R1b1c7

A short note by

Iain Kennedy

Last update March 24^th, 2008

Every living man has a Y-chromosome inherited from his father. It is usually handed down as an exact copy but from time to time a benign mutation occurs – the wrong nucleotide base is selected. From these events come our genetic diversity. The mutations divide men up into groups, each one of which is defined by the mutation that spawned it. Several layers of grouping exist since within one mutated branch further mutations can occur.

Scientists have a nomenclature for these groups, which they call ‘haplogroups’ based on an alternating sequence of letters and numbers. The original haplogroup ‘A’ came into being in Africa; the commonest western European group is a sub-division of group ‘R’ called ‘R1b’. R1b and R1a split off from the parent R1 tens of thousands of years ago. There is much controversy about the date and time of these splits – whatever I were to write here as the current latest theory would quickly get outdated. Maybe it happened in the Zagros mountains in what is now Iran/Iraq.

How are we supposed to work out these locations? Scientists tell us we have to look at two parameters: preponderance and diversity. Simple logic may dictate that the group should occur most heavily where it originated. But since it was there before it was anywhere else, it is easy to see that it should also have evolved or mutated more there than anywhere else too since it has had more time to do so.

Normally we measure the diversity with a slightly different yardstick. Some of the genetic material is in the form of repeated ‘arrays’ of base letters eg GATGATGAT and so on. These repeats also suffer mutational error not in the base letters but in the number of times it is repeated. For example 3xGAT might get copied as 4xGAT and so forth. These mutations happen more often than the first kind I described and so work as a finer yardstick gradation.

Note that what we don’t want to do is use the same yardstick to define the group and measure its internal diversity! Whilst it is possible to use these ‘tandem repeats’ (or ‘STRs’) to estimate the groups which are defined by the base letter mutations, it would not be wise to then use the same yardstick to work out how varied that group is.

It can be useful to draw in haplogroup density maps over geographic ones. Sometimes an interesting pattern may emerge, like a gradient or ‘cline’. A group may be heaviest on one side of a landmass or continent and gradually decline as you cross it. You might think that there is a single and even simple explanation for these clines, but there isn’t.

For example the heaviest concentration of R1b is in western Ireland. Good grief, Ireland is the cradle of civilization! But didn’t we just suggest that R1b originated in the Middle East? Why isn’t it most concentrated there? Unfortunately the answer has already been indicated; there isn’t a single explanation for genetic cline.

Note that it would patently be absurd to study the migration of R1b by overlaying it on a map of Ireland, even if I hadn’t already mentioned the Zagros mountains. We want to see it in as wide a context as possible, subject to the budget of our study.

Further caution must be applied if we use calculations to work out the most recent common ancestor (TMRCA) of our genetic samples:

‘this is a thorny and highly contentious issue, on the cusp of population genetics and phylogenetics. The date of an MRCA is a purely genetic estimate and need not have any obvious correlate in the other prehistorical records. This date does not represent the timing of a migration that spread the chromosomes carrying those haplotypes … the TMRCA is typically much older than the population in which they are found’ (1).

In fact the origin in the Zagros mountains was determined in the first case by diversity measure – variation of STRs within the R1b haplogroup. It is currently thought that R1b spread out from there, eventually reaching western Ireland where it is most dominant; the dropping cline as we move back across Europe is due to replacement by other later haplogroups. If you compare the STR diversity of R1b in Ireland and Iraq the former will be much lower than the latter.

Let us now turn our attention to R1b1c7, defined by the mutation called ‘M222’ (the letters which name these mutations represent the scientist who first describes them and are nothing to do with the haplogroup nomenclature), of which I am a member. How do we work out where and when it originated? By following the exact same technique with the same caveats.

