World Families Forums - so what does the new study mean for L21?

Welcome, Guest. Please login or register.
July 14, 2014, 06:31:44 AM
Home Help Search Login Register

+  World Families Forums
|-+  General Forums - Note: You must Be Logged In to post. Anyone can browse.
| |-+  R1b General (Moderator: rms2)
| | |-+  so what does the new study mean for L21?
« previous next »
Pages: 1 2 [3] 4 5 6 Go Down Print
Author Topic: so what does the new study mean for L21?  (Read 12591 times)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #50 on: September 01, 2010, 04:23:15 PM »

There was large scale migration from the Celtic areas of the Isles into England during the Industrial Revolution:

"After 1840, emigration had become a massive, relentless, and efficiently managed national enterprise. Counting those who went to Britain, between 9 and 10 million Irish men, women, and children emigrated after 1700. The total flow was more than the population at its historical peak in the 1830s of 8.5 million. From 1830 to 1914, almost 5 million went to the United States alone. In 1890 two of every five Irish-born people were living abroad."

"The 2001 UK Census states 869,093 people born in Ireland as living in Great Britain, with over 10% of the population (over 6 million) being of Irish descent."

"10% of the British population has one Irish grandparent."

http://en.wikipedia.org/wiki/Irish_diaspora#Britain
I don't know if you noticed, but the DNA project information I am using is based on MDKA's, mostly of a generation or two or three back.  Some of the very recent numbers (i.e. 2001) you are using are impacted by the current mobility of society as well as changes in the nature of the United Kingdom.

Please also note (as I've already posted) that I checked the GD's of the English M222s.  Several did not have matches in Ireland in the last 1000 years or so, almost all did not seem to have matches in the last 200 years (which includes the Irish Diaspora).  That doesn't mean these guys aren't Irish transplants.  They very well could be.  I'm just saying the origin of M222 is an open question from a genetic data point of view.
« Last Edit: September 01, 2010, 04:24:10 PM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #51 on: September 01, 2010, 04:28:40 PM »

.... When I got my positive result for L-21 inmediately some people in the DNA forums tried to adjudicate me an Irish/British origin...
Yes, it seems some people need L21 to be Britano-Irish.  No doubt, much of it is, but that doesn't mean it's origin is there.  I stepped on a few toes over there pretty hard on the issue this summer.  The thread got locked and some of those going over the deep end seemed to go into hiding.... but I had some inconsistencies well pointed out and was disappointed in the logic of some of the counter-arguments.
Logged

R1b-L21>L513(DF1)>L705.2
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #52 on: September 01, 2010, 05:01:09 PM »

... Can you elaborate a little more on what you said recently about M222 having much more variance in England than Ireland and give an indication of relatively how much more.
Alan,
I tried to answer in this post.  What I did is calculate variance for English and Irish M222 at 25 and at 67 (actually just the 50 non multi-copy) markers.  I also calculated diversity for the same.  The numbers for England came up higher in the ranges I list.
....  The actual calculations indicate English R-M222 ranging from 25% to 70% higher variance than Irish R-M222. ...

I take it that the data for M222 in the continent is far too small to make variance observations.
I don't see the detail of the haplotypes in Germany or France from the Myres study.  If anyone sees it somewhere, let me know.  As far as our DNA projects, we only have the one M222 from Germany and the one from Norway.

I have certainly noticed in the past that variance is always low in a tiny sample and get bigger as the sample improves.  
That may be, but I don't know if there is any logic to that other than luck. The count of the English MDKA M222's from the projects is ten. To me that is a small sample so I wouldn't bet too much on it.  Still, the Myres study confirms some significant presence of M222 in France and in Germany so it is worth investigating.

« Last Edit: September 01, 2010, 05:34:13 PM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
argiedude
Old Hand
****
Offline Offline

Posts: 146


« Reply #53 on: September 01, 2010, 06:28:28 PM »

Does anyone have that y-dna study of France (Santiago de Compostela)?

Here it is:

http://tinyurl.com/2btr8va

I meant if someone can share the study.

I think I just read it online. I don't have it on my computer.

Argiedude I found it in the French L21 thread.

http://download.journals.elsevierhealth.com/pdfs/journals/1875-1768/PIIS1875176809002340.pdf

A couple of us emailed the authors to see if they could give us more data or even just % breakdowns of the clades but we never got a reply.


