PDA

View Full Version : Irish DNA Atlas, Preliminary Results



Pages : 1 2 3 4 5 6 [7]

Finn
09-11-2018, 05:22 PM
As long as Jessie (pretty Irish) and I (pretty North Dutch) resemble so much in much PCA's and admixtures etc.....there are still a lot of uncleared issues.

msmarjoribanks
09-11-2018, 05:49 PM
I mentioned before that I couldn't recall the average distances within Ireland vs. between Ireland and England. This is helpful, from the Irish DNA Atlas:

"...genetic differentiation across Ireland and Britain is subtle, with the greatest genetic distances between Orcadian and non-Orcadian clusters (mean Fst = 0.0032). Ulster appears to be an outlier relative to the other ‘Gaelic’ Irish clusters, consistent with its position in PCA (Supplementary Fig. 2). The Gaelic clusters exhibit fine differentiation between each other (average Fst = 0.00030; average Fst excluding outlier Ulster = 0.00024) which is comparable to the differentiation we see between English clusters (average Fst = 0.00031; average Fst excluding outlier Cornwall I = 0.00024). This level of differentiation is finer than what we observe within Wales (average Fst = 0.00138), or Scotland (average Fst = 0.00250). The level of differentiation we observe in the island of Ireland (Fst = 0.0003), Gaelic and N Ireland clusters included, is almost an order of magnitude smaller than what we observe within clusters found across Great Britain, excluding Orkney (Fst = 0.00135)."

Nqp15hhu
09-11-2018, 08:51 PM
What does that imply?

Nqp15hhu
09-12-2018, 10:07 AM
Thanks for your reply Kevin.

I think there had to be some degree of intermarriage between Protestants and Catholics in N. Ireland, or Ed Gilbert with the Irish DNA Atlas study wouldn't have mentioned the 3 groups that had half Irish half British ancestry which he says happened some time in the 17th or 18th century.

Church records in Northern Ireland, in general, are rare for both Presbyterians and Catholics before the early 1800s, with the Catholic ones being generally later than the Presbyterian ones. Here's the link to PRONI (Public Record Office of Northern Ireland) marriage records.

https://www.nidirect.gov.uk/sites/default/files/publications/Guide_to_church_records.pdf

So there could be at least 200 years where mixed marriages took place with no record of it.

In "Presbyterians and the Irish Language" by Roger Blaney, on pages 16 and 17, he mentioned where some of the Gaelic Catholic Irish converted to Presbyterianism. He mentions some of the surnames from the Templepatrick Presybterian Church (County Antrim) session book from 1646 to 1744. Not only were there a lot of Irish surnames, there were a lot that had the O'! He lists some of the Irish surnames found there. In other instances Blaney speaks of, on page 17, over time, the O' was dropped. In another study of Saintfield Presbyterian Church, founded in 1658, there were found many instances of Irish surnames. It was pointed out that many more persons of Gaelic Irish descent could have had their name subsumed by anglicized surnames on pages 17,18.

So, if someone whose ancestors were Catholic Gaelic Irish, but somewhere down the line, they anglicized their name to where it didn't look Irish, and they no longer spoke Gaelic, and they were Protestant, then how could another Protestant distinguish them from those whose ancestors came from the Scottish Lowlands or England?

My late O'Hair uncle is a fairly strong autosomal DNA match to a Patterson who lives in County Down, N. Ireland. My ancestor Michael O'Hair b. 1749 came from County Down. This Patterson has an O'Hare great grandmother, who was Catholic, but who married a Protestant. Miss Patterson said her family has several mixed marriages and she also said her name Patterson, was Ulster Scot. Miss Patterson is also an autosomal match to an O'Hare from County Down, who appears to have about 3 Ulster Scot, or Scottish surnames in his ancestry.

I was taken aback a little by this, as I thought there was so much segregation in Northern Ireland that these admixture events would be more rare. So, I think mixed marriages were by no means often, but I think that over time they happened enough for the autosomal DNA of the 3 British/Irish clusters to reflect it. Also, I think the great majority of the Catholic Irish remained Catholic. In another book I have, I have read of a few Scottish lowland families that were Catholic that came to N. Ireland during the Plantations. They may have been from the Scottish/English border.

