World Families Forums - R-L21 Clusters - can we identify a few?

Welcome, Guest. Please login or register.
July 25, 2014, 07:04:26 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-L21 Clusters - can we identify a few?
« previous next »
Pages: 1 2 3 [4] 5 Go Down Print
Author Topic: R-L21 Clusters - can we identify a few?  (Read 10419 times)
mcg11
Member
**
Offline Offline

Posts: 38


« Reply #75 on: May 29, 2009, 06:54:03 AM »

[


If L21 so old, why is it so difficult to spot an L21+ haplotype? Why are they so like other P312+ haplotypes? Why is the L21 Modal essentially the same as the U152 Modal, etc.?

My haplotype, and a few others in L21+ are not easy to place.  I have been called eastern european among others.  The predominant number of entries in both L21 and U152 are young, as you know.

When I saw V. Tilroes cluster diagram, it started me thinking.  Why are there a bunch of us that don't cluster?  I have the same problem with the Ian Cam of Clan Gregor.  I am considered to be, by the site manager, a distant relation, but clearly not Ian Cam, most of which are younger than 1500 years.

When Vince and Ken do a intraclade analysis, what do they work with for haplotypes?  Its most probably a modal haplotype?

I am having a discussion with Kerry ODair on DNA Forums and he has found the same problem in their analysis of E Haplogroup entries.  I think, possibly by using the non-clustering haplotypes we may be able to infer the older TMRCA to the clade founder.

Note: when I do an intraclade between myself and the heriditary chief of Clan Gregor I get a TMRCA of over 11k BP, subsequently when I converge the Ian Cam, I first get a convergence at about the time of the reputed clan founder, at the time of Alpin and then finally at the time of the Dal Riada.  It fits well with the historical record and I'm using the same set of mutation rates for both analyses.

All this aside, my basic premise is that modal haplotypes are probably not clade TMRCA's haplotypes.  I think we would be better off estimating the oldest haplotype and using that for our analyses.  JMHO
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #76 on: May 29, 2009, 07:38:48 AM »

Well, like I said, I cannot argue the math with you. Math is something I haven't kept up with since my college days.

But it seems to me someone is making a grievous error - maybe more than one - because you on the one hand and Dr. Nordtvedt, Vince Vizachero, and Dr. Klyosov on the other arrive at such disparate figures. I know that Nordtvedt and the others rely on observed father-to-son ("germline") mutation rates while you rely on Zhivotovsky rates.

The latter give me the impression (as a layman) of having been "cooked" a bit because germline rates were producing figures that were too young to fit expectations.
« Last Edit: May 29, 2009, 07:41:10 AM by rms2 » Logged

mcg11
Member
**
Offline Offline

Posts: 38


« Reply #77 on: May 29, 2009, 10:11:42 AM »

Well, like I said, I cannot argue the math with you. Math is something I haven't kept up with since my college days.

But it seems to me someone is making a grievous error - maybe more than one - because you on the one hand and Dr. Nordtvedt, Vince Vizachero, and Dr. Klyosov on the other arrive at such disparate figures. I know that Nordtvedt and the others rely on observed father-to-son ("germline") mutation rates while you rely on Zhivotovsky rates.

The latter give me the impression (as a layman) of having been "cooked" a bit because germline rates were producing figures that were too young to fit expectations.
I am not the math expert that Ken or Vince are either.  I don't find Klyosov to be very convincing, I've tried his log expression and it doesn't work.  I didn't want to argue mutation rates, I wanted to point out that there may be a flaw in working with modals.  They are biased by the sample set statistics.  A casual look at L-21+ and its major clusters would support a young age.  Certainly, the Ian Cam of the MacGregors, numbering over 60, are such a cluster.  The question is where do the oddballs, like myself, fit in?  If in my clan Gregor analysis, I find that 3 mutations takes the existing chief to the scots modal and that's at least 1500 to 1800 years, then his and my GD separation of 25 has to have a had a blistering rate in the next 1200 years to be consistent with current L-21+ age estimates.  I think our perception of R1b in western europe has to recognize that there may have been a catastrophe around 6K BC, and this changed the genetic/dynamics of R1b in western Europe.
Logged
susanrosine
Senior Member
***
Offline Offline

Posts: 76


« Reply #78 on: May 29, 2009, 02:27:07 PM »

I'd agree with your dubiety regarding modal haplotypes.  Notice that the Modal Haplotype is charted near, but not at the centre of the aforementioned Neighbornet SplitsTree diagram.  It may be close to representing the ancestral haplotype, but probably isn't it exactly.  Maybe using Mean Haplotypes would be slightly more accurate?

