World Families Forums - R-M269 article on Holocene era founder effect on Central and West Europe

Welcome, Guest. Please login or register.
July 30, 2014, 12:50:21 PM
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)
| | |-+  R-M269 article on Holocene era founder effect on Central and West Europe
« previous next »
Pages: 1 [2] 3 4 ... 7 Go Down Print
Author Topic: R-M269 article on Holocene era founder effect on Central and West Europe  (Read 21553 times)
argiedude
Old Hand
****
Offline Offline

Posts: 146


« Reply #25 on: August 25, 2010, 11:30:30 PM »

R-L21*'s least frequent quarter of France is the southeast.   This turns out to be a various diverse area for P312 overall, probably mostly because of U152, which I think is the oldest subclade of P312, along with P312* and SRY2627.. a little of everything.

Agreed. And let's keep in mind that the study didn't test SRY2627 and M153. These results are hidden amongst their P312* samples, giving an erroneous first impression of a bigger than expected presence of P312* in southwest Europe.
Logged

y-dna: R1b L21
mtdna: U5
Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #26 on: August 26, 2010, 02:07:17 AM »


Argiedude writes: "Can't wait to see what Maliclavelli is going to find about ht35,
or what Stevo has to say about L21".

Argiedude, I spoke yesterday first of this paper on this site (see the thread
"R-P312+ et al tracking" and "Dienekes blog").

Living in Italy, I have a different time zone: when you are awake I do sleep.
I have said what I had to say: my theory, supported by your maps, like that of R-L51
and the double presence, ab antiquo, in Italy and in  the Caucasus region of R1b1b2
(but also of R-M420), I think you have illustrated by your maps. For the rest I have
said that I am waiting for the aDNA, above all that of Oetzi probably next year.

About R-L21 in Italy I have said that I have disproved all, except yours, but that
it too is in danger. I have written many postings on this, I don't know if you have read
them. You know I am against who hides himself, even though you probably have some right
motives. To say that in Italy there are one R-L21 (yours) you should test some relative
of yours in Italy and resolve your link with those Callaway I spoke about in other
postings. Unfortunately here we go in the private, but science demands accuracy
and pitilessness.
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #27 on: August 26, 2010, 02:20:32 AM »

Another paper on this argument is announced by "Dienekes":
"Strong intra- and inter-continental differentiation revealed by Y chromosome SNPs M269, U106 and U152 " by the famous Cruciani (et alii).

Unfortunately they aren't for free.

But let me know: these data come from individuals tested for the first time and by chance? or are they from data already available? If they are new, that some R-L21 has been found in Italy demonstrates that there it is, even though at a low frequence.
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

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

Posts: 2012


« Reply #28 on: August 26, 2010, 04:57:35 AM »

so if there were 40 S116 in France and 7 (17.5%) were L21 and 14 (35%) U152, then what were the remaining nearly 50% of French S116?  I would never have guess half of French S116 would be S116* or the Iberian clades.

Got to say too that these stats are absolutely light years away from the rates the L21 project resting of R1b1b2 found although that may partly be due to many being from the north-western quarter of France.  
I don't know if they tested for SRY2627 or M153 but we know there are some SY2627 folks in France.  Also, keep in mind that I don't think the study is claiming to be a scientific sampled "frequency" study.  It's more like spot checking for age calculations.

I went to Table S4 of the study and pulled up their frequencies for M529xM222 which should be L21*.   Even though some of these percentages are low, L21* is in places randy monks and wild geese didn't make it to.

I believe these are percentages of the total population:

54.2% Ireland South
50.0% Ireland West
40.9% Ireland Southwest
33.3% Ireland North
31.3% Ireland East
25.0% England North
24.0% England Southeast
20.8% England Southwest
19.4% France Alpes de Haute Provence
14.3% France West
09.5% Denmark North
09.1% Denmark Southeast
08.2% France Vaucluse (upstream Rhone)
08.0% France East
08.0% England Central
07.9% France South
07.4% Switzerland Northwest
06.3% France Bouches du Rhone (at mouth)
05.8% Sweden South (Malmö)
05.7% Netherlands
05.6% Austria
05.3% Spain Cantabria, Santander
05.3% Denmark West
03.1% Germany North
03.1% Switzerland Northeast
03.0% Portugal, Lisbon
02.9% France, Var (coastal, East of Rhone)
02.4% Spain, Castille and Leon, Leon
02.4% Poland
02.2% Germany South
02.1% Germany East
02.0% Switzerland (Lower Rhone Valley)
01.9% Croatia mainland
01.1% Czech Republic
01.1% Poland Southwest(Wroclaw)
01.0% Germany West
00.9% Spain, Valencia
00.9% Hungary
00.8% Andalusia, Sevilla
00.8% Russia Central
00.8% Italy North
00.6% Romania
00.4% Russia South
00.4% Slovakia