I originally asked because Alan had stated the following:

"I do also wonder again about the figures.  Hobbiest testing has found a lot lot less continental M222 and  think the Santiago de Compostella University paper on French y-DNA only found instances in Paris (hardly surprising) and Tolouse."

I got the impression someone had obtained the haplogroup results by region (or just the results, period!).
Logged

y-dna: R1b L21
mtdna: U5
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #54 on: September 01, 2010, 08:32:55 PM »

. . .
Myres does show on Table S4 that M222 is 5.3% of Germany.  

. . .

* EDIT: I wonder if this is really R-L21*.


That has got to be an error. I think you are right in your "EDIT" note. It seems to me there are quite a few problems with Myres' figures.

I find it impossible to believe that the L21 frequencies they list for Germany are correct. If German L21 were really as low as Myres et al say it is, we wouldn't have had so many men of German ancestry test L21+ with FTDNA.

5.3% L21 (instead of M222) would make a lot more sense, especially given the 5.6% L21 figure in neighboring Austria.

And those are percentages of the total male population, not merely of the R1b1b2.
« Last Edit: September 01, 2010, 08:34:23 PM by rms2 » Logged

alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #55 on: September 01, 2010, 10:35:12 PM »

. . .
Myres does show on Table S4 that M222 is 5.3% of Germany.  

. . .

* EDIT: I wonder if this is really R-L21*.


That has got to be an error. I think you are right in your "EDIT" note. It seems to me there are quite a few problems with Myres' figures.

I find it impossible to believe that the L21 frequencies they list for Germany are correct. If German L21 were really as low as Myres et al say it is, we wouldn't have had so many men of German ancestry test L21+ with FTDNA.

5.3% L21 (instead of M222) would make a lot more sense, especially given the 5.6% L21 figure in neighboring Austria.

And those are percentages of the total male population, not merely of the R1b1b2.

Another issue I think is M222 is not that common in the parts of Ireland facing the continent.  NW Ireland is very remote from the continent.  

Regarding Austria if I recall correctly R1b1b2 is only about 20 or so so that 5.6% of total population is a big chunk of it.  Its actually a real bonus to get a stat for Austria because it must be the most testing-shy country in western Europe.   
« Last Edit: September 01, 2010, 10:39:15 PM by alan trowel hands. » Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #56 on: September 01, 2010, 10:51:54 PM »

Does anyone have that y-dna study of France (Santiago de Compostela)?

Here it is:

http://tinyurl.com/2btr8va

I meant if someone can share the study.

I think I just read it online. I don't have it on my computer.

Argiedude I found it in the French L21 thread.

http://download.journals.elsevierhealth.com/pdfs/journals/1875-1768/PIIS1875176809002340.pdf

A couple of us emailed the authors to see if they could give us more data or even just % breakdowns of the clades but we never got a reply.


I originally asked because Alan had stated the following:

"I do also wonder again about the figures.  Hobbiest testing has found a lot lot less continental M222 and  think the Santiago de Compostella University paper on French y-DNA only found instances in Paris (hardly surprising) and Tolouse."

I got the impression someone had obtained the haplogroup results by region (or just the results, period!).

We tried hard and emailed the authors but no dice.  I cannot understand why they did not even give percentages.  All we had was the statement that U152 is the biggest clade in the east (Alsace) but elsewhere S116* (which in this case would include untested L21*) outnumbered U152.  When you compare this with the new paper it actually is compatible.  We also had the statement that M222 had been found in Paris and Tolouse but not in the other areas (NW, NW, E, S, SE).  Paris of course is a huge city so that tells us nothing.  The M222 presence in Tolouse is interesting but no numbers are quoted.  However, if you  Paris and note that it was absent in the other testing areas other than Tolouse it makes it mighty hard to see how there could be as significant an M222 presence in France as implied by Myers paper.   
Logged
y24
Member
**
Offline Offline

Posts: 16


« Reply #57 on: September 02, 2010, 05:27:01 AM »

There was large scale migration from the Celtic areas of the Isles into England during the Industrial Revolution:

"After 1840, emigration had become a massive, relentless, and efficiently managed national enterprise. Counting those who went to Britain, between 9 and 10 million Irish men, women, and children emigrated after 1700. The total flow was more than the population at its historical peak in the 1830s of 8.5 million. From 1830 to 1914, almost 5 million went to the United States alone. In 1890 two of every five Irish-born people were living abroad."

