World Families Forums - R-L21(S145)* Haplotype Analysis - What do you think the clusters & stats mean?

Welcome, Guest. Please login or register.
September 20, 2014, 07:49:23 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(S145)* Haplotype Analysis - What do you think the clusters & stats mean?
« previous next »
Pages: [1] Go Down Print
Author Topic: R-L21(S145)* Haplotype Analysis - What do you think the clusters & stats mean?  (Read 2146 times)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« on: July 09, 2009, 10:01:21 PM »

I'm going to present for informational purposes some data on R-L21* (aka Ethnoancestry S145, ISOGG R1b1b2a1a2f* or FTDNA R1b1b2a1b5) haplotypes. I'll try to keep this thread updated as I have major updates.

A few notes on the process. I download all of the the confirmed R-L21* from FTDNA projects and YSearch that I can find (Many thanks to VinceT and RMS2 for their help.) This should include all of the R-L21* confirmed folks from the R-L21Plus FTDNA project at a minimum. I try to match up FTDNA kit #s and YSearch ID's so as to eliminate duplication. I also reduced (filtered haplotypes out) where there were perfect matches with the same surnames and apparent identical Most Distant Known Ancestors. I pretty much trust the data input by the haplotype owners. I also eliminate the R-L21* designated modal or identifiable artificial Ysearch ID's that I can.

I categorized in "Old World" geographic locations down to what a I call a "locality." I defined localities somewhat granularly where the current R-L21* data is dense (i.e. the Isles) and less granularly elsewhere, i.e France. Anything genealogically roadblocked in the "New World" I categorized as a unknown origin. These categories are not meant to be politically correct or balanced or anything like that. I just attempted to define the geography to a hopefully useful level. The categories can be changed easily.

I have good news in that the R-L21* folks are doing a good job of upgrading to 67 STR markers, a "must" I feel. It is hard to identify differentiated clusters without 67 markers. The number of confirmed R-L21* folks seems to be growing daily. It must be a large portion of the old R1b1c*.

I will speculate on possible clusters and I'm interested in any feedback as to the origin or nature of a cluster. A cluster, the way I describe it, is nothing more than a similar set of haplotypes that have a common "signature" of some off-modal markers, where off-modal means off the Western Atlantic Modal, which is essentially equivalent to R-L21*'s current modal. A cluster does not assure a close relationship in the same way an SNP can. Undoubtedly many clusters identified will "fall apart" with deeper clade testing in the future. Of course, any haplotype, without a known genealogical tie, may appear in a cluster inappropriately simply because of possible convergence STR markers over time. The implication to me is that it is normally of no value to search for clusters that cross known SNP marked sub-clades.

Caveat: I could easily have some errors, so please let me know of corrections or omissions.

Stay tuned... more to come.
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #1 on: July 09, 2009, 10:18:21 PM »

Here are the totals I have currently...

615 apparent unique haplotypes. 354 have 67 markers done.  Great job! as the 67 markers helps identify clusters and relationships.

Andorra - 1
Croatia - 1
Denmark - 1
England - 96
Finland - 3
France - 12
Germany - 20
Ireland (including N.Ireland) - 143
Isle of Man - 1
Italy - 3
Lithuania - 2
Netherlands - 2
Norway - 7
Scotland - 75
Spain - 8
Sweden - 4
Switzerland - 1
Wales - 25

The above doesn't add up to 615.  All of the rest have unknown Old World origins and are genealogically blocked in the New World.
« Last Edit: July 09, 2009, 10:43:14 PM by Mike » Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #2 on: July 10, 2009, 12:07:01 AM »

RMS2, you'll like this.

I've been looking for off-modal haplotype signatures in R-L21* but that neglects the people closest to the Western Atlantic Modal.

For grins, I went through the slowest markers looking for WAMH dead-on matches.  Of the 355 (justed added one) haplotypes with all 67 (FTDNA) markers, 75 of them were dead-on matches for the 28 slowest markers.

And now for the winners....

Here are the top 12 WAMH folks in R-L21*.  They match WAMH perfectly for the 36 slowest markers.

VT2R6    Beddoes? - Eng W Midlands
M6PHZ    Churchman - ENG East
PR7A7    Henderson - unknown
Q6YSK    Johnson - unknown
T9UTE    Kepler - Ger South & West
2MBF4    Morrison - unknown
HQ6GM    Morrison - Ire Ulster
Z6R4A    Morrison - unknown
6NPWB    Muir - Ire Ulster
HX9ZF    Stevens - unknown
JJ2PD    Uren - Eng South West
H2DAM    Wolken - Ger North
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
GoldenHind
Old Hand
****
Offline Offline

Posts: 731


« Reply #3 on: July 10, 2009, 03:37:01 PM »

RMS2, you'll like this.