Some very odd things there.  Only half of R1b1b2 coming in as L21 in western Ireland an 40% in SW Ireland??? That is a long way from the usual understanding.  Only 30% if SE Ireland is L21?? That area is about 75 % R1b1b2 so that is like saying 80% of the eastern Irish are not L21.  2.2% in Germany south seems almost impossible to tally with the project results for the Rhineland.  While it can be argued that the there is a bias towards the Rhineland due to migration patters I cannot see how that factor could bias the clade count in the project for the Rhineland.  In the project L21 is up their with U152 etc in that area.  On the other hand the showing the Denmark is pretty respectable. There is no northern France stat in the list above which is where we would expect the continental L21 max.  Is that an oversight??
« Last Edit: August 26, 2010, 04:59:45 AM by alan trowel hands. » Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #29 on: August 26, 2010, 05:06:15 AM »

so can this still fit the beaker model??
Logged
Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #30 on: August 26, 2010, 05:15:57 AM »

What have I to say about R-L23, above all in Italy? Rich will respond by himself about R-L21, but I confirm what I have said thus far: that in Italy probably it isn’t and I am waiting a confirmation of
that of Argiedude. I am seeing now that Myers hasn’t found it in Italy: the 0,008% of another spreadsheet  comes from previous studies and we should investigate if they are always the same: Bonnet, Argiedude, Basile, all sub judice re. their reliability. As you know, this confirm my theory about the “invasions” of Italy: Celts, Germans etc. I have always said that they haven’t lost practically anything. The massive presence of R-U152 (they say in North Italy, but actually overall, and on Rootsweb David Faux complains about the fact that Sardinia hasn’t been tested, as the two SNP-tested are come out R-U152). The high percentage of R-U152 amongst Bashkyrs has at last resolved our questions: as I have said many times in the past, not only they are very recent, they have suffered a founder effect, probably have something to do with Tokharians, and other similar haplotypes near the Urals I have analysed in the past as R-U152 (perhaps Argiedude remembers). If many are thinking that R-U152 is the most ancient subclade of R-P312, this is a good sign for my Italian refugium.

Re R-L23 (repeating what  has rightly said Vizachero quoting the paper about the rough reliability of the data for the fortuitousness of the chosen samples) I can say that Italy has, also in this inquiry, the highest variance in Western Europe. The fact that there was (and is) an ancient presence in the Caucasus we did know (and Argiedude’s maps have demonstrated), but my theory was that the Western European subclades were born from the Italian R1b1b2* and not from the Eastern one, for this I am asking that is tested the “Rozen’s SNP”, that I think does separate a Western R1b12* from an Eastern one. Anyway (for Vizachero) also in this paper nothing to do with Middle East: Asia Minor, Caucasus aren’t Palestine. Also for R-L51 (M412) Italy has an high presence, that doesn’t come out from the “ht35 project”, where only an Italian is present amongst many not Italians.

Who knows me knows that I like to call persons for name and surname, and for pedigree.
When Dante meets Farinata degli Uberti (Inferno, X), probably a good R-U106 of German descent,
asks: “Chi fur li maggior tui?” (Who were your ancestors?) A research, done by chance, without knowing History and Geography, which is the usual among these illustrious scholars, is at least rough. My theories presuppose the knowledge of infinite disciplines. Then, of course, I may be wrong, but the paper of Ferri et alii on Modena (the Etruscan Mutina) found R1b at 67,7% and R1b1b2a (L23+) was at least 15/16% and the variance one of the highest I have ever seen.
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #31 on: August 26, 2010, 05:36:36 AM »



Some very odd things there.  Only half of R1b1b2 coming in as L21 in western Ireland an 40% in SW Ireland??? That is a long way from the usual understanding.  Only 30% if SE Ireland is L21?? That area is about 75 % R1b1b2 so that is like saying 80% of the eastern Irish are not L21.  2.2% in Germany south seems almost impossible to tally with the project results for the Rhineland.  While it can be argued that the there is a bias towards the Rhineland due to migration patters I cannot see how that factor could bias the clade count in the project for the Rhineland.  In the project L21 is up their with U152 etc in that area.  On the other hand the showing the Denmark is pretty respectable. There is no northern France stat in the list above which is where we would expect the continental L21 max.  Is that an oversight??