I think the closest we can get to visualizing how a bottleneck may look like is by referring to the same diagram, where the M222+ group connects to everyone else.  You've got a fairly bare branch with a nice big bush at the end; plus it coincides with a rather distinctive motif.  Mutations keep occurring regardless of who lives and who dies.  They're kind of apathetic that way.
Interesting idea!! It would be fascinating to see the differences using mean, mode AND median.
I don't know anything about M222+.  What's the estimate on how long between L21 and M222? Isn't M222 strongest in Ireland and Scotland? If it's NOT, then doesn't that just add to the theory that L21 emerged on the continent?
Logged

Dad: JAMES:  Ysearch QSCQ3;  R-P312, L21+ (R1b1b2a1b5*)
Dad: mitosearch QSCQ3; T1a; no matches HVR2 or FGS
Mom's brother: LEWTER: Ysearch FYFDA;  R-U106, L48+ (R1b1b2a1a*)
Mom's brother: mitosearch FYFDA, U5b2; 1 exac
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #79 on: May 29, 2009, 07:58:48 PM »

Well, like I said, I cannot argue the math with you. Math is something I haven't kept up with since my college days.

But it seems to me someone is making a grievous error - maybe more than one - because you on the one hand and Dr. Nordtvedt, Vince Vizachero, and Dr. Klyosov on the other arrive at such disparate figures. I know that Nordtvedt and the others rely on observed father-to-son ("germline") mutation rates while you rely on Zhivotovsky rates.

The latter give me the impression (as a layman) of having been "cooked" a bit because germline rates were producing figures that were too young to fit expectations.
I am not the math expert that Ken or Vince are either.  I don't find Klyosov to be very convincing, I've tried his log expression and it doesn't work.  I didn't want to argue mutation rates, I wanted to point out that there may be a flaw in working with modals.  They are biased by the sample set statistics.  A casual look at L-21+ and its major clusters would support a young age.  Certainly, the Ian Cam of the MacGregors, numbering over 60, are such a cluster.  The question is where do the oddballs, like myself, fit in?  If in my clan Gregor analysis, I find that 3 mutations takes the existing chief to the scots modal and that's at least 1500 to 1800 years, then his and my GD separation of 25 has to have a had a blistering rate in the next 1200 years to be consistent with current L-21+ age estimates.  I think our perception of R1b in western europe has to recognize that there may have been a catastrophe around 6K BC, and this changed the genetic/dynamics of R1b in western Europe.

I think that puts the two of you 3500-4000 years apart, not over 11,000. I'm not an expert, but I don't think you can take a comparison of two haplotypes and make a rule from them.

Logged

mcg11
Member
**
Offline Offline

Posts: 38


« Reply #80 on: May 30, 2009, 08:21:10 AM »

I don't believe it is incorrect to estimate the time between two haplotypes?  It doesn't provide any founder TMRCA information.  I believe it simply presents the mutation time accumulated between the two haplotypes.  There is a similar thread on DNA.Forum and Kerry ODair makes the same comment.

My simple point re: clusters is that you 1. Have to know the ancestral haplotype to make a good estimate.  2. If we use modals based on clusters and don't include all the "odd" haplotypes, we will miss part of the tree and underestimate the founders TMRCA.

My current thinking on what we should do is to converge all the "odd" haplotypes and estimate TMRCA to their modal?
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #81 on: May 30, 2009, 11:28:45 AM »

I don't think there is much point in my continuing to post in this thread, but I would like to say that I don't think much that you're suggesting will do any good unless and until there is some consensus on mutation rates.

If you are thinking L21 is a least 11,000 years old, and Nordtvedt, Vizachero and the others are thinking it is 3,500 years old, something is amiss.

Little can be accomplished until the root of that discrepancy is attended to.
Logged

mcg11
Member
**
Offline Offline

Posts: 38


« Reply #82 on: May 30, 2009, 11:59:28 AM »

I'll tell you where I am coming from.  I think you can appreciate it.  I am trying to ascertain the founder of the Ian Cam of clan Gregor.  From History, we may have 3 choices:  1. Griogair of the Golden Bridles, born c. 1300 AD, 2. King Alpin, first Scotti king of the Scots or, c.850 AD. 3.  One of the Dal Riada boys, c. 300-500 AD.

