CLAN DONNACHAIDH DNA NEWS

CLAN DONNACHAIDH DNA REPORT

2007 – No 1

2006 was a very exciting year for the Clan Donnachaidh DNA surname project. It grew by almost 100 participants and it is providing an impressive amount of information together with the advances in DNA research.

The DNA results can be seen on the Clan Donnachaidh International website: http://www.donnachaidhinternational.com/

The project and the Clan Donnachaidh Society

The project, which covers the names Robertson, Reid, Duncan Stark, Duncanson, and several others associated with Clan Donnachaidh, now has almost 300 participants. The project was set up in 2002 by Bill Robertson, the International Vice-Chairman of the Clan Donnachaidh Society, to cover the names associated with the clan.

Participants include society members and others who are interested in discovering more about their remote or more recent origins. There is no requirement to be a member of the clan society to take part but if you would like to know more about the society, please see the Clan Donnachaidh International website:

http://www.donnachaidhinternational.com/

If you would like to join the society you are very welcome (an application form can be found on the website home page).

If you would like to join the project you can enrol on the Family Tree DNA website at http://www.familytreedna.com/ under the project name Donnachaidh. Participants for the Y chromosome test must be men descended in the male line from one of the clan surnames (Donachie, Duncan, Duncanson, Reid, Robertson, Stark, some variants of these and several other names).

Project administrators

Tim Duncan MiTnacnuD[at]aol.com

Bill Robertson billrobertson[at]donnachaidhinternational.com

Stephanie Robertson dna[at]pt.lu

Results to date

At the time of writing (mid-January 2007) we have 246 sets of results representing a number of surnames and a wide range of results. There are now nearly 300 who have enrolled. There is no predominant result that can be said to represent the clan or even one of the surnames within the clan.

Among these results we have

138 different 12-marker haplotypes

157 different 25-marker haplotypes

136 different 37-marker haplotypes

However, the results are starting to coalesce. There are eight Duncan subgroups, one Reid subgroup, six Robertson subgroups and some larger Robertson and Reid groupings of similar haplotypes.

Some dedicated testing by people researching the name Duncan in the United States has produced some excellent matches that confirm documentary family research. Some Robertson families have been working along similar lines.

By selective testing the Duncan researchers have been able to identify which Duncan families that settled in the United States have a common origin. It has proved possible to link many of them, though the name and location of the common ancestor is still unknown.
Using DNA results to help in genealogical research

The results of Duncan Group A are used as an example here.

Certain basic principles are involved in genetic genealogy

• It looks at a timeframe over the period of surname development.

• Follows lineages by SURNAME

• Uses the Y-DNA test for male research

Results for the surname Duncan

• The Duncan participants are coalescing into eight defined subgroups.

• Over 20 men do not yet have DNA matches.

• The DNA results show that the Duncan surname developed along different male lines over the period of surname development.

The Duncan Group A results for 25 markers are shown below. It will be noted that, with a few variations, the results are very similar. This is the sort of result that is expected if the people concerned share a common ancestor within an historical timeframe. One interesting feature is that some members of this group have an additional marker at 464e.

Duncan Group A with 25-marker results

Through documentary research, the participants had traced their ancestry back to six different men. It was not known how these men were related.

Duncans in Walton County, Georgia

Genealogical question:

• Were the Duncan men living in Walton Co., Georgia during the time period 1820-1860 related?

– Do they share a common ancestor?

• Genealogical research has built a strong circumstantial case, though no documentary proof has been found, that George, Anderson Elbert and Brantly Duncan were brothers and the son of Nelson Duncan.

How can DNA testing help?

• By testing male descendants from each known Duncan line from Walton Co., Georgia, this will give a biological ‘fingerprint’ of each participant’s DNA.

• This DNA ‘fingerprint’ is called a haplotype.

• Compare the DNA test results from each descendant.

• Look for near identical DNA haplotype/signature.

Results of Walton Co., Georgia DNA testing:

• DNA testing shows that the male descendants of George, Anderson Elbert, and Brantly Duncan are indeed biologically related.

• DNA results show that these men share a common male ancestor.

• DNA validates the genealogical research that has been amassed on these Duncans.

With genetic testing, researchers had proof that all these lines were related, although documentary proof has not yet come to light.

Duncan researchers in the United States have tested a number of Duncan lines to see whether they are related. In many cases a relationship has been demonstrated, though it is not always known how and when the relationship occurred.

Using genetic testing for your family research

If you are interested in using DNA testing for family research:

Seek out men from families with the same surname who seem likely to be related to you (e.g. they come from the area historically associated with your family or they have other characteristics in common with your family, such as the same unusual forenames).
If you have already compiled a family tree, it could be useful to test different branches to obtain a collection of results to define the ancestral haplotype.
If not much money is available for testing, choose a 12-marker test to assess families that may be related. This will indicate whether there is likely to be a close relationship or not. An upgrade can always be ordered later if the results look interesting.