From what I can see, it looks like the testing in France was heavily skewed to the South and East. Naturally, that would inflate the U152 figures and deflate the L21 figures.

But I think this study's results are bizarre. Something very odd has occurred. Maybe I'm wrong, but it just doesn't jive with what we have seen so far in the project. I'm baffled, frankly. And I don't like the way this study's report is organized. Things could have been laid out more clearly.

Also, when I look at table S4 in the supplementary data, all I get are figures for Asia and none for Europe. Of course, I am inexperienced with spreadsheets, so I'm probably not handling the thing correctly.
Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #32 on: August 26, 2010, 05:50:02 AM »

I've got to say this report is very suspicious to me. You could say I think this merely because things didn't turn out as I expected, and that is partly true. But my expectations were based on observations of test results through FTDNA over the past year and a half or so since L21 was discovered.

I don't know exactly what is wrong with this report, but something is. Unfortunately, now it's out there, and it will be considered by many to be the last word on the subject.

I wonder if Myres et al didn't go to SMGF once again and, as a consequence, get a largely Mormon sampling for Western Europe. If so, I wonder what effect that could have on the outcome.

Maybe I'm grasping at straws, but I'm admittedly flabbergasted by the low frequency of L21 on the Continent. This report, quite frankly, makes me feel I have been wasting a lot of time, hard work and money on the R-L21 Plus Project.

Perhaps I should have just let it be and let it become the Irish social club so many wanted it to be.

I'm pretty close to washing my hands of the whole mess.
« Last Edit: August 26, 2010, 05:52:49 AM by rms2 » Logged

vineviz
Old Hand
****
Offline Offline

Posts: 191


« Reply #33 on: August 26, 2010, 06:05:36 AM »

Maybe I'm grasping at straws, but I'm admittedly flabbergasted by the low frequency of L21 on the Continent. This report, quite frankly, makes me feel I have wasting a lot of time, hard work and money on the R-L21 Plus Project.
I wouldn't be so despondent.  We've always known that the frequency of R-L21 is higher in the Isles than in the Continent, but that's not evidence for an Isles origin.

Myres do not have even sampling on the Continent, or within France.  But they do have concentrations of R-L21 (as a % of R-M269) pushing 30% in some places which is, itself, an important finding.  More than other clades, R-L21 diversity is about the same everywhere but is still a little higher on Continent which is good news for you I think.
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #34 on: August 26, 2010, 06:39:34 AM »

I was finally able to manipulate Table S4 so that I can see everything. Those frequencies are percentages of the R1b1b2, not of the total male population, right?
Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #35 on: August 26, 2010, 06:49:44 AM »

I was finally able to manipulate Table S4 so that I can see everything. Those frequencies are percentages of the R1b1b2, not of the total male population, right?


Do the references that sometimes say, "This study", mean that Myres et al pulled samples from previous studies and tested them?

Obviously some of those references are to studies that were conducted too early to include L21 testing. Surely if they used figures from those studies to determine frequency, they must have tested the samples for L21, right?
Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #36 on: August 26, 2010, 07:11:06 AM »

I was finally able to manipulate Table S4 so that I can see everything. Those frequencies are percentages of the R1b1b2, not of the total male population, right?


Okay, I think I can answer my own question. Those frequencies are percentages of the entire y-dna population, not just the R1b1b2.