I have 26+ entries with 67 dys loci.  The clan chief is modal for the group.  I do successive convergences (7 of them) starting with entries of 0 and 1 mutations.  The results when I use Zhivs rate make sense but to include everyone I have to go back to 600 AD.  If I divide by 3.6X (germ-line), I get 388 BP for persons with seven mutations, everyone else is less and the zero and one mutation entries had a common ancestor some 20 years ago.  None of this makes sense. (the germ line results).  The chief has no known close ancestor on the list.  The desired result of the list members is that 385a mutated from 11 to 10 i 1300, so that everyone is a descendant of the clan founder.  I can't get that result using either approach.  You are a site manager, you know how people are sensitive to this data and their place in the pecking order.  I have done the same analysis on Charles Kerchners family, again, a known short term entity and get similar results.  Zhivs rates make more sense.

All, the other TMRCA cases that are presented are basically , hearsay, noone at this time really knows.

So, I don't know what to do?  I am not sure of my results either, but they do make some sense and they can be loosely correlated with historical events.  I have asked VV and Ken to work these two cases, I think VV has worked the Kerchner data but he doesn't publish his work.  Ken ignores my request.  Note that there are over 60 Ian Cam with 37 dys loci measured.

I am just trying to perform an analysis, which on the surface, seems tractable.  If you are aware of a fairly large cluster of entries with similar attributes and known historical background, I would be more than glad to evaluate them, using both rates, and see what I get.  The results should speak for themselves.  Apparently the Kerchner and Clan Gregor results are insufficient?
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #83 on: May 30, 2009, 12:12:16 PM »

Honestly, I don't know what the problem is, but obviously there must be one.

Is it certain that everyone in the Ian Cam of Clan Gregor is R-L21*?

My first impression is that you are starting with some assumptions that may not be true. What if you are trying to converge people on a common ancestor - one of those three alternatives you posed - that they do not actually share?

What happens when you take the variance of those haplotypes using germline rates without making any assumptions?
Logged

mcg11
Member
**
Offline Offline

Posts: 38


« Reply #84 on: May 31, 2009, 03:06:17 PM »

I really don't know?  I know myself, the chief and the website mgr are (he is not Ian Cam). (this is re: L-21+). 

Richard, the web site mgr, is very careful who he places in the Ian Cam, its kind of a badge of honor for some folks.

Rather that just use the 3.6X factor, I'm going to go back to Johns rates and use them directly on the same set of data and see what I get.

Note: In my derived rates and Johns I will use 33 year per gen.  It better fits the historical data we have about MacGregors.
Logged
mcg11
Member
**
Offline Offline

Posts: 38


« Reply #85 on: June 02, 2009, 07:17:45 AM »

Well, I'm prepared to eat "humble pie".  I approached the 26, 67 dys loci entries of Ian Cam MacGregor the same way.  The approach is to work from the least mutations to the most in increments, and in each case discard dys loci where I have to guess whether two mutations are independent or unique.  At the end, I did the case of all 26 at once as a sanity check.

It is clear to me that the results using Chandlers rates better correlate with historical data than the Zhiv rates.  Both techniques predicted a very small number of years for the first convergence because only two CDYa,b mutations are apparent.  It is apparent that using the Chief as ancestral, i.e. modal is the best I can do, but it is also apparent that he probably had mutations at these two dys loci, but they are hidden.  Zhivs rates gave convergences that kept going back in time as I increased the number of mutations, Chandlers did also but they converged on 700 BP, which, based on historical data is believed to be correct.  Note all the other scotti clans have a 11 at 385a, MacGregor is unique at that site, but further the MacGregors along with several other clans form the Siol Alpin, who are all believed to be descendants of one of the Sons of King McAlpin, who lived about 850 AD.  To make sense out of this history, the 385a mutation from 11 to 10 had to occur about 1300 AD, and then MacGregors, Buchanans, McMillans and some Grants I believe converge into each other.

The results of the Chandler rates convergence support this historical scenario, the Zhiv rates do not.  Ergo, I conclude that the Chandler rates are closer to correct than Zhivs.

