ANALYSIS OF RESULTS
My intention is to attempt an analysis of all results posted in this project. The major limitations on any analysis are firstly the lack of SNP testing (that is, we don't know for sure the Haplogroup), and secondly the limited level of testing undertaken by many participants. The fact is that few people will achieve their aim of identifying relatives unless they undergo testing at the level of 67 or 111 markers. This is particularly true of the various groups assessed by FTDNA as R1b1a2, many of whom I guess to be R1b1a2a1a1b4, at this time also known as L21. But the absence of the SNP test severely limits my capacity to analyse their result. It should be understood that there are now many known sub-clades to L21, and knowledge of where a person fits among these sub-clades would be of great assistance to their progress.
Of those tested, a major group is those identified as R1b1a2a1a1b4b, alternatively known as M222, which is the SNP that identifies their Haplogroup. There are, as at November 2011, no known sub-clades of M222. However, as of December 2011, it is assumed that a newly discovered SNP, DF23, is ancestral to M222, arriving after L21. There are several significant aspects to the various Haplotypes seen in Group One. Firstly, all of these people are, with almost no chance of error, carriers of the M222 SNP, even though not all are tested for it. This is because all carry (so far as the respective tests go) a majority of the particular markers that differentiate M222 from its ancestral L21. These markers are: (DYS)390=25, 385b=13, 392=14, 448=18, 449=30, 464b=16, YCAllb=23, 607=16, 413a=21, 534=16, 481=25, 710=35, 714=24, 549=12, and 513=13. No other Haplogroup carries this combination.
The second aspect is that all clearly descend from one man. This is clear because all carry (the odd mutation excepted) a set of markers that are exceptional within the M222 Haplogroup but are uniform throughout this group. That is to say, the probability that all, or indeed any, of these men (or their ancestors) could independently simultaneously develop this combination of mutations is minuscule, and can be ignored for our purposes. The set of markers that identify the descendants of that man are: (DYS)458=18, 464=15-16-17-17, 576=18 (but see later), 570=18 (see later), 444=13, 533=12, Y-GATA-A10=14, 712=22, and 715=23. The latter four results are limited by the small number of test results, but are uniform in those tested at 111 markers. That man (the founder) is estimated, using conventional as well as DNA genealogy, to have lived about 650 years ago. He is the Most Recent Common Ancestor (MRCA) for this group. He is most likely the ancestor of the Griersons of Lag, and all who descend from them.
The third matter is the distinct difference in M222 between what I identify as the Scots cluster and the Irish cluster, which is sometimes known as the Niall cluster. These Group One results are not only dissimilar to the Irish cluster, but have certain similarities with other SW Scots families, to the extent that I have identified them as the Nith cluster. However, DYS444=13 is quite rare in M222, and doesn't appear in related families in Scotland, such that in this cluster I regard 444=13 as a Grierson identifier.
I have broken Group One into sub-groups so as to illustrate the relationships:
Group 01 A
represent the oldest strain among the descendants of the founder, and is also the group that seems to have sustainable but relatively recent Scottish records. As the oldest, it also exhibits, as would be expected, the greatest diversity. The representative of an ancestor for Groups One C and One D may ultimately be found within this group. It is also within this group that a non-DNA genealogical connection to the Lag family is found. One of the members of this Group carries the DF23 SNP, and it is assumed that all M222 do.
Group 01 B
is a branch, not in the line of descent, but carries the DYS389i=14 mutation seen among the further groups. However, it is identifiable by the additional pair of mutations 391=10 and 570=19 which do not carry forward. It is likely that Group One B members are recently related.
Group 01 C
has a MRCA within nearer times. They are united by 389i=14 and 576=19, as well as an indication that CDY is stabilising at 37-41. CDY is regarded as unstable, so a group carrying close results cannot be far apart in time. Member 199344 has two significant mutations at DYS 385 and 464, suggesting a branching over 200 years ago. 209737 has a RecLOH at DYS 425; this null return can be researched at http://en.wikipedia.org/wiki/RecLOH
, and the member can join another project if he wishes (subject to conditions), see http://www.familytreedna.com/public/null425/
. 57611 has had insufficient testing to be sure of the proper location in the chart. This group may represent the branch that migrated from Scotland to Ireland about 400 years ago.
Group 01 D
shows the probability of perhaps three family descents, with a much nearer MRCA probably among the Group One C cohort. It is united by 389i=14 (indicating a probable descent from Group One C), 576=20, and CDY almost completely stabilised at 37-41. I feel that this group could be the American strand, and would look for the MRCA to be near the time of the migration from Ireland.
Group Four is the so-called Viking group, because it is believed that in Scotland and Ireland the majority of Haplogroup I1 derives from the Viking incursions to those countries in the period 800-1100AD. That is, by virtue of, perhaps, rape and pillage, some "native" women were taken as wives or concubines by the invaders, children of those unions became part of the local clan, and in the fullness of time their descendants lost the knowledge of that period. They just lived as members of their family/clan, and in due course took the surname that local families also adopted, perhaps hundreds of years later. There are other possibilities, for example, the earlier period in England when a considerable part of the country was ruled by Denmark (Danelaw).
Professor Ken Nordtvedt, an acknowledged expert researcher in the I Haplogroup, has this to say about Group Four:
"They could have come over from Denmark or from one of the post-Roman Germanic populations. Or, of course, they could have come over later with Danes or Norwegians, etc. Or they could be a form started after being in the British Isles."
In my opinion, the number of mutations is low enough to suggest that these men are related more closely than the earlier invasions would allow. I favour the Viking raids of about 1000 years ago as being the Scandinavian source for this group. However, there are two clear groups here, differentiated by the DYS389i mutation from 12 to 13. Remember that 389 has special qualities, and the change from 12-28 to 13-29 reflects just the one mutation. But that may well have happened after the original incursion, and so a member of Group 04 A might be ancestral to Group 04 B. However, I've shown them separately as they could also represent two different visits from related Vikings.
R1b1a2 SNP untested
Unless results for an individual are predicted or proven to be L21+ by FTDNA, a person cannot join the L21 project. But, I believe that to gain any value from YDNA test results, the person should be in the most advanced project dealing with their haplogroup. The bottom line, therefore, is to find out what your haplogroup is, and then (in my opinion) join the L21+ project.
Although very likely L21, this group is relatively remote from any seen in the L21 project. A deep clade test by one member would have significant value in determining ancestry.
We see a very rare combination here with 23 at DYS 290 combined with 9-10 at DYS 459. The only way to advance now is to take a deep clade test.
Member 57019 has a very similar haplotype to a tested R1b1a2a1a1b4g member of the L21 Plus project. This haplogroup is that of the L193 SNP, which is downstream of the L513 SNP (both very recently discovered). I would suggest a test for L193 and L513 as a short cut to identifying the member's haplogroup. The L21 Plus member has a 230 year old Northern Irish heritage, identifying Fermanagh
as a location. Member N97315 has been provisionally added to Group 9 subject to further testing.
These members have similarities to several members of the L21 project who carry the L513 SNP.
These members have similarities to several members of the L21 project who carry the L193 SNP.
I am unable to identify any near relations in L21 and therefore recommend a deep clade test.
J David Grierson 23 June 2012