Interesting that L21 is 19.4% of the total y-dna in Alpes de Haute Provence in SE France.
Logged

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

Posts: 2012


« Reply #37 on: August 26, 2010, 07:38:14 AM »

I think though before getting too blown away by the lower percentages for L21 we should also note that U152 which seems at first like the winner in all of this is also not exactly spectacular (below 10 and often below 5%) in many areas.  It only gets big scores in a few places.  I think to some extent we are just not used to seeing them as a percentage of total populations.  All of the R1b clade percentages seem low (except a couple of places like U152 in Switzerland) and maybe we have all gotten a bit carried away.  I also think this is an eye opener in terms of how much western European R!b is not L21, U152 or U106 in many areas.  I have already posted about some of the results seeming plain wrong or lets put it this way hugely different than expected.  If taken at face value then even the more remote west and south-west of Ireland where the population is largely 'native' is less than half L21 whereas it has become commonplace to assume its 90=odd % L21

In terms of L21 I think project testing essentially proved a decent amount of L21 in the north of France and the Rhineland provence in SW Germany and adjcacent areas with a thin string linking the two.  I do not think it ever was showing much L21 elsewhere.  Not much in east and south France, not much in much of Germany, Belgium, Holland east of the original Rhine line, practically zilch in Italy and the whole eastern Europe (except Jewish). Much of the recent Iberian L21 was uncovered by chasing matches and a signifcant amount seemed a recent cluster.  

So the bottom line of the L21 project was an extremely high northern French hit rate in fairly random testing of R1b2b2 (which is undeniable) and a respectable showing of L21 in and adjacent to the Rhineland province (the only bit of Germany west of the Rhine). That seemed solid because there is no reason why one clade should be overrepresented in an area. Its areas that migration patterns bias towards not the clade count within the area.  

So did the study contradict the core findings of the project.  Firstly, there is no northern France stat.  That is hugely important because that is where most continental L21 is.  So there is no contradiction there.  The result of this study does seem to contradict project expectations in the SW Germany. In the projects L21 competes successfully with U152  but is trounced by it in this study there.  That is very hard to explain.  There is absolutely no reason to see why new world SW Germans should tend to be unrepresentative.  Its weird.  

One area that is interesting is that L21 does better in the north (and weirdly Austria) than it does in inland west-central Europe.  I think the project was begining to uncover this aspect.  L21 has a decent (if unspectacular) presence across the northern coast of Europe.  I had in recent times started to think that the maritime side was important for L21 with so much in the west of the isles, NW France, Norway and little (orobably very small) pockets found in the Basque area and around Portugal.  This study strengthens the feeling the North Sea coastal areas also had a significant L21 presence and if anything shifts the centre of L21 to the north.  

It hard to interpret something with no agreed dates.  One thing that is clear though even from this study as well as others is that L21 did not originate in Ireland.  What evidence there is from variance points to France.  Its not and never has been about numbers.  We should all know by now that a higher count does not tell you anything about the point of origin.  I would still feel that L21 probably occurred as S116 was passing westwards into northern France which is what I have always felt.  That is why its relatively rare elsewhere.  There are not that many 'out of northern France' movements in prehistory other than towards the isles.  The area did have some wider contacts which would explain movement by sea from there to leave small to moderate traces all along the North Sea and Atlantic.  What this study seems to point to though is that L21 is more a north Atlantic, English Channel, North Sea thing rather than an 'Atlantic' one that took in Iberia to any major degree.

In terms of interpretation, the paper does focus on the LBK-descended cultures of northern France as pivotal in the final spread of R1b1b2 in the middle Neolithic, including the isles.  That still fits well with the idea that L21 occurred at the end of the LBK trail as it entered northern France.  So, although the beaker theory was getting the upper hand this study much more easily fits into the LBK model that the authors have suggested, something I have generally felt more comfortable with.  

Finally, I would not be too despondent about the low L21 count in much of the continent.  We never had a very high L21 count or any huge reason to suspect one other than in the north of France which this project seems to have missed out in its sample.  If its the truth and not bad sampling then at least we now know where to focus testing.  Targeted testing was until recent months focussed on France and we should be thankful for the fact that we picked the right spot in terms of return for the money.  Even France without the north still came in relatively high for L21 compared to most other places and there was a surprise that the Alps area seems to have a decent amount.  If that is true then that is interesting.  France for me remains the best area in terms of return for investment in testing L21.  What we now know is (if we can trust the sample, which I doubt) that testing elsewhere (other than by close matches) would be very much long-shot.    
 
Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #38 on: August 26, 2010, 07:43:24 AM »

I was finally able to manipulate Table S4 so that I can see everything. Those frequencies are percentages of the R1b1b2, not of the total male population, right?


Okay, I think I can answer my own question. Those frequencies are percentages of the entire y-dna population, not just the R1b1b2.

Interesting that L21 is 19.4% of the total y-dna in Alpes de Haute Provence in SE France.