A large number of different haplotypes have emerged from the testing to date. Groups are starting to emerge. However, as yet we almost certainly do not have results from all lineages for all the surnames. If you are interested in using DNA results for genealogical research, we would encourage you, where possible, to identify families who are likely to be related to your family and approach them about testing.

Making the most of your results in other ways

Don’t forget to keep your contact details up to date. If you move or change your e-mail address, remember to correct your page on the FTDNA site.

You can help genetic research in general by uploading your results to the Y-search site. You can do this from your Y-DNA matches page, where you will find the link under the following text.

Additional possibilities for searching matches:

While our database is not open to the wide public for privacy reasons, Family Tree DNA has created Ysearch.org as a free public service so that people who have tested with the different companies can compare their results. You will be able to determine what portion of your personal information you want to disclose. Please note that a new user ID will be created for you and you will be asked to choose a new password. This new set of ID/Password is exclusive for Ysearch.org. You will also be given the opportunity to upload your GEDCOM if you have one.

Click here to upload to Ysearch.org

An overview of the results

As has been stated earlier, most Clan Donnachaidh results fall into Haplogroup R1b. This is the most common haplogroup in western European populations. It is believed to have expanded throughout Europe as humans recolonized after the last Ice Age. The latest research suggests that humans returned to Britain only about 11 500 years ago; the first arrivals almost certainly belonged to certain categories of Haplogroup R1b.

We also have quite a number of results in Haplogroup I. Haplogroup I spreads up through central Europe and represents about 10-15% of men in north-western Europe. We also have one or two sets of results in Haplogroups R1a, E and J.

Within the R1b results most participants come into certain broad categories. When defined by six markers the main categories are:

Principal clan surnames associated with haplotype	Haplo-group	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392	% of total
Duncan, Duncanson, Reid, Roberts, Robertson, Stark	R1b	13	24	14	11	12	13	19%
Duncan, Reid, Robertson	R1b	13	25	14	11	12	14	15%
Donachie, Duncan, Duncanson, Inches, Reid, Roberts, Robertson, Stark	R1b	13	24	14	10	12	13	11%
Reid, Robertson	I	13	22	14	10	14	11	6%
Duncan, Robertson	R1b	13	24	14	10	12	14	6%
Duncan, Reed, Robertson, Stark	R1b	13	23	14	10	12	13	5%
Duncan	R1b	13	26	14	10	12	13	4%
Reid, Robertson	R1b	13	23	14	11	12	13	3%
Reid, Robertson	R1b	13	25	14	11	12	13	3%
Duncan, Reid, Robertson	R1b	12	24	14	10	12	13	2%

It will be seen that, even at six markers, there is no haplotype that represents the majority of the clan results. Consequently, so far there is no evidence of a genetic profile that represents a typical Duncan, Reid or Robertson. The highest number of results is found in the 13 24 14 11 12 13 category, which is the category that occurs most frequently in Britain and probably represents the earliest settlers.

Matches with other surnames (on 37 and 67 markers)

Most participants have achieved fairly close matches with other surnames on 12 and 25 markers. There are some who have matches with other surnames at 37 and in a few cases even 67 markers.

So far all the 37 and 67 matches occur among participants who come into certain categories of very frequently occurring haplotypes. They will be identified below, when the principal haplotypes are discussed in detail.

Trying to identify origins from matches with other surnames

Family Tree DNA points out in a section on Understanding matches with different surnames:

The range of generations for the common ancestor extends to 76.9 generations, or almost 2000 years for those cases where there is not a surname in common. Therefore the importance of a surname link is paramount to provide a comfortable conclusion of relatedness. Most of the time random matches with people with different surnames do not stand the test for extended DNA testing.

Some of these surname matches seem to come from particular areas (perhaps a number of Scottish or a number of Irish results).

Surname maps

The distribution of these names can be studied by means of the surname maps that are now available on the Internet. The most recent for Great Britain can be found on the Surname Profiler site:

http://www.spatial-literacy.org/UCLnames/

This was developed from a research project based at University College London.

If you click on Start a surname search at the top of the home page, you can see how the distribution of a surname has developed by comparing results in 1881 and 1998. If you click on the Geographical Location or Frequency and Ethnicity headings, you can find more information about the surname.

There does not seem to be an easily accessible Internet option for Ireland at present. There is a site that lists the main Irish surnames by county and barony but does not provide maps

http://www.rootsweb.com/~irlkik/ihm/

For charting the spread of names in the USA, the Hamrick site is easily available: http://www.hamrick.com/names/index.html

No matches (even on 12 markers)

At the other end of the spectrum, some people have no matches, even at 12 markers.