"The 2001 UK Census states 869,093 people born in Ireland as living in Great Britain, with over 10% of the population (over 6 million) being of Irish descent."

"10% of the British population has one Irish grandparent."

http://en.wikipedia.org/wiki/Irish_diaspora#Britain
I don't know if you noticed, but the DNA project information I am using is based on MDKA's, mostly of a generation or two or three back.  Some of the very recent numbers (i.e. 2001) you are using are impacted by the current mobility of society as well as changes in the nature of the United Kingdom.
If you noticed, the main Irish migration was more than two or three generations back.
http://en.wikipedia.org/wiki/Irish_diaspora#Britain
That doesn't mean these guys aren't Irish transplants.  They very well could be.  I'm just saying the origin of M222 is an open question from a genetic data point of view.
I'm also just saying, being familiar with the history of the Industrial Revolution and some of our family trees here in England, they could well be 'Irish transplants'. Many of us have a bit of Irish in us, especially in the bigger cities and old industrial towns. It's not unusual, as Tom Jones says.
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #58 on: September 02, 2010, 09:20:14 AM »

That doesn't mean these guys aren't Irish transplants.  They very well could be. I'm just saying the origin of M222 is an open question from a genetic data point of view.
I'm also just saying, being familiar with the history of the Industrial Revolution and some of our family trees here in England, they could well be 'Irish transplants'. Many of us have a bit of Irish in us, especially in the bigger cities and old industrial towns. It's not unusual, as Tom Jones says.
I think it is quite likely there are many Irish transplants throughout England and probably some on the continent.  I'm fine with that. I have no bone to pick and in fact my ggg-grandfather was a Kilkenny Irishman.  However, I don't think the book is closed on M222's origin.  I think some of the Myres data is hard to understand, if not questionable....  but I don't want to assume it away.  M222 is not as young as people think, in my opinion.  The older it is (and/or if the older the TMRCA calculation methods prove correct) then the more likely it originated further east, possibly across the Channel or even the Rhine.

P.S. Tom Jones would be Welsh, right?
« Last Edit: September 02, 2010, 09:21:54 AM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
Jdean
Old Hand
****
Offline Offline

Posts: 678


« Reply #59 on: September 02, 2010, 09:44:49 AM »

P.S. Tom Jones would be Welsh, right?

A proper Valleys boyo
Logged

Y-DNA R-DF49*
MtDNA J1c2e
Kit No. 117897
Ysearch 3BMC9

pconroy
Member
**
Offline Offline

Posts: 6


« Reply #60 on: September 02, 2010, 01:21:55 PM »

Thanks for the recommendation. It sounds like an interesting book.  The beautiful Turoe stone is near the site of my ancestors in Co. Galway.

Also check out this R-M222 discussion thread:
http://archiver.rootsweb.ancestry.com/th/read/DNA-R1B1C7/2009-01/1231241661
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #61 on: September 02, 2010, 03:08:53 PM »

.... We do have one individual in the FTDNA project data that is from Germany - N14949 Lominac.
A second German MDKA has now shown up and joined the M222 project although the name looks French to me.  I wonder how many continental M222's are in FTDNA but wouldn't to join the "Northwest Irish" project (or maybe not any project) because it must have nothing to do with them?  Maybe it's okay now since Myres says there may be some M222 on the continent.
« Last Edit: September 02, 2010, 04:23:27 PM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #62 on: September 02, 2010, 03:30:28 PM »

"I was told by my great aunt that three brothers came over from Alsace Lorraine to South Carolina in about 1840. This is currently France today but has been swapped back and forth with Germany over the centuries.

I had a German professor who had fled Nazi germany, but who had earned a doctorate before the war in Slavic languages. He told me that Lominac(k) is a Czech name, and I notice there is a Lominacki (sp.?) river in the western (German-speaking) portion of the Czech Republic".

This from WEB. Also to me it would seem a slavic surname, but I am not expert of these surnames like  the Italian ones.
But the ancient origin by a genetic point of view is worth more than a word.


Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #63 on: September 02, 2010, 03:48:06 PM »

”Common misspellings and typos for this name: Lomincak, Lmoinack, Loimnack, oLminack, Lominac, Lominaca, Lominsck, Lominack, Lomonack, Lominacka, Lominacke, Lominacki, Lominacko, Lominnack”.

Interesting the metathesis “Loimnack” for “Lominack” like that “Leimone” for “Lemoine” we have discussed on another thread.
 

Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

Heber
Old Hand
****
Offline Offline

Posts: 448


« Reply #64 on: September 02, 2010, 06:24:59 PM »

.... We do have one individual in the FTDNA project data that is from Germany - N14949 Lominac.
A second German MDKA has now shown up and joined the M222 project although the name looks French to me.  I wonder how many continental M222's are in FTDNA but wouldn't to join the "Northwest Irish" project (or maybe not any project) because it must have nothing to do with them?  Maybe it's okay now since Myres says there may be some M222 on the continent.

Mike,
Myres has placed M222 firmly by frequency, diversity and age in Ireland as have all serious studies prior to this one.
Logged

Heber


 
R1b1a2a1a1b4  L459+ L21+ DF21+ DF13+ U198- U106- P66- P314.2- M37- M222- L96- L513- L48- L44- L4- L226- L2- L196- L195- L193- L192.1- L176.2- L165- L159.2- L148- L144- L130- L1-
Paternal L21* DF21


Maternal H1C1



jrw
New Member
*
Offline Offline

Posts: 1


« Reply #65 on: September 02, 2010, 07:58:07 PM »

I am not a statistician, but it appears from some of the earlier posts in this thread that there is too much significance being attributed to some of the figures in Table S4.  Regarding the sample population "Germany," the 5.3% of M222 means that just ONE male out of a sample of 19 tested M222 positive.  If the entire sample population for all of the German cohorts is considered, there is just ONE M222 positive male out of a population sample of 321, or 0.3% -- a much different, and more statistically valid result.  

However, unless there was bias in drawing the population samples,  the result of only 7 males being L21 out of a population sample of 321 for all of the German cohorts, or 2.2%, should be indicative that L21 is rare in present-day Germany.
« Last Edit: September 02, 2010, 08:36:58 PM by jrw » Logged
y24
Member
**
Offline Offline

Posts: 16


« Reply #66 on: September 02, 2010, 09:16:04 PM »

P.S. Tom Jones would be Welsh, right?
Correct.
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #67 on: September 02, 2010, 09:22:26 PM »

I am not a statistician, but it appears from some of the earlier posts in this thread that there is too much significance being attributed to some of the figures in Table S4.  Regarding the sample population "Germany," the 5.3% of M222 means that just ONE male out of a sample of 19 tested M222 positive.  If the entire sample population for all of the German cohorts is considered, there is just ONE M222 positive male out of a population sample of 321, or 0.3% -- a much different, and more statistically valid result.  

However, unless there was bias in drawing the population samples,  the result of only 7 males being L21 out of a population sample of 321 for all of the German cohorts, or 2.2%, should be indicative that L21 is rare in present-day Germany.

That is one of the reasons I think something is very wrong with Myres et al. We have quite a few men of German descent in the R-L21 Plus Project who tested on their own and were not recruited or "cherry picked" by me based on matches with known L21+ men. Several of them are German citizens, currently living in Germany.

If L21 were so rare in modern Germany, I don't think we would have seen the results we have seen.

I'm not buying it, say what you will. Something is wrong with Myres et al.
« Last Edit: September 02, 2010, 09:26:10 PM by rms2 » Logged

Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #68 on: September 02, 2010, 09:39:03 PM »

.... We do have one individual in the FTDNA project data that is from Germany - N14949 Lominac.
A second German MDKA has now shown up and joined the M222 project although the name looks French to me.  I wonder how many continental M222's are in FTDNA but wouldn't to join the "Northwest Irish" project (or maybe not any project) because it must have nothing to do with them?  Maybe it's okay now since Myres says there may be some M222 on the continent.

Mike,
Myres has placed M222 firmly by frequency, diversity and age in Ireland as have all serious studies prior to this one.
Heber,

How many serious studies have even tested for M222 in Germany?  How many studies have done a good cross-sectional, random sampling of German Y DNA, period?  Once upon a time, all the serious scholars thought the earth was flat.