It is very interesting.  There is no way of explaining that presence across the Alps on the Med. away with mad monks etc (although a few will try).  You any idea why the north of France is not in the table. Seems odd because there map for L21 looks very very like the project map.  I think they may have borrowed it.   
Logged
Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #39 on: August 26, 2010, 08:47:28 AM »

Alan writes: “There is no way of explaining that presence across the Alps on the Med. away with mad monks etc (although a few will try”.

Don’t forget Argiedude. I made two hypotheses re. his haplotype: either he is linked with Callaway, a NPE happened in Argentine or elsewhere, or his ancient haplotype testifies the first origin also of this haplogroup in the Alpine region, in what is for me the Italian Refugium.

Don’t underestimate Gioiello!
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #40 on: August 26, 2010, 08:53:57 AM »

Of course the course was from Italy to, via Rhein, British Isles: see the K1* of Oetzi and infinite samples, clear above all at mtDNA level. This explains why  R-L21 isn't in Italy. It probably went out from here and never came back.
« Last Edit: August 26, 2010, 08:56:43 AM by Maliclavelli » Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

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

Posts: 2012


« Reply #41 on: August 26, 2010, 09:05:08 AM »

One interesting aspect of the LBK theory touched on in this paper is their idea that R1b1b2 in western Europe would have in the early Neolithic been confined for a long period to the LBK areas.  I think their model would demand that S116 and U152 happened in central Europe with L21 not happening until it was crossing into norther France or possibly happening only after some S116* people who had settled there developed L21 in subsequent generations.  What the report seems to be suggesting is that there was a secondary expansion from the LBK area of inland northern  and eastern France in the middle Neolithic that spread R1b1b2 into new areas.  In the north of Europe these were areas not previously settled by hunters.  The L21 map they present does have a very close resemblance to the area of northern Europe that the farmers expanded into in the European middle Neolithic.  So some sort of role for L21 among that expansion seems to be indicated.  In terms of what we know from the project this would include a very strong expansion in northern France and on to the isles (what a pity the project did not give stats for northern France). There is a considerable area where L21 is lacking in landlocked eastern France between this known north French hotspot and the unexpected hotspot south of the Alps in France.  There is an intermediate area in the Rhineland Palatinate that the FTDNA project has a small but strong hotspot but for some reason the sampling in this paper has not picked it up (did they test in the Rhineland?).  Its an important hotspot even if it is small because it links the northern France area to the an area that is not too far from the Rhone and access to the south of France.Otherwise we need to look at founder effects.  They clearly happen as is shown by the isles.  Another aspect if martime travel.  Could it be that small L21 groups ruled the waves and settled in pockets (some very small, some larger) along trade routes?  Even this new French pocket south of the Alps might fit in.
« Last Edit: August 26, 2010, 09:09:05 AM by alan trowel hands. » Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #42 on: August 26, 2010, 09:42:47 AM »

I think though before getting too blown away by the lower percentages for L21 we should also note that U152 which seems at first like the winner in all of this is also not exactly spectacular (below 10 and often below 5%) in many areas.  It only gets big scores in a few places.  I think to some extent we are just not used to seeing them as a percentage of total populations. ...
 
Quote from: vineviz
... More than other clades, R-L21 diversity is about the same everywhere but is still a little higher on Continent
I'm excited about the data, just because we have more than we had before.  It is clear to me that this is not a good, cross-sectional, random sampling that could be used to project frequencies accurately for whole countries.

As Alan pointed out, the numbers are also presented as a % of the total population.  L21+ is a relatively young, high resolution clade when compared to what we are used to looking at on haplogroup maps.

My opinion is that the sampling, although widespread, is not strong in many locations (where is N. France? why only 4 locations in Spain?)  and therefore is hit or miss "spotty".  I don't know if you call the sampling scientific or not, but it was not a statistically sound cross-sectional, random sampling of all of Europe.   The expense of such an endeavor would be enormous.