Note this analysis did not use ASD/Variance, it simply counted mutations.  The one or two multi-steps were counted as one mutation.  Also, one other small quibble, I do not believe Chandlers rate estimate for 388 is correct, It should be the same as 426.  In all databases I've used they usually have the same number of mutations, I believe some I1a data contaminated his 388 estimate?  Further, for this cluster, I assumed that one person was the ancestral haplotype so that the convergences could be estimated.  In general, if a good candidate isn't known, I would use the modal haplotype with the understanding that there may be an error in the estimate.  There appears to be a small one in this example.  Note that I computed the SD of this estimate and got 701 +/- 57 years as the estimate.  The SD was obtained by taking the individual dys loci year contributions and taking the square root of the sum of the squares.

This has been a minor setback in my studies which have previously only used Zhiv rates.  I still have some reservations about the use of Variance and converging clusters/clades where a ancestral haplotype is unknown, but I fully acknowledge my error in using Zhivs rates.
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #86 on: June 02, 2009, 07:24:48 AM »

That is good news to me, Robert.

Are you rethinking the ideas about the origin of L21?
Logged

mcg11
Member
**
Offline Offline

Posts: 38


« Reply #87 on: June 02, 2009, 12:15:01 PM »

I will go back and redo the 393 = 12 first to see what I get.  The approach I used should probably work for any tight knit cluster.  If you have any requests, I can try and see what I get.  It's taken quite a while to develop this approach.
Logged
Maliclavelli
Guru
*****
Offline Offline

Posts: 2146


« Reply #88 on: June 02, 2009, 01:05:40 PM »

Zhiv rate isn't useful under a few thousands of years, but I think that cannot be disregarded for many thousands. If it hasn't been the time for backmutations, it is unreliable.
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #89 on: June 08, 2009, 10:18:39 PM »

The following list has not be deep clade R tested, although a number of them have tested to 67 markers.

My bet is that they are all R-L21* and they meet the 406S1=11/617=13 criteria for what some people call the "Ros Mascy Super Family Combo Group" (at least as I understand it.)    There are a lot of 19/394=15 and YCAIIa/b=18/23 on this list, which are both off-modal and both fairly slow moving.

RXYKH Walsh - Co. Kilkenny, Ireland/Wales
WTDAE Morgan - Ponty Pool, Wales
KRXND Evans - Merthyr, Wales
SD7EF Livingston - Scotland
7F56N Anderson/Wallentinus - Östergötland, Sweden
32J89 Welsh - Co. Kilkenny, Ireland
VGX5F Banks - Colonial US
JTK5M Banks - Colonial US
CPUNG Stephens - Colonial US
UQCFX Barrett - Colonial US

Below, I've got the relevant modals and then a Ysearch string you can use to look at these haplotypes.
XN33E L21 Combo Cluster
KEFGX Wales Modal 3

RXYKH, WTDAE, KRXND, SD7EF, 7F56N, 32J89, VGX5F, JTK5M, CPUNG, UQCFX, XN33E, KEFGX
Logged

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

Posts: 5023


« Reply #90 on: June 09, 2009, 07:46:32 AM »

Here's another one (406S1=11/617=13) for you: Keasler (Germany), YSearch M9JF4.

Mueller of Germany (2ZU3D) also has it, but he's already a known R-L21* and a member of the R-L21 Plus Project.

Maybe you mentioned these two earlier?

« Last Edit: June 09, 2009, 08:04:39 AM by rms2 » Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #91 on: June 09, 2009, 08:11:43 AM »

Hammerlaen (4YRVX) of the Netherlands also has 406S1=11/617=13. He is R-L21* and a member of the project, too.
« Last Edit: June 09, 2009, 08:12:23 AM by rms2 » Logged

Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #92 on: June 09, 2009, 09:50:21 AM »

Hammerlaen (4YRVX) of the Netherlands also has 406S1=11/617=13. He is R-L21* and a member of the project, too.
Yes, the cluster with 406S1=11/617=13 is much larger than the list I put together just below.   Part of the list below doesn't have past 37 markers so we can't really tell if they are 406S1=11/617=13.  It's just that because they are also 19=15 and YCAIIa=18, like myself, I think they are probably 406S1=11/617=13 as well.   Whenever I find R1b1b2 with 19=15 and YCAIIa=18, if it there are 67 markers, it seems like they are all 406S1=11/617=13 as well.  That is my speculation along with the idea they must be L21* as well.
Logged

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

Posts: 2963


WWW
« Reply #93 on: June 11, 2009, 11:21:24 PM »