So we will NOT

(i) draw a map of a single country and start overlaying R1b1c7 distribution on it

(ii) estimate people to be R1b1c7 using STRs then measure the diversity and thus age of ‘R1b1c7-estimated’ using STRS too

(iii) only quote one possible interpretation for any geographic cline we find

As of late 2007, the nearest we have come to respectable analysis of R1b1c7 is by Sykes in his recent book ‘Blood of the Isles’. He at least took samples throughout Scotland and compared them with customer data of his from Ireland. The analysis by Moore et al. (sometimes referred to as ‘McEvoy’) [6] is poor for a number of reasons; his hasty conclusion at the end that there is a link with the semi-historical figure of Niall of the Nine Hostages, coupled with falling back on Capelli’s ultra-low resolution Scottish data. Capelli had earlier sampled Scotland at various points on a grid he imposed on a map of the country but decided to omit the Scottish Highlands altogether. Analysis by amateurs is flawed by self-selection and a determination to interpret everything through an Irish prism, and the constant claims that the group only exists in lowland Scotland. The lone voice of reason amongst amateurs is John McLaughlin of the United States, who has pointed out the lack of a logical explanation of how R1b1c7 could get from Donegal to SW Scotland without also appearing heavily in Ulster, where it is less prominent; or why Irish and/or Scottish Dal Riata should share a genetic motif with the Ui Neill when there is no pedigree link between them. None of the ‘experts’ have ever responded to this, but are content to paint their results tables a lurid green colour as if everything is settled. It cannot be emphasized enough that a test that is now marketed as an ‘Irish’ DNA test will attract people who suspect or wish to verify they are from Ireland. All an attempt to plot their results in a table achieves is to demonstrate that a lot of Irish people have taken the test! The huge ranges quoted for the age of R1b1c7 show that it could far predate the Ui Neill and maybe go back as far as the first settlers in Ireland and Scotland.

In the meantime most of the research, scholarly or amateur, is concentrated on Ireland or on Scottish clans with an Irish origin tradition, who the same Irish researchers can look up in the traditional but often suspect Irish annals and other genealogies. This means that Scottish clans or families who lack such a tradition, like the Kennedys, must probably await lower level SNPs splitting M222 into sub-trees so that work can continue on such issues as how our ancestors moved around and between the lowlands and highlands. In the case of the Highland Kennedys which of course interests me most, the jury is still out on their origins, either surname-based or genetic.

Breaking news 24^th March 2008

As we await low level SNPs which will sub-divide R1b1c7, the next best thing may have arrived – a SNP directly above it, which will probe the deep history of the group. This SNP was revealed by comparing results of early testers at DecodeMe. Over the last few days interest in the SNP built up to such an extent that pioneering testing company Ethnoancestry, now a wholly UK-based organization, offered the individual test for just £20. I immediately signed up for it and will report the results back here as soon as I get them. It is expected that I will be positive (derived) for the mutation. The results may tell us more about when M222, the marker for R1b1c7, occurred, or at least set a limit on the date.

Probably of more use to the Kennedy study will be the results of our first tester with proven ancestry from Co. Donegal. Of course R1b1c7 is a minority even in its ‘heartland’ so it is by no means certain if he will be in this group. If he is, interpretation will need to proceed with great caution; it will be more important than ever to distinguish between connections before and after the adoption of the Kennedy surname. Hopefully my ongoing work in the Glasgow Poor Law records, which give the exact place of birth of each applicant and hold over 50 Kennedys from Co. Donegal, will give me a better idea of the impact Donegal has had on the Kennedys as a group.

References

(1) Human Evolutionary Genetics (Jobling, Hurles, Tyler-Smith, 2004) p182

(2) John McLaughlin’s incisive analysis of the problems with the current theories of the origins of R1b1c7 can be found here.

(3) Bryan Sykes’ book Blood of the Isles.

(4) The opening message from David Wilson’s R1b1c7 rootsweb mailing list immediately excludes anyone from the Scottish Highlands (and thus Dal Riata [modern day Argyllshire]!).

(5) Even Highland Scots get their results painted Shamrock Green on John McEwan’s results page

(6) A Y-chromosome signature of hegemony in Gaelic Ireland, American Journal of Human Genetics 78 (2006), L. Moore, B. McEvoy, E. Cape, K. Simms, D. Bradley

(7) Cenel Conaill and the Donegal Kingdoms, Brian Lacey, Dublin, 2006. A recent re-examination of the origins of Niall Noigiallach and the so-called ‘northern Ui Neill’.