Frequency (% count) of the population is a red herring.  It is not useless by any means, but the diversity of the haplotypes is more important for determining potential age and migration paths.
« Last Edit: August 26, 2010, 09:49:25 AM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #43 on: August 26, 2010, 09:54:13 AM »

The more I read the paper the more exciting it is.  The variance points to an origin in SW Asia for R1b1b2, S116 in eastern France, U152 in Germany (presumably south or south-west) and L21 in France.  This confirms several other hobbiest calculations that France edges out England with Ireland further behind.  However, it also seems to confirm that L21 once it had ccurred in France spread pretty rapidly to the isles.  This remains in line with the idea that L21 occurred at the north-west terminus of LBK.  LBK is only really known in France where its eastern borders touched the Rhine , along the Meuse which led into the Seine and Loire river systems of northern France.  So if L21 occurred in France it must have occurred in those areas if there is an LBK link.  I still suspect that northern France may have been where it first expanded rather than its origin point.  The problem with origin points is that if an SNP  happens on a trail west but the line its on is not expanding then the chances are that no descendants will be left from the pre-expansion part of the story.  In fact I suspect that in many cases in a period of migration the early part of the trail of movement of an SNP may be completely lost.  On expansion ensured that we still have descendants to test.  In pre-expansion periods when he numbers carrying a new SNP are low and there is a lot of line extinction then the MRCA could be a lot later than the the SNP.  One advantage f linking R1b1b2 to an expansion like the LBK is that it was so spectacular that lines had a much higher chance of survival.  

One further thought.  Perhaps L21 happened in northern France among the descendants of LBK people but after the LBK period.  I think there is evidence in the variance for this.  If L21 happened as it was spreading into France then it would be somewhere between 7000-8000 years old (if I recall correctly).  Now we know that the Neolithic spread to England around 4400BC (new early dated sites) and Ireland by just after 4000BC.  If L21 was up to 8000 years old in France but under 6000 years old in Ireland we would expect French L21 to be a third greater in France than in Ireland. It seems a lot closer than that so  I think the idea floated in this report that the spread of R1b1b2 (including L21) into the isles was due to post-LBK (but LBK-descended) middle Neolithic groups is interesting. I think they are hinting at something similar regarding U152 and its spread into the Alps, Italy etc.  Unfortunately I think the middle Neolithic of France and neighbouring areas is one of the most uncertain periods and very hard to find much of a recent vintage in print.  
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #44 on: August 26, 2010, 09:57:08 AM »

.... the report seems to be suggesting is that there was a secondary expansion from the LBK area of inland northern  and eastern France in the middle Neolithic that spread R1b1b2 into new areas.  In the north of Europe these were areas not previously settled by hunters.  The L21 map they present does have a very close resemblance to the area of northern Europe that the farmers expanded into in the European middle Neolithic. ...
Which expansions are you referring?  Would they include or be akin to the Rössen?
http://en.wikipedia.org/wiki/Rössen_culture

We had a conversation on this topic back in January:
Quote from: anotherblog
The Rossen Culture may have spawned the Neolithic ("New Stone") Age in Britain. Neolithic farmers invaded Britain around 4000 BC. By about 3900 BC, they had pushed as far north as Yorkshire, and brutal border conflicts erupted with the previous Mesolithic ("Middle Stone") Age hunter-gatherers*. The invading Neoliths built Causewayed Enclosures like those seen earlier in Germany**, the oldest begun around 4000 BC***. These Causewayed Enclosures gradually developed from mere earthworks to wooden, and later stone, henges. They continued in unbroken use for thousands of years, only falling into disuse after 1000 BC#, around the time of the first Celtic invasions that brought the Iron Age to Britain.

Quote from: Mikewww
Is the Michelsberg culture as one of your "LBK-descended middle Neolithic cultures"?
If so, there may be evidence for your position with Hg I-L38, who may have been "tribal brother" of R-L21*. Hans De Beule makes the case that I-L38 and R-L21* have coincident distribution patterns in his paper and that they may have come out of the Michelsberg culture as it blossomed from c. 6400 to 5500 ybp.
Origins of Hg I-L38 (I2b2) Subclades - Hans De Beule – 5th of april 2009
BTW, I-L38 is also found in the Lichenstein Cave.
http://sites.google.com/site/haplogroupil38/

umm... I see Hans is now speculating "Paper III: November 2009: Early Bronze Age Origin and Late Iron Age (La Tène) Migrations of I-L38".  Whatever the case, he might as well speculate with us.  It appears I-L38 and R-L21 were in the same boat, or at least same wagon.
« Last Edit: August 26, 2010, 10:09:22 AM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #45 on: August 26, 2010, 10:13:14 AM »

I think though before getting too blown away by the lower percentages for L21 we should also note that U152 which seems at first like the winner in all of this is also not exactly spectacular (below 10 and often below 5%) in many areas.  It only gets big scores in a few places.  I think to some extent we are just not used to seeing them as a percentage of total populations. ...
 