I've been looking for off-modal haplotype signatures in R-L21* but that neglects the people closest to the Western Atlantic Modal.

For grins, I went through the slowest markers looking for WAMH dead-on matches.  Of the 355 (justed added one) haplotypes with all 67 (FTDNA) markers, 75 of them were dead-on matches for the 28 slowest markers.

And now for the winners....

Here are the top 12 WAMH folks in R-L21*.  They match WAMH perfectly for the 36 slowest markers.

VT2R6    Beddoes? - Eng W Midlands
M6PHZ    Churchman - ENG East
PR7A7    Henderson - unknown
Q6YSK    Johnson - unknown
T9UTE    Kepler - Ger South & West
2MBF4    Morrison - unknown
HQ6GM    Morrison - Ire Ulster
Z6R4A    Morrison - unknown
6NPWB    Muir - Ire Ulster
HX9ZF    Stevens - unknown
JJ2PD    Uren - Eng South West
H2DAM    Wolken - Ger North
Who do you suppose Stevens is?
Logged
vtilroe
Project Coordinator
Old Hand
*****
Offline Offline

Posts: 150


« Reply #4 on: July 10, 2009, 11:16:38 PM »

RMS2, you'll like this.

I've been looking for off-modal haplotype signatures in R-L21* but that neglects the people closest to the Western Atlantic Modal.

For grins, I went through the slowest markers looking for WAMH dead-on matches.  Of the 355 (justed added one) haplotypes with all 67 (FTDNA) markers, 75 of them were dead-on matches for the 28 slowest markers.

And now for the winners....

Here are the top 12 WAMH folks in R-L21*.  They match WAMH perfectly for the 36 slowest markers.

VT2R6    Beddoes? - Eng W Midlands
M6PHZ    Churchman - ENG East
PR7A7    Henderson - unknown
Q6YSK    Johnson - unknown
T9UTE    Kepler - Ger South & West
2MBF4    Morrison - unknown
HQ6GM    Morrison - Ire Ulster
Z6R4A    Morrison - unknown
6NPWB    Muir - Ire Ulster
HX9ZF    Stevens - unknown
JJ2PD    Uren - Eng South West
H2DAM    Wolken - Ger North
Who do you suppose Stevens is?
The "unknown" one? I believe he is one of those guys that lurks on genetic genealogy forums a lot.  Call it a hunch.
Logged

YSearch & MitoSearch: 2GXWW


yDNA: R-U106*


mtDNA: U5a1a1 (Genbank# GQ368895)


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

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #5 on: July 12, 2009, 05:43:29 PM »

Who do you suppose Stevens is?

Just a poor guy whose wife has been working him to death on the "honey-do" list for the last couple of weeks!

Almost done installing new flooring in the thatched hut.
Logged

Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #6 on: July 13, 2009, 01:24:23 AM »

I've been going through the STR markers, slowest to fastest, and playing with patterns of off-modal markers.   Here is one:

The R-L21* 426=13 GATAH4=11 460=10 Cluster.

DYS426 is very slow moving and only 2% of R-L21* has 426=13.  If this is "one" branch off of R-L21* it could have happened a long time ago.  Since there are couple of other markers that match off-modally I think this is a decent cluster. Also, two of these guys have a 3 above WAMH off-modal allele on 449.

EC6P9 Abernethy - Ire Ulster
AQJ6J Allen - Eng S&W
MBX33 Leonard - Ire Connacht
Y7QYG Werner - unknown

Any similarities in the heritages of the surnames? Does this look beginnings of a true subclade?
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #7 on: July 13, 2009, 01:26:13 AM »

Here is another possible cluster based off a very slow moving marker.

The R-L21* 578=10, 393=23, 442=13, 607=17,18 Cluster.

I like this one because 578 is very slow and besides the markers listed, there are some other similarities.

2FGYD Burns - Ire Ulster
XBCZD Erskine - Ire Ulster
VFQ39 McConnell - unknown
T2ZJ7 Svensson - Sweden/Baltic Sea*
*He's more on the Baltic Sea side than the Strait.