The people concerned tend to have more unusual results in that it is not possible to identify the haplogroup by analysing the haplotype. They come into the categories of less common haplotypes, which registered only one or two occurrences even in a 2003 six-marker survey in Britain and Ireland (see below). There are no matches for these results in the Clan Donnachaidh project, even at six markers.

Comparing the Clan Donnachaidh results with other results in Britain and Ireland

In May 2003 a team of scientists published the results of tests in 23 small towns in Great Britain among men whose paternal grandparents were born within a 20-mile radius of the central point[1]. These results were compared with similar samples from men in Norway (Bergen and Trondheim), Denmark together with neighbouring Schleswig-Holstein in Germany, the Basque country and two towns in Ireland. Castlerea in central Ireland was chosen because it had no known history of contact with Anglo-Saxon and Viking invaders. Rush, to the north of Dublin, is on the coast.

Because earlier research had indicated the similarity of Celtic and Basque Y chromosomes, the Basque samples were included to help provide a representation of the Y chromosome of the indigenous population of the British Isles. Similarly, the samples from northern Germany/Denmark and Norway were included to provide a comparison for groups whose ancestors came into Britain from the north and east.

Six markers were tested (DYS393, 390, 19, 391, 388, 392). The results provide an overview of the distribution of certain haplotypes in Britain. Even when only six markers are considered there is a marked variation in the frequency of results.

The most frequently occurring Clan Donnachaidh results have been compared with the results of this survey. Charts have been drawn up for the six-marker haplotypes concerned on the basis of the 2003 survey results, showing the percentage of the total sample in the particular localities.

The places where samples were taken are listed below, with their abbreviations.

(Chart and locations based on sample results from ‘A Y Chromosome Census of the British Isles’ (2003). The names above indicate which Clan Donnachaidh surnames are associated with this category of results.)

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	24	14	11	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category represents 18% of the total sample population in Britain and Ireland, 7% of the population in its lowest occurrence, rising to 20% and over in several areas of Britain. The result of the sample in the Basque country was over 30%. This category includes the short form of the Atlantic Modal Haplotype, which represents 2% of the men in western Europe.

The percentage of the total within the Clan Donnachaidh project is slightly higher, representing 19% of the results and a number of surnames.

Some members of this category have several 37-marker matches with other surnames. In some cases these matches are obviously Scottish names such as Maclean and Sinclair, others seem to be associated with other areas of Britain and even beyond. The chart shows that variations of this haplotype occur throughout Britain.

Curiously, other members of this category so far have no 37-marker matches, although the six-marker matches are well represented throughout Britain and it is much the most numerous of the six-marker categories, both in the Clan Donnachaidh project and in the population as a whole. Variations of this haplotype occur in very high frequencies throughout western Europe. Several participants who have Family Tree DNA Western Atlantic Modal Haplotype badges have no 37-marker matches with other surnames. A smaller number of WAMH participants do have 37-marker matches with other surnames. More analysis is required to see whether the 37-marker matches with other surnames fall into well-defined categories.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	25	14	11	12	14

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category covers people who match the Niall of the Nine Hostages haplotype and others who come close. It will be noted that the results from Castlerea in Ireland are strikingly high, whereas this combination was absent from the Basque sample and was found at only very low frequencies in northern Germany/Denmark and Norway.

The combination DYS390/391 = 25/11, particularly when combined with certain other markers, has been noted as being strongly associated with Ireland.

It will be noted that this category is not found throughout Britain and, outside Ireland, it is most strongly represented at between 4 and 8% in the Western Isles, Stonehaven, Pitlochry and Oban, also in the Isle of Man and Haverfordwest.

Within the Clan Donnachaidh results, this category has a number of 37 matches with other surnames, many Irish.

The proportion of participants who come into this category in the Clan Donnachaidh results is substantially higher than in the total sample population (15% of the Clan Donnachaidh result but only 2% of the total sample in Britain and Ireland). Robertsons, Reids and Duncans are all found in this category.

It will be interesting to see whether it appears to have a wider significance in the formation of the clan.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	24	14	10	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

Proportions in the sample taken in Britain and Ireland and in the Clan Donnachaidh project are identical at 11%

Within the Clan Donnachaidh results, members of this category have quite numerous 37-marker matches with other surnames. The matches are with various different surnames but they include a number of specifically Scottish surnames such as MacGregor, Ferguson, Buchanan.

