World Families Forums - Wow! R-L21+'s twelve founders (sons) - Anatole's view

Welcome, Guest. Please login or register.
April 20, 2014, 12:10:29 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)
| | |-+  Wow! R-L21+'s twelve founders (sons) - Anatole's view
« previous next »
Pages: [1] 2 3 Go Down Print
Author Topic: Wow! R-L21+'s twelve founders (sons) - Anatole's view  (Read 5335 times)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« on: October 16, 2009, 09:59:19 AM »

I throw out the caveats that speculations and proposals about haplotype clusters and hierarchial trees are based the limitations of statistics and the validity and representative nature of the data sampled.

However, Anatole Klyosov, is clearly a sharp fellow (google him if you wish) and I think you must consider his views carefully.

He has done some initial analysis on the R-21* data and has made the following comment:

"It turned out that L21 tree is perfectly symmetrical and points at only one common ancestor (see attached). As you see,  it contains at least 12 local branches, all of them descended from one common ancestor."

I posted his diagram in the "FILES" section of the R-L21 Yahoo group at
http://tech.groups.yahoo.com/group/RL21Project/
along with the spreadsheet itself.     You do have to join the group but you don't have to receive any email messages from it.  Under "Edit Membership", select the option "Web Only - Don't get notified of the latest happenings. Read messages only on the web" if you are worried about email in-basket traffic.

BTW, he also feels R-U152, R-L21's large brother clade within R-P312, originated practically at the same time.  
"By the way, it exactly as in R-U152. It shows that they have formed practically at the same time. R-U106 are different by only two mutations in the 25-marker format, that is only 600 years from THEIR common ancestor."
« Last Edit: October 16, 2009, 09:59:49 AM by Mike » Logged

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

Posts: 2963


WWW
« Reply #1 on: October 16, 2009, 11:38:24 AM »

Here is the methodology that Anatole used:
corrected link:  http://www.springerlink.com/content/e7701424635633p7/

He also sent me a legible "linear diagram" where it actually shows the FTDNA & Ysearch ID's and the proposed phylogenetic tree that leads back up (or I should say to the left) to the 12 founders and ultimately to the L21 TMRCA haplotype.

I'll post this diagram also.
« Last Edit: October 16, 2009, 11:55:10 AM by Mike » Logged

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

Posts: 731


« Reply #2 on: October 16, 2009, 02:58:36 PM »

Here is the methodology that Anatole used:
corrected link:  http://www.springerlink.com/content/e7701424635633p7/

He also sent me a legible "linear diagram" where it actually shows the FTDNA & Ysearch ID's and the proposed phylogenetic tree that leads back up (or I should say to the left) to the 12 founders and ultimately to the L21 TMRCA haplotype.

I'll post this diagram also.
I am very curious whether he posits a geographic location for the expansion.
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #3 on: October 16, 2009, 03:04:54 PM »

Here is the methodology that Anatole used:
corrected link:  http://www.springerlink.com/content/e7701424635633p7/

He also sent me a legible "linear diagram" where it actually shows the FTDNA & Ysearch ID's and the proposed phylogenetic tree that leads back up (or I should say to the left) to the 12 founders and ultimately to the L21 TMRCA haplotype.

I'll post this diagram also.
I am very curious whether he posits a geographic location for the expansion.

Ditto. Double ditto in fact.
« Last Edit: October 16, 2009, 03:05:13 PM by rms2 » Logged

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

Posts: 2012


« Reply #4 on: October 16, 2009, 03:22:55 PM »

The very modest amount of the non-isles L21 relative to the isles tested to 67 markers is still a big limitation.  We cannot possibly be seeing much of the continental L21 picture when a whole continent is far less represented than a couple of islands off its NW edge.  I will be interested to see Anatoles thoughts but I could not take him seriously if he comes up with an isles origin given the sheer amount of L21 turning up in NW and west-central Europe.   
Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #5 on: October 16, 2009, 03:25:54 PM »

Is an origin point implied in his diagram.  I cant see it properly and am far too lacking in basic computer literacy to work out how to. 
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #6 on: October 16, 2009, 04:31:50 PM »

Is an origin point implied in his diagram.  I cant see it properly and am far too lacking in basic computer literacy to work out how to.  
The diagram does not list geographies, just haplotype ID's.   He does talk a little bit about geography in his email to me.  He said it was okay to quote him so here it is.
Quote from: Anatole Klyosov. date=1255721154
Dear Mike,
 
As a follow-up of your thoughtful discussions on forums, attached is a haplotype tree composed of 509 of 67-marker R-L21 haplotypes.  