As far as YDNA, there are many Irish surnames and many Scottish surnames that are M222>S588. I think there was much back and forth movement between Scotland and Ireland in early Medieval times, as well as before that, and later. I myself am M222>S588>S603 and there are some Scottish surnames that are S603 as well, although it seems like there are much more Irish names that are S603.

There are Ulster Scot Ewings whose ancestors came to N. Ireland (I think probably came in during the Plantations), who are confirmed S603. Also, I have read of other people who have Ulster Scot ancestry and are M222.

This is a very interesting subject. Blaney also names some Presbyterian ministers who had the O' prefix in their Irish surname.

Kind Regards

I agree with you. Though I have to question the contribution a distant converted ancestor contributes to their ancestry? Say you are a Protestant and your GGGG granny was Catholic, surely that isn’t going to contribute much to your gene pool? No?

The intermixing must’ve came from more than one convert.

Nqp15hhu
09-12-2018, 10:13 AM
I also wonder if some of the Northern Ireland 2 group may represent "natives" in Ulster.
The Northern Ireland 1 group only shows up in Dublin and around Bantry:noidea: on the map, but there was something written about them mentioning 1 tester being from southern Ireland. Does that symbol represent only 1 person?

The Northern Ireland 3 cluster only shows up around Derry, Belfast and Dublin and mostly in N England with a trend toward Cumbria.

Meanwhile there seems to be a lot of the Dublin cluster in Northern Ireland, from the Route, Belfast and in N Tyrone, it looks like.
Could it also possibly just be unrelated migrations/intermixing of people with similar backgrounds?

Some Protestants in NI have Anglo Irish ancestry from the Republic.

Nqp15hhu
09-12-2018, 10:16 AM
Was anyone else besides me surprised that all three of the Planter clusters in Northern Ireland had substantial amounts of Gaelic Irish ancestry? The authors of the Irish DNA Atlas article mention their techniques for discovering this in the electronic supplementary material.

https://static-content.springer.com/esm/art%3A10.1038%2Fs41598-017-17124-4/MediaObjects/41598_2017_17124_MOESM1_ESM.pdf

However, in the next to the last paragraph under the "Discussion" section, the authors also say or indicate that some of this admixture could be caused by migrations before the 17th and 18th Centuries A.D. (before the Ulster Plantations) from Scotland to Ulster or vice versa. They specifically mention the migration of the Scottish Gallowglass mercenaries and numerous Redshanks (from about the 13th Century A. D. to about 1600 A. D.).

So, I believe that most of this admixture happened some time during the 17th and 18th centuries A. D. (like their calculations indicate), with some of it occurring during the Gallowglass migration into Ireland, also from some of the Normans or their retinue (although there weren't large numbers of Normans in Ulster), and some of it as far back as the kingdom of Dal Riada.

In the 13th century, Alan and Thomas of Galloway had some lands in Ulster, but I don't know if they were there long enough for their colonies to have any genetic input. They were said to be descendants of Fergus, a ruler of Galloway.

Kind Regards

Not really no.

- Protestants in NI have been cut off the GB gene pool for a few hundred years now.
- Flings and other endeavors go unknown.
- I have a few native Irish names in my tree.
- In my county there are a lot of Protestant: O’Niell, Mullan, McLaughlin and Dohertys.

MacUalraig
09-12-2018, 12:28 PM
Not really no.

- Protestants in NI have been cut off the GB gene pool for a few hundred years now.


Glasgow is full of people from NI and beyond (Donegal especially). Some of them go back, you can verify that for yourself in the Irish census. My Kennedy study is chockablock with em.

Nqp15hhu
09-12-2018, 02:07 PM
That was an Irish migration.

msmarjoribanks
09-12-2018, 03:24 PM
A few thoughts:

The Irish DNA Atlas is a specialized population, with all great-grandparents born within 30 km (or 50 km in less dense areas) of each other. I'm curious how much of a percentage of the overall Irish population would fit those criteria. Presumably it depends on where you are (likely less common in Dublin, for example).