Quote from: vineviz
... More than other clades, R-L21 diversity is about the same everywhere but is still a little higher on Continent
I'm excited about the data, just because we have more than we had before.  It is clear to me that this is not a good, cross-sectional, random sampling that could be used to project frequencies accurately for whole countries.

As Alan pointed out, the numbers are also presented as a % of the total population.  L21+ is a relatively young, high resolution clade when compared to what we are used to looking at on haplogroup maps.

My opinion is that the sampling, although widespread, is not strong in many locations (where is N. France? why only 4 locations in Spain?)  and therefore is hit or miss "spotty".  I don't know if you call the sampling scientific or not, but it was not a statistically sound cross-sectional, random sampling of all of Europe.   The expense of such an endeavor would be enormous.

Frequency (% count) of the population is a red herring.  It is not useless by any means, but the diversity of the haplotypes is more important for determining potential age and migration paths.


In a situation where sampling is too patchy to really be sure and so broad brush that whole hotspots can be missed (including L21's in unsampled Northern France and the Rhineland Palatinate) then I think the best approach is accept the positive evidence for presence for the spots it did sample but do not accept it as negative evidence for absence in areas it did not or as a generalisation for the country or wider area.  That is the approach I take to project data too.  There is no question in my mind that L21's stomping grounds on the continent were missed.  I do however accept that in all probability that the big picture that L21 is lower in central Europe than expected is strong enough that that negative evidence can be accepted broadly.  We can clearly take pleasure in the positive evidence of higher than unexpected L21 areas like southern France, Denmark and Austria.  If you combine that with other areas of higher L21 identified by the FTDNA project (and I think the projects data should be taken in tandem not set in opposition) then L21 has an almost bizarre pattern of strong pockets separated by very low patches.  If this dotty nature of L21 is typical then a very dotty sampling as in this paper could easily miss important L21 pockets as it certainly has for Northern France and surely also has for the Rhineland Palatinate.  I also would not be too hasty to throw out evidence for small but distinctive pockets in eastern and western fringes of Iberia.  They may be small but important and could be easily missed in a study such as this.  IMO, this paper has enough sampling issues to suggest that we should not think it replaces the project findings.  the two sources should be combined and where there seems to be contradiction there may actually not be any.  
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #46 on: August 26, 2010, 10:21:43 AM »

umm... I see Hans  (De Beule) is now speculating "Paper III: November 2009: Early Bronze Age Origin and Late Iron Age (La Tène) Migrations of I-L38".  Whatever the case, he might as well speculate with us.  It appears I-L38 and R-L21 were in the same boat, or at least same wagon.
I do recommend reading Hans' "Late Iron Age" paper.  Whether you agree with his conclusions or not, he provides some good background for the geography and situation that is relevant to R-L21.   I find it amusing that he struggles with the same issues we do.  A critical factor is TMRCA.  What TMRCA ranges should you trust? If you trust the "younger ages" that don't use the Zhiv rates, etc. and are more in line with the 2008 Behar study then you come back to thinking Bronze/Iron Age things like this speculation...
http://sites.google.com/site/haplogroupil38/home/III.Nov2009_I_IL38_EBA_and_LateIronAge_Origin.pdf?attredirects=0&d=1
« Last Edit: August 26, 2010, 10:36:21 AM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #47 on: August 26, 2010, 10:27:50 AM »

.... the report seems to be suggesting is that there was a secondary expansion from the LBK area of inland northern  and eastern France in the middle Neolithic that spread R1b1b2 into new areas.  In the north of Europe these were areas not previously settled by hunters.  The L21 map they present does have a very close resemblance to the area of northern Europe that the farmers expanded into in the European middle Neolithic. ...
Which expansions are you referring?  Would they include or be akin to the Rössen?
http://en.wikipedia.org/wiki/Rössen_culture

We had a conversation on this topic back in January:
Quote from: anotherblog
The Rossen Culture may have spawned the Neolithic ("New Stone") Age in Britain. Neolithic farmers invaded Britain around 4000 BC. By about 3900 BC, they had pushed as far north as Yorkshire, and brutal border conflicts erupted with the previous Mesolithic ("Middle Stone") Age hunter-gatherers*. The invading Neoliths built Causewayed Enclosures like those seen earlier in Germany**, the oldest begun around 4000 BC***. These Causewayed Enclosures gradually developed from mere earthworks to wooden, and later stone, henges. They continued in unbroken use for thousands of years, only falling into disuse after 1000 BC#, around the time of the first Celtic invasions that brought the Iron Age to Britain.