In my "world history" shaped in a way by haplotype analysis, R1b might have entered Europe by two (at least) routes. The starting point was in Asia, 16 thousand years ago. That what R1b haplotypes (rather, their mutations) show. They had migrated across the Russian Plain (6,900 years to their common ancestor), Anatolia (6,000 ybp), Lebanon (5,200 ybp; 5,500 ybp in Jewish R1b haplotypes as well), then via North Africa to Algeria (3800 ybp) to Iberia (3700 ybp) and to the Isles (3600 ybp). I have all those numbers with plus-minus 95% confidence intervals, etc. Here they are given just as some illustrations, without details.
 
However, Central Europe has R1b1 with 4500 years to a common ancestor. It does not fit to the route described above, unless that common ancestor lived in the Middle East, and "brought" to Europe in haplotypes of R1b1b2 migrants. It might be that another migration wave of R1b had entered Europe directly from the Russian Plain, or directly from the Middle East, not via Northern Africa. However, it is only 4200 ybp in Italy, that is a "common" figure.  
 
Therefore those L21 were of a particular interest for me. I figured that if L21' common ancestor lived around 3700 ybp, the route might have been the same. If L21 are of 4500 ybp, there certainly was another route, such as described above.
 
In my analysis of those L21, I had to establish first if the haplotype series is "right", that is has only one (!!) common ancestor, hence, can be calculated properly. There are several criteria for this. One is a symmetrical haplotype tree, coming from ONE common ancestor. If there were two or more of "common ancestors", a tree typically would have two or more distinct branches, sitting apart from each other on the tree.
 
It turned out that L21 tree is perfectly symmetrical and points at only one common ancestor (see attached). As you see,  it contains at least 12 local branches, all of them descended from one common ancestor. All those branches are "younger" than the tree itself, and they shows local (geographical) distribution of various L21 subfamilies. Maybe deeper subclades, maybe just subfamilies. Apparently, those groups you have mentioned on the RootsWeb. They can be followed with a map in hand. You cam play with them, if you wish. I have a "linear" tree in which all numbers are visible, as well as all the subbranches.
 
The second criteria concerns mutations in the whole tree (all 509 haplotypes) vs. a number of "base" (ancestral) haplotypes in the series. They should be in an agreement (mutations and a number of base haplotypes).  The more time passed from a common ancestor, the more mutations are accumulated, and the less of those "intact" haplotypes remain. There is a mathematical formula connecting the two (mutations vs. a number of base haplotypes). It is published in my paper a few days ago in Human Genetics. For example, among all 770 of 12-marker L21 haplotypes on the list there are 49 "base" haplotypes
 
13 24 14 11 11 14 12 12 12 13 13 16
 
They are ancestral ("base") haplotypes, hence, they are the most frequent in the series. This is already enough to calculate a time span to a common ancestor, even without considering mutations. As you see, one does not need even to count mutations. The formula is [ln(770/49)]/0.022 = 125 generations to a common ancestor, and after a correction for back mutations it gives 143 generations (25 years per generation, this was a condition in the calibration; 0.022 is the mutation rate constant per 12-marker haplotype), or 143x25 = 3575 years to a common ancestor. At 95% confidence interval it is 3575+/-620 years bp.
 
If we consider 25-marker haplotypes, there are only two base haplotypes left
 
13 24 14 11 11 14 12 12 12 13 13 16 -- 17 9 10 11 11 25 15 19 29 15 15 17 17
 
By the way, it exactly as in R-U152. It shows that they have formed practically at the same time. R-U106 are different by only two mutations in the 25-marker format, that is only 600 years from THEIR common ancestor.
 
Anyway, with 2 base haplotypes among 712 of 25-marker haplotypes in the series, [ln(712/2)]/0.046 = 128 generations without a correction for back mutations, or 147 generations with the correction, that is 3675 years to a common ancestor. 0.046 is the mutation rate constant for 25-marker haplotypes. As you see, it is practically the same ybp as the 3575 ybp for 12-marker L21 haplotypes. It is a very good sign that the calculations are correct.
 
Finally, all 509 haplotypes contain 2924 mutations in the first 25-marker haplotypes. These is no need to count all mutations in all 67-marker haplotypes, the 25-markers are the most precise for calculations. It gives 2924/509/0.046 = 125 generations without the correction, or 143 generations with the correction, that is 143x25 = 3575 years to a common ancestor (3575+/-370, to be meticulous). As you see, it is exactly the same number. This indicates that there was indeed only one common ancestor for the whole series, and the calculations are justified.
 