I don't think the Myres study data is definitive and probably does have serious errors, but the truth is there is no "deeper" pan-European set of data available for R1b1b2 so I think it is useful, just not conclusive.

I guess, you are saying you are 100% sure that M222 originated in Ireland.  Is that true?  If so, show me the study and the specific information that proves this.  I think you are talking about some of the Trinity College stuff, but show where you think M222 is proven to be of Irish origin.

Don't get me wrong, I think the chances are good that M222 originated in Ireland or the western side of Britain, but I think there is an excellent chance it could be on the continent as well.
« Last Edit: September 02, 2010, 09:43:36 PM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
argiedude
Old Hand
****
Offline Offline

Posts: 146


« Reply #69 on: September 02, 2010, 11:04:42 PM »

I looked at yhrd, and there are 6300 German samples. Most M222+ have all 3 of the following off modal values: 390=25, 385=11/13, 392=14. The 2 M222+ from the continent in Myres study are like this (though 385 wasn't tested). Only 1 of the yhrd German samples had all 3 values.
Logged

y-dna: R1b L21
mtdna: U5
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #70 on: September 06, 2010, 05:30:13 AM »

It does not appear that the Myres paper or indeed what we already knew really supports a major Iberian component in the spread of R1b1b2.  It still looks to me north and western France  are probably the main point of connection with the Atlantic seaways of the British Isles.  Myers gives a pretty high L21 count for western France and although Myres didnt cover it we know beyond a shadow of doubt from the project testing that the north of France is considerably higher still.  However other than the north-easternmost Iberian sample from Santander (which basically confirms the project's identification of a minor Iberian peak in the extreme NE corner of Spain) there is very very little L21 in most of Iberia.  I also note that S116 * is very low in variance in Ireland, far lower than S116* there,  so it certainly does not suggest and Atlantic isles origin for L21 is even remotely possible either.   
Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #71 on: September 06, 2010, 06:38:56 AM »

Looking at the Myres paper one thing stands out about L21 and S116* which although a paragroup must contain the ancestral forms to L21.  In Ireland S116* is very low in variance.  Indeed it is much lower than the Irish variance for L21*!!  The same is also true although to a less extreme degree for England.   L21 cannot have arisen in a place where it is older than the ancestral form S116*.  This youth of Irish S116* has been noted before in terms of the surnames of S116* people often being not native Irish. 

The expected picture for the area where L21 was born from an S116* ancestor is that S116* will be (at least a little) older.  That fits far better the French and German picture than anywhere else.  L21* there is fairly old but not as old as S116.* 

So, I think it is still very much supported that L21 happened on the continent, probably in France among the older S116* groups there.
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #72 on: September 06, 2010, 09:38:51 AM »

I went to Ysearch and looked for additional R-M222 (R1b1b2a1b5b with FTDNA as test co.) from England.  I found an additional five beyond the ten I had. I re-ran several variance calculations.  England keeps coming up as having more variance than Ireland, even though England's sample is quite limited.  Whether England is a cross-roads or not, if Ireland is THE true origin point it should have more variance.

M222 still comes up half the the age of R-L21. This is given in large sample sizes in the 100's and 1000's so I think this is pretty accurate.

So which is it?  Is L21 6000 BC and M222 2000 BC?  or is it that L21 is 2000 BC and M222 is 0 AD?     of course there is in between but M222 seems to be no younger than 0 AD at the latest.

.... We do have one individual in the FTDNA project data that is from Germany - N14949 Lominac.
A second German MDKA has now shown up and joined the M222 project although the name looks French to me.  I wonder how many continental M222's are in FTDNA but wouldn't to join the "Northwest Irish" project (or maybe not any project) because it must have nothing to do with them?  Maybe it's okay now since Myres says there may be some M222 on the continent.

Mike,
Myres has placed M222 firmly by frequency, diversity and age in Ireland as have all serious studies prior to this one.
Heber,

How many serious studies have even tested for M222 in Germany?  How many studies have done a good cross-sectional, random sampling of German Y DNA, period?  Once upon a time, all the serious scholars thought the earth was flat.

I don't think the Myres study data is definitive and probably does have serious errors, but the truth is there is no "deeper" pan-European set of data available for R1b1b2 so I think it is useful, just not conclusive.