Two guys from Ulster, a Scottish surname and a Swede!   What does that mean?
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
GoldenHind
Old Hand
****
Offline Offline

Posts: 731


« Reply #8 on: July 13, 2009, 02:42:21 PM »

Who do you suppose Stevens is?

Just a poor guy whose wife has been working him to death on the "honey-do" list for the last couple of weeks!

Almost done installing new flooring in the thatched hut.
I hope the hard packed earth floor worked out OK for him.
Logged
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #9 on: July 13, 2009, 06:01:40 PM »

The R-L21* 531=12 YCAIIb=24,25 389i-ii=17,16 391=10 444=11 (GATAH4=12 413a=22) Scots "classic" type. The parenthesis just mean that these values are very common but not always present in this cluster.

I looked at the Scots Modal and tried to glean a set of off-modal (off WAMH) markers that would form a signature that applied to R-L21*. This is a pretty strong set of markers with a lot of commonality across these haplotypes.

Notice that there a couple of Scandinavians on the list, including up and in the middle of Sweden.

77P34 Anderson - Eng York
131060 Babson - Eng South West
W3WXS Buchanan - Sco Strathclyde
KJYA8 Buchanan - Sco Strathclyde
BUQ4H Campbell - Sco Strathclyde
RQMZ7 Campbell - Sco unk
4MNKD Enix - unknown
NSP69 Erickson - Sweden/Baltic Sea
J4WRB Eunson - Sco Isles
6AUEE Graham - unknown
UE75B Halligan - unknown
X69Z8 Hamilton - Sco unk
ATNT8 MacLahian - Sco Strathclyde
PYMQZ McDonald - Sco Strathclyde
G3ERH Mitchell - unknown
XDDA9 Neilson - Sco Strathclyde
4NU4A Norton - unknown
VVCYQ Rannekleiv - Norway
XFA7Y Roberts - Sco unk
3HWW9 Rock - Ire Ulster
DHU8U Rogers - unknown
S3NK2 Rogers - unknown
KG2FY Stanley - Sco unk
QA78V Stewart - Sco unk
MABKF Tagert - unknown
NZWQD Turner - Sco unk
6WPQA Watson - Sco Grampian
CSRF9 Young - Sco Strathclyde
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #10 on: July 24, 2009, 01:04:46 PM »

This is the R-L21* 531=12 YCAIIb=22 GATAH4=12 391=10 444=11 (440=30 413a=22) Scots variation cluster. This group almost, but doesn't quite fit with what I was calling the classic or mainline Scotts cluster and I'm not at all sure this group represents a Scottish origin although a couple of people in the group are.

The main difference here is a YCAIIb being less than WAMH versus greater than (in the main Scots group.) YCAII is slow moving so I think the 2-3 step difference is significant but I'm not proficient on palindrome markers.

9ZWK9   Arnold - Eng South West
MU26C   Hunt - unknown
VNRPW   MackMillion - unknown
KKVJF   MacMillan - Sco unk
4AKUT   McCurdy - Sco Isles
QDNJP   McDowell - Ire Ulster

Arnold and Hunt look a little different than the others, haplotype-wise (and I guess surname-wise as well.) I've also got these two in another potential cluster so I'm not positive they belong here although this signature has a large number of off-modal matches.

Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #11 on: July 24, 2009, 01:17:39 PM »

I'm copying this from another forum from Dennis Wright:

The Irish Type III cluster has proved to be R-L21*
DYS439=11, DYS459=8,9 DYS464=13,13,15,17 DYS456=16 DYS463=25 (FTDNA nomenclature) DYS557=15 DYS494=9 DYS533=13 DYS636=12 DYS714=25 DYS716=24

Tested R-L21+ or S145+
CD2C7 Bryan
6JN52 Bryant
QZ9GK Collier
3FYBQ Hogan
8AQ3Q Kennedy
ZBEBW King
DWB7A Mabe
ZHG9N Noonan
WC3H8 O'Connor
BNAHR Reilly
UFXCS Walker
6PWCD Wright
WVRPF Wright

I'm only looking at the FTDNA 67 markers, but I get this cut of Irish Type III with a slightly more limited criteria.  However,  I don't understand 456 though as I get 15 for these people, not 16.  Does Ethnoancestry have a different way of representing 456 vs FTDNA?  Since my criteria is less restrictive, I get a slightly larger group.