There are 14 haplotypes with null 425 marker. they all contain 60 mutations, which gives 60/14/0.046 = 93 generations (w/o correction), or 103 generations with the correction, that is 2575 years to a common ancestor. However, that mutation was formed 4400 years back, and descendants carry it on since then. It was formed elsewhere (the Middle East?), and, apparently, only one person (or only one survivor) brought it to Europe 2575 years ago. He became a common ancestor of that lineage. So, this null mutation is not formed randomly, as it was suggested in this forum. It is inherited and carried on.
 
You see how much can be extracted from a list of haplotypes?
 
Regards,
 
Anatole Klyosov
 
P.S. Feel free to quote the above at the RootsWeb or elsewhere.
« Last Edit: October 16, 2009, 04:33:03 PM by Mike » Logged

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

Posts: 2012


« Reply #7 on: October 16, 2009, 05:39:32 PM »

So, Anatole is saying all L21 from the west of Ireland to the Alps has a common ancestor in post-Roman times.  That is the historic period when any movement would have been known.  The various high L21 areas like Ireland, northern France, SW Germany etc share absolutely no common denominator in that period. 
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #8 on: October 16, 2009, 06:33:05 PM »

So, Anatole is saying all L21 from the west of Ireland to the Alps has a common ancestor in post-Roman times.  That is the historic period when any movement would have been known.  The various high L21 areas like Ireland, northern France, SW Germany etc share absolutely no common denominator in that period. 
I'm not following where you are getting the "post-Roman".  The quote below seems to indicate about 1700BC for a TMRCA for L21 or 2500BC.  Where you interpreting "post-Roman"?

Quote from: Anatole
I figured that if L21' common ancestor lived around 3700 ybp, the route might have been the same. If L21 are of 4500 ybp, there
Logged

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

Posts: 5023


« Reply #9 on: October 16, 2009, 06:56:38 PM »

I don't buy his route for R1b into Europe. It doesn't make sense.

Russia to Anatolia to Lebanon to North Africa to Iberia to the Isles?

From what I have seen, the SNP track shows a SE to NW track out of Anatolia or the Balkans.

I wonder what Vince Vizachero would say about Anatole's R1b route.

He didn't really say where he thinks L21 came from, not clearly anyway.
« Last Edit: October 16, 2009, 06:56:58 PM by rms2 » Logged

Nolan Admin - Glenn Allen Nolen
Project Coordinator
Old Hand
*****
Offline Offline

Posts: 292


WWW
« Reply #10 on: October 16, 2009, 11:08:32 PM »

So, the methods used by Ken and Anatole basically agree with an age for R-L21 and R-U152 of roughly 3,700 years.

The route will be argued about for years to come as will these dates. We will never satisfy skeptics about the dating methods or the routes, but I'm satisfied with the results.  

"The quote below seems to indicate about 1700BC for a TMRCA"

Quote from: Anatole

"I figured that if L21' common ancestor lived around 3700 ybp,"

http://archiver.rootsweb.ancestry.com/th/read/GENEALOGY-DNA/2008-05/1211221129

"I get an age back to the common MRCA for S21 and S28 to be 3780 years."  

And yes, I agree Anatole is a very smart man as is Ken. We should thank each of them for their dedication to this field.
« Last Edit: October 16, 2009, 11:28:23 PM by Nolan Admin - Glenn Allen Nolen » Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #11 on: October 17, 2009, 04:02:45 AM »

So, Anatole is saying all L21 from the west of Ireland to the Alps has a common ancestor in post-Roman times.  That is the historic period when any movement would have been known.  The various high L21 areas like Ireland, northern France, SW Germany etc share absolutely no common denominator in that period. 
I'm not following where you are getting the "post-Roman".  The quote below seems to indicate about 1700BC for a TMRCA for L21 or 2500BC.  Where you interpreting "post-Roman"?

Quote from: Anatole
I figured that if L21' common ancestor lived around 3700 ybp, the route might have been the same. If L21 are of 4500 ybp, there

Sorry I had a brainstorm there and didnt read it properly. He does indeed suggest those dates for a MRCA and for the mutation itself respectively.
Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #12 on: October 17, 2009, 04:41:51 AM »

I think the route Anatole seems to be suggesting was based on upstream R1b not L21.  He cannot have based his Med. route of L21.  Just look at the map.  There is virtually no L21 on the Med. nor any in Russia, Anatolia, Lebanon or Africa and indeed little in Iberia.  I am sure about 99% of L21 is north of the Alps and Pyrenees. I think Anatole's route is his theory based on upstream clades.

