procedure using my modal calculator, and I got the variance for R-L23(xM412) for Turkey to be 0.2828. Their variance for the Caucasus R-L23(xM412) was 0.292, mine was 0.3063.

Thanks for looking at the data. I posted this elsewhere today but it belongs here.

.. I did go back to the Myres and Barlaresque papers and reread where they got their data from and how they put it together.  I can repeat the variance calculations, etc., but I can't at all defend how they got their data from different sources.  It makes me nervous.
I should add that I can't defend that it is representative. I don't think anyone say we have truly representative data.