R-L21* 459a/b=8/9 439=11 456=15 (557=15) Irish T3 "plus" variation

16274   Anderson - Eng London
XZXCC   Bretheim - unknown
6JN52   Briant - unknown
QURQS   Brown - unknown
6JN52   Bryant - unknown
474NH   Casey - Ire Munster
QZ9GK   Collier - unknown
3FYBQ   Hogan - unknown
8AQ3Q   Kennedy - Ire Munster
37DRT   Noland - unknown
ZHG9N   Noonan - Ire Munster
WMBJR   O'Brien - Ire unk
WC3H8   O'Connor - Ire Munster
PFNY5   O'Hara - Ire unk
BNAHR   Reilly - Ire Munster
UFXCS   Walker - Eng Yorkshire
WVRPF   Wright - unknown
6PWCD   Wright - Ire unk

What do you think of the expanded group?
« Last Edit: July 24, 2009, 01:24:24 PM by Mike » Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #12 on: July 24, 2009, 01:32:59 PM »

The way I cut what I call R-L21* Irish T3+ (forgive me for misappropriating classic Irish Type III if that's what I'm doing) I end up with 2 clear sub-clusters

Sub-cluster A has the primary alleles for classic Irish Type II at 464.
R-L21* Irish T3+A  459a/b=8/9 439=11 456=15 & 464=13/13/15/17,18 (557=15)

16274   Anderson - Eng London
XZXCC   Bretheim - unknown
QURQS   Brown - unknown
474NH   Casey - Ire Munster
QZ9GK   Collier - unknown
3FYBQ   Hogan - unknown
8AQ3Q   Kennedy - Ire Munster
37DRT   Noland - unknown
ZHG9N   Noonan - Ire Munster
WMBJR   O'Brien - Ire unk
WC3H8   O'Connor - Ire Munster
UFXCS   Walker - Eng Yorkshire

Here is the other half.  Sub-Cluster B.

R-L21* Irish T3+B 459a/b=8/9 439=11 456=15 & 464=13/13/15/16 (557=15)

6JN52   Briant - unknown
6JN52   Bryant - unknown
PFNY5   O'Hara - Ire unk
WVRPF   Wright - unknown
6PWCD   Wright - Ire unk

I don't know, perhaps the differences at 464d are not that important and I'm splitting hairs.
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
Mike Walsh
Guru
*****
Offline Offline

Posts: 2964


WWW
« Reply #13 on: August 18, 2009, 04:38:50 PM »

Here are some general statistics on all of the people I have in the R-L21*. They are from the FTDNA R-L21Plus project and from Ysearch. I removed all of the duplicate MDKA haplotypes and all of the modal/artificial haplotypes that I could identify. We have:

689 apparently unique haplotypes. 658 of them are confirmed SNP L21* tested, 31 are "predicted".

430 have tested to 67 markers.

The modal for R-L21* matches exactly the Western Atlantic Modal for all 67 markers.

The median and average for R-L21* match the Western Atlantic Modal for 65 of 67 markers. The exceptions are:
449=30 (instead of 29)
CDYa=37 (instead of 36)
These are fast moving markers.

Keep in mind the testing rates vary widely from one country to the next, but of the people who know their old world MDKA,
 12% are from Western Continental Europe
 4% are from Scandinavia
 1% are from Eastern Europe
 22% are from England
 35% are from Ireland
 19% are from Scotland
 7% are from Wales

Honestly, I don't see hardly any differentiation for the haplotype modal, median or averages from one geography listed above to the next.   Eastern Europe is the exception but there are so few participants there I don't think that should be considered.

The only geographic differentiation of any consequence I can see is 391=10 is more frequent than 391=11 for both Scotland and Scandinavia.
Logged

R1b-L21>L513(DF1)>S6365>L705.2(&CTS11744,CTS6621)
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #14 on: August 25, 2009, 09:37:38 AM »

I found a few more men with that 406S1=11, 617=13 combination yesterday:

Z69DA - Hetlet (Norway)
6C3G6 - Merrill (original surname Conrardy; Luxembourg)
A8F7K - Gruetzmacher (unk origin)
TA9UH - St. Claire (France)

St. Claire actually has 406S1=12, but I think he's still in that combo cluster, since that is one move up from 11 and he has 617=13.

Thus far, I have heard from one of them, Hetlet. He has ordered the Deep Clade-R, so time will tell whether he is L21+ or not.
« Last Edit: August 25, 2009, 09:53:27 AM by rms2 » Logged

Pages: [1] 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 0.131 seconds with 18 queries.