He does hint at a 2nd route through central Europe direct from Russia.  I could broadly live with a two route theory of a simultanious spread along the Med. and central Europe of S116* with L21 happening among those S116*who took the Central European (probably Danube) route. 

The dates with confidence intervals touch the beaker period, the only international culture which could possibly correlate with a very widespread east-west population movement in this period.  Again beakers are a mystery and origins and direction of spread have been long debated.  There is a school of thought that they travelled at least parly along the Med. reaching Iberia early but this has never been easy to correlate with the northern and eastern beakers.  Perhaps there were two routes - A Med. one and a Danubian one, as I suggested above for S116, and L21 occurred among the group travelling through central Europe.

I also have to say I am not clear why he starts the trail in the Russian Plains. I could maybe have understood the Caucuses . I also am not sure I understand the Lebanon and Africa parts.  Makes me wonder if there has been too much of an attempt to assume 100% correlation of modern STR patterns with past movements as though no distortions or loss of diversity has happened.     


Logged
Nolan Admin - Glenn Allen Nolen
Project Coordinator
Old Hand
*****
Offline Offline

Posts: 292


WWW
« Reply #13 on: October 17, 2009, 11:31:36 AM »

"I think the route Anatole seems to be suggesting was based on upstream R1b not L21."
 

I'm not going to debate archaeology with you. The dates by Anatole and Ken support my research and the Catastrophe of Avellino as a source for the Irish. So be it, if you want to speculate about this and that and ignore the obvious pink elephant in the room. Speculate away, but the pink elephant isn't going anywhere. 

I'll leave you to your archaeology ramblings and no longer post to this thread.


http://freepages.genealogy.rootsweb.ancestry.com/~nolenancestry/page12.html

Of the Nolans (Nola)

Origins of the Irish and Scottish: Corca Luighe (Corca Laoidhe) and Dál Riada (Dál Riata)



http://freepages.genealogy.rootsweb.ancestry.com/~nolenancestry/the_catastrophe_of_avellino_of_the_1780_bc.txt

The Catastrophe of Avellino 3780 years ago

http://www.areavesuvio.org/modules.php?artid=34&name=Sections&sop=viewarticle

http://web.archive.org/web/20070930043810/www.areavesuvio.org/modules.php?name=Sections&sop=viewarticle&artid=34

September 30, 2007   
 
La Catastrofe di Avellino del 1780 a.C. quale Peggiore Scenario per una Futura Eruzione del Vesuvio



http://www.archaeology.org/0203/newsbriefs/bronzeage.html

A publication of the Archaeological Institute of America

Move Over, Pompeii

Volume 55 Number 2, March/April 2002, by Jarrett A. Lobell
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #14 on: October 17, 2009, 11:44:58 AM »

I think Anatole was referring to upstream R1b, too, which is why he wrote R1b and not R-L21* or R-P312.

My point is that I don't think R1b entered Europe via North Africa and the Pillars of Hercules. I think it entered as R1b1b2 via the Balkans or possibly even farther north in Eastern Europe.

It is possible there was a two-pronged entry, however, with some R1b1b2 coming via the Mediterranean and Iberia and the rest coming in SE to NW via the Balkans and Eastern Europe, probably mostly up the Danube Valley.

I think that latter route is what Vince Vizachero's work with the Ht35 Project shows.

And I think that somewhere along the Danube route is where the brothers L21 and U152 were born.
Logged

Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #15 on: October 17, 2009, 01:37:57 PM »

.... It is possible there was a two-pronged entry, however, with some R1b1b2 coming via the Mediterranean and Iberia and the rest coming in SE to NW via the Balkans and Eastern Europe, probably mostly up the Danube Valley.
I think that latter route is what Vince Vizachero's work with the Ht35 Project shows.
And I think that somewhere along the Danube route is where the brothers L21 and U152 were born.
Assuming an Danube River route, moving one step further back, where do you think this R1b1b2 group was sourced?    Did they come up from the south and the Balkan Peninsula (in land route) and Greece?  Did they come the northern side of the Black Sea or from the southerly coast via Anatolia?
Logged

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

Posts: 2963


WWW
« Reply #16 on: October 17, 2009, 01:48:25 PM »

.... I also have to say I am not clear why he starts the trail in the Russian Plains. I could maybe have understood the Caucuses . I also am not sure I understand the Lebanon and Africa parts.  Makes me wonder if there has been too much of an attempt to assume 100% correlation of modern STR patterns with past movements as though no distortions or loss of diversity has happened. 
I can not read Anatole's mind, but I do agree he is talking about R-M343, M269, etcl and not just R-L21* or he would have specifically cited R-L21* in those earlier quotes.