The POBI was less restrictive, focusing only on where grandparents were born, and one issue they had with including Ireland is that they couldn't sort out whether similarities were due to Irish movement to England/Scotland or English/Scottish movement to Ireland. Since they wanted to try to pick out outside contributions to the population (which from the continent mostly went one way, they considered emigration from the British Isles to the continent and decided it would not have had a significant effect), including Ireland where the movement was both ways was difficult.

The three N Ireland clusters are all mixed, but they are not all primarily found in Ulster. Specifically, the cluster called N Ireland I consists of 7 Irish and 26 English; N Ireland II consists of 53 Irish, 19 Scottish, and 22 English; and the cluster called N Ireland III consists of 28 Irish, 9 Scottish, and 1 English. The Irish in N Ireland II are found in Ulster, whereas the Irish in N Ireland I and III are found both in Ulster and Dublin (with one from N Ireland I in southern Ireland). The people in these clusters have 25x the number of Scottish surnames and 6x the number of English surnames as the proportion found in the other Irish clusters.

Based on the youtube video I watched given by a grad student who had worked on the study, N Ireland II was looked at for the timing of the mixture, and the evidence suggested both a significant event around the time of the Plantations and earlier mixing, which is consistent with the idea that there is longstanding movement back and forth in that area and connections between the populations.

I think it's a mistake to assume that N Ireland I-III = Protestant, they don't seem to have taken that information, although I would imagine that the percentage of Catholics in the Ulster cluster is higher than the percentage of Catholics in the N Ireland II cluster.

fridurich
09-13-2018, 03:14 AM
I agree with you. Though I have to question the contribution a distant converted ancestor contributes to their ancestry? Say you are a Protestant and your GGGG granny was Catholic, surely that isnÂ’t going to contribute much to your gene pool? No?

The intermixing mustÂ’ve came from more than one convert.

I agree with you. If a Protestant in Northern Ireland today had one Catholic gggg grandparent then they would only have about 1/64th of the Dna of the Catholic ancestor. Because of the random nature of recombination they may have even less of that ancestorÂ’s Dna, or they could have more. My understanding of the techniques used by the Irish Dna Atlas Project is that they were able to use a very fine tooth comb to detect very small autosomal Dna differences. However, I think most or many of the people in the Planter clusters Northern Ireland One, Two, and Three have multiple Catholic ancestors in addition to their obvious Protestant ones. Can you imagine how many ancestors one has, going back 9-12 generations? So, just like people everywhere, the descendants of the Planters would have many ancestors living in the 17th or 18th centuries, which would increase the chances for Irish admixture. The same in reverse, for the descendants of the Gaelic Irish, the number of ancestors 9-12 generations back would be large , increasing their chance for admixture with the Planters.

Kind Regards

MacEochaidh
09-13-2018, 12:02 PM
A few thoughts:

The Irish DNA Atlas is a specialized population, with all great-grandparents born within 30 km (or 50 km in less dense areas) of each other. I'm curious how much of a percentage of the overall Irish population would fit those criteria. Presumably it depends on where you are (likely less common in Dublin, for example).

The POBI was less restrictive, focusing only on where grandparents were born, and one issue they had with including Ireland is that they couldn't sort out whether similarities were due to Irish movement to England/Scotland or English/Scottish movement to Ireland. Since they wanted to try to pick out outside contributions to the population (which from the continent mostly went one way, they considered emigration from the British Isles to the continent and decided it would not have had a significant effect), including Ireland where the movement was both ways was difficult.

The three N Ireland clusters are all mixed, but they are not all primarily found in Ulster. Specifically, the cluster called N Ireland I consists of 7 Irish and 26 English; N Ireland II consists of 53 Irish, 19 Scottish, and 22 English; and the cluster called N Ireland III consists of 28 Irish, 9 Scottish, and 1 English. The Irish in N Ireland II are found in Ulster, whereas the Irish in N Ireland I and III are found both in Ulster and Dublin (with one from N Ireland I in southern Ireland). The people in these clusters have 25x the number of Scottish surnames and 6x the number of English surnames as the proportion found in the other Irish clusters.