Hammerlaen (4YRVX) of the Netherlands also has 406S1=11/617=13. He is R-L21* and a member of the project, too.
... It's just that because they are also 19=15 and YCAIIa=18, like myself, I think they are probably 406S1=11/617=13 as well.   Whenever I find R1b1b2 with 19=15 and YCAIIa=18, if it there are 67 markers, it seems like they are all 406S1=11/617=13 as well.  That is my speculation along with the idea they must be L21* as well.
This speculation held true in this case. Anderson/Wallentinus, Ysearch 7F56N. His deep clade R came in recently and he is L21*. 
« Last Edit: June 11, 2009, 11:21:55 PM by Mike » Logged

R1b-L21>L513(DF1)>L705.2
vtilroe
Project Coordinator
Old Hand
*****
Offline Offline

Posts: 150


« Reply #94 on: June 12, 2009, 12:40:44 AM »

This is pretty cool!  Boy, I hope you find a SNP in WTY to cinch this cluster down!
Logged

YSearch & MitoSearch: 2GXWW


yDNA: R-U106*


mtDNA: U5a1a1 (Genbank# GQ368895)


R-P312-WTY Project Admin http://tinyurl.com/daertg

susanrosine
Senior Member
***
Offline Offline

Posts: 76


« Reply #95 on: June 12, 2009, 12:38:44 PM »

The following list has not be deep clade R tested, although a number of them have tested to 67 markers.

My bet is that they are all R-L21* and they meet the 406S1=11/617=13 criteria for what some people call the "Ros Mascy Super Family Combo Group" (at least as I understand it.)    There are a lot of 19/394=15 and YCAIIa/b=18/23 on this list, which are both off-modal and both fairly slow moving.

RXYKH Walsh - Co. Kilkenny, Ireland/Wales
WTDAE Morgan - Ponty Pool, Wales
KRXND Evans - Merthyr, Wales
SD7EF Livingston - Scotland
7F56N Anderson/Wallentinus - Östergötland, Sweden
32J89 Welsh - Co. Kilkenny, Ireland
VGX5F Banks - Colonial US
JTK5M Banks - Colonial US
CPUNG Stephens - Colonial US
UQCFX Barrett - Colonial US

Below, I've got the relevant modals and then a Ysearch string you can use to look at these haplotypes.
XN33E L21 Combo Cluster
KEFGX Wales Modal 3

RXYKH, WTDAE, KRXND, SD7EF, 7F56N, 32J89, VGX5F, JTK5M, CPUNG, UQCFX, XN33E, KEFGX

Mike, so are you in both Wales Modal 3 group AND the Ros Mascy Super Family Combo Group? If they aren't the same thing, can you explain how they differ?
 Did you ask Morgan and Evans to join the Wales_Cymru project? If not, I'll ask them. That way you won't be alone in your cluster LOL.
Logged

Dad: JAMES:  Ysearch QSCQ3;  R-P312, L21+ (R1b1b2a1b5*)
Dad: mitosearch QSCQ3; T1a; no matches HVR2 or FGS
Mom's brother: LEWTER: Ysearch FYFDA;  R-U106, L48+ (R1b1b2a1a*)
Mom's brother: mitosearch FYFDA, U5b2; 1 exac
mcg11
Member
**
Offline Offline

Posts: 38


« Reply #96 on: June 17, 2009, 09:07:32 AM »

I will go back and redo the 393 = 12 first to see what I get.  The approach I used should probably work for any tight knit cluster.  If you have any requests, I can try and see what I get.  It's taken quite a while to develop this approach.

I have spent quite a bit of time working with two Scottish Clusters.  First Clan Gregor, whom I get converging to a founder in 1308 +/- 57 years.  I also converged a set of 177 campbells, using only 25 dys loci and got 511 +/- 33 years.  Using a different approach for the Campbells, their modal haplotype of 67 dys loci, I had estimated 1425 AD as their clan founder date.  I then converged the Clan Gregor modal to the Scots modal and get a date of about 850 AD, that also fits in with the Campbell modal to scots modal.  Finally I estimated the distance to L-21+ and using two different approaches I get 3700 -4000 BP as the time of the founder of L-21+(note this agrees well with Tim Janzens estimates, but is a little older than his estimates).  Based on historical data about where clan gregor came from I would guess the benelux countries may be where it occurred.  Reading other posts about Iberia on this forum, suggests to me, that Iberia may be a source also, but it may also imply that there was a lot of travel between the Benelux countries and other countries by water?
Logged
mcg11
Member
**
Offline Offline

Posts: 38


« Reply #97 on: June 17, 2009, 09:13:33 AM »

The following list has not be deep clade R tested, although a number of them have tested to 67 markers.