I think he starts the trail in Russia because of a common thought among some researchers is that R, R1, R2 along with P and Q are all possibly from Central Asia.  National Geno's Dr. Wells outlines this in "The Eurasian Heartland: A continental perspective on Y-chromosome diversity" -  http://www.pnas.org/content/98/18/10244.abstract

The timing and direction of movements/expansions from their are what the story is all about, but Central Asia seems to be long lost home.
Logged

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

Posts: 2012


« Reply #17 on: October 17, 2009, 05:18:28 PM »

Glenn
Your body of work includes looking at interpretation of the parts of Irish mythology and ancient clan genealogies and other sources that purport to refer to the prehistoric period, although written down a lot later in Medieval times.   Clearly you have done a lot of in-depth research and a lot of work and thought and have compared it to the DNA evidence.  My knowledge is in a different discipline but I am basically doing the same thing as you - comparing a body of knowledge from another discipline with genetics.  

People researching within the same discipline often disagree with each other's conclusions. So, it is even less likely people will agree if its comparing two separate bodies of knowledge with a common third discipline (DNA).  If we have different ideas and conclusions then so be it.  We do not need to agree.  I like this hobby as I can see that its the efforts of many dedicated hobbiests that drives progress but there would be no fun in this hobby if we all agreed.  
« Last Edit: October 17, 2009, 05:55:48 PM by alan trowel hands. » Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #18 on: October 17, 2009, 05:23:41 PM »

[[/quote]
Assuming an Danube River route, moving one step further back, where do you think this R1b1b2 group was sourced?    Did they come up from the south and the Balkan Peninsula (in land route) and Greece?  Did they come the northern side of the Black Sea or from the southerly coast via Anatolia?[/quote]

I get the impression that the route from the south side of the Black Sea to SE Europe is slowly coming more into favour than a steppes one from what I read on the net but I think only extensive testing along both of those possible routes can answer this.  If anyone has insight on this it would be Vince Vizachero but he doesnt seem to post here.    
« Last Edit: October 19, 2009, 03:50:49 PM by alan trowel hands. » Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #19 on: October 17, 2009, 05:45:22 PM »

I don't think the North African route makes sense, given the R1b1b2 SNP trail. That trail leads from east to west through Europe, or southeast to northwest, but the North African route is just a variation on the old out-of-Iberia theme. If it's true, then we should see greater splitting off at upstream nodes in Iberia and just across the Pillars in North Africa.
« Last Edit: October 17, 2009, 05:45:56 PM by rms2 » Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #20 on: October 18, 2009, 07:49:16 AM »

The diagrams aren't legible to me, especially the linear one. Even when I magnify it 400%, I can't read it.

Sorry.
Logged

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

Posts: 2012


« Reply #21 on: October 18, 2009, 09:43:42 AM »

The diagrams aren't legible to me, especially the linear one. Even when I magnify it 400%, I can't read it.

Sorry.

I cant read the early part of the diagram either.  Can anyone summarise in cluster terms what clusters are descended from what sons and which clusters of type of L21 is nearest the roots.  Is there a geographical pattern? 
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #22 on: October 18, 2009, 03:21:35 PM »

The diagrams aren't legible to me, especially the linear one. Even when I magnify it 400%, I can't read it.
I cant read the early part of the diagram either.  Can anyone summarise in cluster terms what clusters are descended from what sons and which clusters of type of L21 is nearest the roots.  Is there a geographical pattern? 
I played around with different tools to open it with and when I use "Windows Fax and Picture Viewer" I can zoom in enough to discern which "single thread" line is to the left and right as they layer.  It is just very difficult to scroll back/forth to read the kit IDs and up and down and stay with each line's relative position/latyer.  It would be great to have a hi-res printer and lay it all out on a desk... or a table.   A rainy day type project when everyone is out for a few hours.
Logged

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

Posts: 191


« Reply #23 on: October 18, 2009, 11:57:21 PM »

I wonder what Vince Vizachero would say about Anatole's R1b route.

I'd say its pretty wacky, and unlikely besides.
Logged
alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #24 on: October 19, 2009, 03:53:48 PM »

I wonder what Vince Vizachero would say about Anatole's R1b route.

I'd say its pretty wacky, and unlikely besides.

Vince

Have you an opinion on where P310 and P312 happened.  It seems to me that their origins are worse defined that upstream of them.
Logged
Pages: [1] 2 3 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 0.132 seconds with 19 queries.