Based on the youtube video I watched given by a grad student who had worked on the study, N Ireland II was looked at for the timing of the mixture, and the evidence suggested both a significant event around the time of the Plantations and earlier mixing, which is consistent with the idea that there is longstanding movement back and forth in that area and connections between the populations.

I think it's a mistake to assume that N Ireland I-III = Protestant, they don't seem to have taken that information, although I would imagine that the percentage of Catholics in the Ulster cluster is higher than the percentage of Catholics in the N Ireland II cluster.

My father and three of my grandparents were from Belfast. Four years ago DNA testing revealed that my father was adopted, so my tree is shaky at best. From what I can interpret I am a North of Ireland II, and on another map, NICS (North of Ireland, Cumbria, South Scotland). I am also 25% French Canadian through my maternal grandmother. Yesterday I received my AncestryDNA Ethnicity Update and it shows me as 39% Ireland and Scotland and 33% England and Wales, along with 28% France.

My Y-DNA line is Doherty, but most of my surnames on all three lines are Scots and/or English: Steele, Erskine, Fulton, Leighton, Fulton, McDonald, Martin, et. al. I was raised as a Catholic with strong ties to New Lodge and Ballymurphy, Befast.

MacEochaidh
09-13-2018, 12:03 PM
Duplicate.

Nqp15hhu
09-13-2018, 01:35 PM
What is your name on ancestry? My mum is from outside Derry.

MacEochaidh
09-13-2018, 03:09 PM
What is your name on ancestry? My mum is from outside Derry.

My name on AncestryDNA is Miles Kehoe Doherty. My great grand father and 2X great grandparents were from Eglinton and Tullyverry; Doherty and Steele.

Nqp15hhu
09-13-2018, 06:20 PM
No match unfortunately which is quite surprising as I have Doherty distantly and from Derry.

MacEochaidh
09-13-2018, 08:56 PM
No match unfortunately which is quite surprising as I have Doherty distantly and from Derry.

I believe Doherty is the most popular name in Derry, so there must be hundreds of unrelated lines. My Doherty Y-DNA matches indicate that my line may have been in Moville, Donegal prior to Derry, but since I found out my Dad was adopted, I'm a bit hesitant to go with others' family trees. Moville is just across Lough Foyle from Tullyverry, so it's possible. Also, my Doherty Y-DNA is DF23 > ZP149, upstream of M222, which is the overwhelming Haplogroup for Doherty men. It could be my Y line was something other than Doherty hundreds of years back. DF23 > ZP149 is a very small branch and most of the men are from Antrim.

CillKenny
09-14-2018, 02:20 PM
Interesting panel discussion in TCD at 5pm on 9 October including Jim Mallory, Dan Bradley, Lara Cassidy and Rowan McLoughlin. Subject is the origins of the Irish people. Hosted by the Centre for New Irish Studies. You have to register to attend

Dubhthach
09-14-2018, 03:18 PM
Interesting panel discussion in TCD at 5pm on 9 October including Jim Mallory, Dan Bradley, Lara Cassidy and Rowan McLoughlin. Subject is the origins of the Irish people. Hosted by the Centre for New Irish Studies. You have to register to attend

A quick google returns this:
https://www.tcd.ie/trinitylongroomhub/whats-on/details/event.php?eventid=129024758

Eventbrite:
https://www.eventbrite.ie/e/interdisciplinary-research-on-the-origins-of-the-irish-people-tickets-50267926722

I believe I might be able to attend this, I've got some hours built up from work, plus given I'm gonna be in Amsterdam the whole following week for Work I can probably swing a half day and go have a look.

alan
09-14-2018, 07:50 PM
In the big cities in NIreland, especially Belfast, the numbers of people with a surname that ‘mismatches’ expected religion is huge. I’ve frequently seen School registers where nearly half the class have such mismatches. The level of mixing in at least the urban parts of NI is grossly underestimated. I’ve actually stopped using the concept of planter and native for modern people because I can see that so many are clearly a mixture of both.
Not really no.