Quote from: Mikewww
Is the Michelsberg culture as one of your "LBK-descended middle Neolithic cultures"?
If so, there may be evidence for your position with Hg I-L38, who may have been "tribal brother" of R-L21*. Hans De Beule makes the case that I-L38 and R-L21* have coincident distribution patterns in his paper and that they may have come out of the Michelsberg culture as it blossomed from c. 6400 to 5500 ybp.
Origins of Hg I-L38 (I2b2) Subclades - Hans De Beule – 5th of april 2009
BTW, I-L38 is also found in the Lichenstein Cave.
http://sites.google.com/site/haplogroupil38/

umm... I see Hans is now speculating "Paper III: November 2009: Early Bronze Age Origin and Late Iron Age (La Tène) Migrations of I-L38".  Whatever the case, he might as well speculate with us.  It appears I-L38 and R-L21 were in the same boat, or at least same wagon.

Sorry but I really had tried hard to find out more about the latest ideas on the western European middle Neolithic but I really have struggled.  Its a pretty specialist area and I suspect a lot of information is in French or German language specialist journals which I really do not have the time to look into.  Problem with archaeology is that in recent decades a lot more information and data has been uncovered but its now too big so good summary books of a period are now rare.  The same phenomenon means that books from even say the 1980s are hopelessly out of date.  In the past useful summaries were more common but basically too much is now know for many to contemplate overviews.  Fortunately in relatively recent times there have been good summaries on Europe's first farmers (early Neolithic) and the Mesolithic but the cultures of that period were very widespread.  The problem with the middle Neolithic on the continent (contemporary with the early Neolithic of the isles) is that Europe fractured into so many different cultures.  

Yep Rossen is one of the many.  I posted the name of a few others names last night but I could tell you very little about them.  I was broadly aware of the suggestion that Chassee was involved in some possible expansion from France into Spain and Italy but could find out very little.  Noone has ever exactly pinpointed one single middle Neolithic culture as responsible for the settling of the isles so it may have been settled on a broad front from Brittany to the Rhine.
Logged
vineviz
Old Hand
****
Offline Offline

Posts: 191


« Reply #48 on: August 26, 2010, 10:35:33 AM »

In a situation where sampling is too patchy to really be sure and so broad brush that whole hotspots can be missed (including L21's in unsampled Northern France and the Rhineland Palatinate) then I think the best approach is accept the positive evidence for presence for the spots it did sample but do not accept it as negative evidence for absence in areas it did not or as a generalisation for the country or wider area.  That is the approach I take to project data too.

I agree.  I think we accept the samples for exactly what they are, but should refrain from aggregating them up to countries or regions unless the individual samples from table S4 are demonstrably representative.  In some cases (e.g. France) they clearly are not, but in others they might be.

The daunting thing is that this paper started out with 10,000 Y-DNA samples.  To get the kind of resolution we would really be happy with we'd probably need 5x-10x the samples.  I'll be surprised if we see a truly random, cross-sectional Y-DNA study of that magnitude any time soon.  Something the scope of the Genographic project maybe could have done it, but I'm afraid they've blown their chance.
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #49 on: August 26, 2010, 10:46:58 AM »

The maps are a bit confusing given the data tables.

Apparently they found no U106 in their Swedish sample, and Norway wasn't tested, yet you wouldn't know that from looking at the U106 map in the report.

There were ten S116 in the Swedish sample. Of those, 7, or 70%, were L21+ (M529+). All of those were xM222 (M222-).

In Denmark, 20 were U106+ and 16 were S116+. Of those latter, 7 were L21+ (xM222). Six of the Danes were S116*, which leaves three S116 who were probably SRY2627+ (but we don't know that for sure).

Interesting that their Scandinavian sample, apparently limited to Denmark and Sweden, contained not even a single U152.
« Last Edit: August 26, 2010, 10:54:59 AM by rms2 » Logged

Pages: 1 [2] 3 4 ... 7 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 0.211 seconds with 19 queries.