It will be noted that the six-marker result represents over 25% of the sample taken in Pitlochry and over 20% of the sample taken in Stonehaven and Oban.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
I	13	22	14	10	14	11

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

Haplogroup I is represented in about 12% of the Clan Donnachaidh results. It has been associated with origins in Denmark and the adjacent area of Germany. In the Scottish sample it was found particularly in Orkney and Stonehaven, which perhaps is not surprising. It was absent from the Basque and Irish samples.

The 13 22 14 10 14 11 haplotype represented 3% of the total survey results in Britain and Ireland and 6% of the Clan Donnachaidh results.

The peaks in Scotland in the 2003 survey were in Orkney and Stonehaven.

Within the Clan Donnachaidh results, this category has a number of 37-marker matches with other surnames.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	24	14	10	12	14

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category represents only 1% of the 2003 survey results, but 6% of the Clan Donnachaidh results. Most of the results came from extensive testing among Duncans in North America, but a couple of Robertsons also come into this category. (The Duncan and Robertson results diverge on additional markers.)

Within the Clan Donnachaidh results, this category has registered no 37-marker matches with other surnames.

As can be seen from the chart, this result did not appear in many areas. The peak was in Orkney, with the second highest proportion in Scotland being found in Stonehaven. It will be noted that the result was not found in the continental sample.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	23	14	10	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category has about the same proportion of results in the 2003 survey and in the Clan Donnachaidh results. In the 2003 survey, it represented 4% of the sample in Britain and Ireland and it represents 5% among the Clan Donnachaidh results.

The survey reveals a presence of under or just over 5% in most of Britain with an interesting peak in the sample taken around Durness. The sample in Stonehaven was also higher than most.

Within the Clan Donnachaidh results, some participants in this category have 37-marker matches with other surnames; others have 37-marker matches only with their own surname.

-o0o-

13 26 14 10 12 13 (chart not available)

The next category (13 26 14 10 12 13) occurred only once in the 2003 survey – in Castlerea in Ireland. It represents 4% in the Clan Donnachaidh survey on the basis of a group of Duncans in the USA who are undoubtedly all related, although the common ancestor has not been identified.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	23	14	11	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

Comparison between the 2003 survey results and the Clan Donnachaidh results reveals the opposite of the comparison in some other categories. This result represents 9% of the survey total but only 3% of the Clan Donnachaidh total.

Despite the low result among Clan Donnachaidh participants so far, this haplotype represents nearly 10% of the sample in Pitlochry and about 7% of the sample in Stonehaven.

Not much information is available about possible 37-marker matches with other surnames within this category. The only result that could be checked reveals no such matches.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	13	25	14	11	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category represents 4% of the 2003 survey result and 3% of the Clan Donnachaidh result.

It will be noted that although the main peak is in Wales, there are secondary peaks in the Western Isles and Stonehaven.

Within the Clan Donnachaidh results, there are no 37-marker matches with other surnames.

Haplo.	DYS 393	DYS 390	DYS 19	DYS 391	DYS 388	DYS 392
R1b	12	24	14	10	12	13

Results from the continental areas from which samples were taken for comparison are on the far right: Basque country = blue; northern Germany/Denmark = green; Norway = yellow.

This category is small and was sparsely distributed throughout Britain in the 2003 survey. However, it represents 2% of the Clan Donnachaidh survey. In Scotland it was found near Durness, but it may well occur elsewhere.

This category shows a link with Denmark and the adjacent area of northern Germany.

No 37-marker matches with other surnames have been noted within the Clan Donnachaidh results.

-o0o-

Thank you for your interest in the Clan Donnachaidh DNA surname project. We look forward to another exciting year in the growth of our project and our knowledge of DNA for genealogy.

Return to DSA Home Page

[1] [1] A Y Chromosome Census of the British Isles. Current Biology, Volume 13, Issue 11, Pages 979-984, May 2003, C. Capelli, N. Redhead, J. Abernethy, F. Gratrix, J. Wilson, T. Moen, T. Hervig, M. Richards, M. Stumpf, P. Underhill (http://www.ucl.ac.uk/tcga/tcgapdf/capelli-CB-03.pdf).

*Code*	*Sample*
Shet	Shetland
Ork	Orkney
Dur	Durness
W.Isles	Western Isles
STH	Stonehaven
PTL	Pitlochry
Oban	Oban
MPT	Morpeth
PNT	Penrith
I.of Man	Isle of Man
York	York
Sow	Southwell
Utx	Uttoxeter
LDL	Llanidloes

*Code*	*Sample*
Llangefni	Llangefni
Rush	Rush
Castlerea	Castlerea
Norf	Norfolk
Hwf	Haverfordwest
Chip	Chippenham
FAV	Faversham
MDH	Midhurst
DCR	Dorchester
PNZ	Cornwall
Ch.Isles	Channel Islands
Bas	Basques
NG/D	Germany/Denmark
Norway	Norway