My bet is that they are all R-L21* and they meet the 406S1=11/617=13 criteria for what some people call the "Ros Mascy Super Family Combo Group" (at least as I understand it.)    There are a lot of 19/394=15 and YCAIIa/b=18/23 on this list, which are both off-modal and both fairly slow moving.

RXYKH Walsh - Co. Kilkenny, Ireland/Wales
WTDAE Morgan - Ponty Pool, Wales
KRXND Evans - Merthyr, Wales
SD7EF Livingston - Scotland
7F56N Anderson/Wallentinus - Östergötland, Sweden
32J89 Welsh - Co. Kilkenny, Ireland
VGX5F Banks - Colonial US
JTK5M Banks - Colonial US
CPUNG Stephens - Colonial US
UQCFX Barrett - Colonial US

Below, I've got the relevant modals and then a Ysearch string you can use to look at these haplotypes.
XN33E L21 Combo Cluster
KEFGX Wales Modal 3

RXYKH, WTDAE, KRXND, SD7EF, 7F56N, 32J89, VGX5F, JTK5M, CPUNG, UQCFX, XN33E, KEFGX

 

Scots modal entries seem to be characterized by 531 = 12.  I am 10 at 406 S1 and 12 at 617 and am L-21+?
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #98 on: June 17, 2009, 03:43:25 PM »

The following list has not be deep clade R tested, although a number of them have tested to 67 markers.

My bet is that they are all R-L21* and they meet the 406S1=11/617=13 criteria for what some people call the "Ros Mascy Super Family Combo Group" (at least as I understand it.)    There are a lot of 19/394=15 and YCAIIa/b=18/23 on this list, which are both off-modal and both fairly slow moving.....
 
Scots modal entries seem to be characterized by 531 = 12.  I am 10 at 406 S1 and 12 at 617 and am L-21+?
Just did a quick check.  406S1=10 & 617=12 are modals so there are 257 out of 603 (total R-L21*)* that have this configuration.     I selected for just 531=12 out of that group and came up with 44.  13 of those 44 were from Scotland and the rest were mostly unknown, but there was a scattering - Ireland, Norway, Sweden, Germany and England all had representatives with 531=12.

* Only about 320 of the 603 haplotypes have 67 markers. 

Really, you should look at it as 44 of 320 known R-L21* that have tested for 406S1 & 617 & 531 match you.   19 of those 44 were colonial type in origin so the true proportion of those matching this signature from Scotland is 13 out of 25 (of known European origin).   
Logged

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

Posts: 2963


WWW
« Reply #99 on: June 17, 2009, 11:45:20 PM »

The following list has not be deep clade R tested, although a number of them have tested to 67 markers.

My bet is that they are all R-L21* and they meet the 406S1=11/617=13 criteria for what some people call the "Ros Mascy Super Family Combo Group" (at least as I understand it.)    There are a lot of 19/394=15 and YCAIIa/b=18/23 on this list, which are both off-modal and both fairly slow moving.

RXYKH Walsh - Co. Kilkenny, Ireland/Wales
WTDAE Morgan - Ponty Pool, Wales
KRXND Evans - Merthyr, Wales
SD7EF Livingston - Scotland
7F56N Anderson/Wallentinus - Östergötland, Sweden
32J89 Welsh - Co. Kilkenny, Ireland
VGX5F Banks - Colonial US
JTK5M Banks - Colonial US
CPUNG Stephens - Colonial US
UQCFX Barrett - Colonial US
....
Mike, so are you in both Wales Modal 3 group AND the Ros Mascy Super Family Combo Group? If they aren't the same thing, can you explain how they differ?
 Did you ask Morgan and Evans to join the Wales_Cymru project? If not, I'll ask them. That way you won't be alone in your cluster LOL.
KRXND Evans did join the Wales DNA project.  WTDAE Morgan has not, but I just asked her again. 

Great news, though.  Both this Evans and Morgan have ordered deep clade R testing so we'll get to see if they are really R-L21*.

I wish I could get 32J89 Welsh to upgrade.  He responds to my emails but that's it.   His MDKA is from the same county as mine.  Even though our GD's aren't that close at 37, I think we must be of the same tribe or something because of the matching off-modal signature.
Logged

R1b-L21>L513(DF1)>L705.2
Pages: 1 2 3 [4] 5 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 0.135 seconds with 19 queries.