I guess, you are saying you are 100% sure that M222 originated in Ireland.  Is that true?  If so, show me the study and the specific information that proves this.  I think you are talking about some of the Trinity College stuff, but show where you think M222 is proven to be of Irish origin.

Don't get me wrong, I think the chances are good that M222 originated in Ireland or the western side of Britain, but I think there is an excellent chance it could be on the continent as well.
Logged

R1b-L21>L513(DF1)>L705.2
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #73 on: September 06, 2010, 10:27:29 AM »

...  L21 cannot have arisen in a place where it is older than the ancestral form S116*. ....
Alan makes a very good point there.  As a platform from which to evaluate, below are the R-P312*(S116*) 25 STR relative variance scores by region of France and other parts of Europe.  The sample sizes are also shown and you can see they are sometimes limited.

France North & Central _ 1.58 __ 008
France Southeast _______ 1.21 __ 004
Aquitaine & Pyrenees ___ 1.18 __ 007
Iberia exc. Pyrenees ___ 1.10 __ 038
East Europe Continental_ 1.09 __ 023 (see note below)
Alpine & Cisalpine _____ 1.07 __ 006
Germany Middle & South _ 1.03 __ 015
France North Atlantic __ 0.97 __ 006
Scandinavia ____________ 0.86 __ 015
Low Countries __________ 0.85 __ 011
Italy, Greece, Malta ___ 0.76 __ 007
France Northeast __________ only 002


The East Europe sample includes Bulgaria, Czech Rep, Hungary, Lithuania, Poland, Romania, Russia, Ukraine. The above numbers are based on straight Sum of the Variance so higher variance markers have moe impact.  If I normalize STR's so that all are weighted equally, East Europe comes up as a much higher variance, higher than France North & Central.  This is the equivalent of saying that East Europe has THE highest variance on the slow moving markers.

Notice that Mediterranean Europe's R-P312* looks relatively youthful. I don't think P312 came from south of the Pyrenees or the Alps.

As far as the topic of the origin of P312, I think U152's data needs to be considered with P312 hand in hand as U152 seems to be close to as old as P312 itself.
« Last Edit: September 06, 2010, 10:34:04 AM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #74 on: September 06, 2010, 02:57:26 PM »

Apparently Myres et al made little effort at obtaining a truly representative sample in France. Just look at Table S4 and the results for L21 versus U152. Small differences in terms of absolute numbers look more impressive when they are chalked up as the percentages of small sample sizes.

L21 (xM222):

France East  N=25  8% (2 individuals out of 25)

France [?]     N=16  0% (0 individuals out of 16)

France West  N=14  14.3% (2 individuals out of 14)

France South N=38   7.9% (3 individuals out of 38)


U152 (subclade unspecified):

France East  16% (4 individuals out of 25)

France [?]      18.8% (3 individuals out of 16)

France West  21.4% (3 individuals out of 14)

France South 10.5% (4 individuals out of 38)


Notice a number of things. First, Northern France was not represented at all, which is clear from the maps that show the locations from which the samples were taken. Second, nearly 68% (63 of 93) of the French samples were drawn from the South and East, precisely where U152 is believed to be most frequent. Third, the sample sizes are small.

Thus, in a relatively small sample heavily skewed to the South and East of France, L21 (xM222) came out with 7.5% of the total, and U152 came out with 15% of the total.

Had Northern France been sampled, and if Western France had been sampled more extensively, what might the results have been?

Note that in Western France the difference between L21 and U152 was a single individual: there were two L21s out of 14 and three U152s out of the same 14.

The biggest differences between L21 and U152 in Myres et al came in France East (believed to be U152's stronghold anyway), where L21 got 2 individuals and U152 got 4 (out of 25), and in (just plain) "France" (whatever that means), where L21 got zero and U152 got 3 (out of 16).

Again, notice that the sample sizes were small, and the differences in absolute results are likewise small (2 versus 4, for example).


« Last Edit: September 06, 2010, 03:00:44 PM by rms2 » Logged

Pages: 1 2 [3] 4 5 6 Go Up Print 
« previous next »
Jump to:  


SEO light theme by © Mustang forums. Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC

Page created in 0.158 seconds with 18 queries.