- Protestants in NI have been cut off the GB gene pool for a few hundred years now.
- Flings and other endeavors go unknown.
- I have a few native Irish names in my tree.
- In my county there are a lot of Protestant: O’Niell, Mullan, McLaughlin and Dohertys.

rms2
09-15-2018, 12:43 AM
Years ago, when I was a young policeman on a certain large urban police department which shall remain nameless, I can remember talking with my sergeant, Ray O'Beirne, whom I nearly worshiped. I can remember it was a gray, gloomy, rainy day. Sgt. O'Beirne said, "You and I are from the same neck of the woods. That's why I think we both feel at home on days like today". I may be off on the exact words, but that was basically it. And he was right.

As you were.

Sikeliot
09-15-2018, 01:27 AM
Someone I know with ancestry from SW Ireland (Kerry) received 100% Irish/Scottish on their results this week.

JonikW
09-15-2018, 09:09 AM
Someone I know with ancestry from SW Ireland (Kerry) received 100% Irish/Scottish on their results this week.

What test?

Jessie
09-15-2018, 10:54 AM
Someone I know with ancestry from SW Ireland (Kerry) received 100% Irish/Scottish on their results this week.

That's common with Ancestry now. Most Irish appear to be getting 100% Ireland and Scotland and have lost all their trace regions.

cilldara
09-15-2018, 11:30 AM
I went from 81% to 95% Irish on Ancestry. My British went from 11% to 5%.

JonikW
09-15-2018, 01:26 PM
Oh, I see, with Ancestry from Kerry. I just read it as with ancestry from Kerry and am not familiar with what all the testing companies offer by way of breakdown...

Nqp15hhu
09-15-2018, 04:46 PM
I went from 81% to 95% Irish on Ancestry. My British went from 11% to 5%.
Same here. Do you have known GB ancestry?

Just trying to work out what my 5% means.

Dubhthach
09-15-2018, 06:11 PM
Someone I know with ancestry from SW Ireland (Kerry) received 100% Irish/Scottish on their results this week.

I'm now 100% Ireland/Scotland in Ancestry (up from 89%), my father is 100% Ireland/Scotland and my mother is 98% Ireland/Scotland & 2% England/Wales & NW European

cilldara
09-16-2018, 04:28 PM
Same here. Do you have known GB ancestry?

Just trying to work out what my 5% means.

I have a few British names on my family tree. I can't place anyone from coming outside of my county, never mind outside of Ireland.

Dubhthach
09-16-2018, 04:51 PM
I have a few British names on my family tree. I can't place anyone from coming outside of my county, never mind outside of Ireland.

Well Kildare was part of the 'English Pale' after all (leaving aside the Earldom of Kildare) so it wouldn't surprise it's case of medieval/early-modern admixture.

https://upload.wikimedia.org/wikipedia/commons/thumb/2/2f/The_Pale_According_to_the_Statute_of_1488_edit.jpg/616px-The_Pale_According_to_the_Statute_of_1488_edit.jpg

Nqp15hhu
09-16-2018, 05:07 PM
I have a few British names on my family tree. I can't place anyone from coming outside of my county, never mind outside of Ireland.

I am the same (several) and trying to work out what this means.

fridurich
11-21-2018, 02:43 AM
Has anyone heard how close the Irish DNA Atlas project people are to publishing their results from testing and doing comparisons using their new samples? The new samples being 500 samples from mainland Scotland and 250 from the Scottish Western Isles and the Isle of Man.

Kind Regards

sktibo
11-22-2018, 05:21 AM
Has anyone heard how close the Irish DNA Atlas project people are to publishing their results from testing and doing comparisons using their new samples? The new samples being 500 samples from mainland Scotland and 250 from the Scottish Western Isles and the Isle of Man.

Kind Regards

Wow, they managed to get 500 from the mainland?! Very exciting.

fridurich
11-24-2018, 07:06 AM
Wow, they managed to get 500 from the mainland?! Very exciting.

Yes, it is exciting. I think the first mention of this new testing was in either the “How Celtic is Scotland” thread or the “Is Scotland More Like England or Ireland?” thread. Seems more like it was in the first one, and I may not have the threads named 100 percent correct. I think the source for this information is pretty reliable.

Kind Regards
Fred

fridurich
04-22-2019, 01:51 AM
Has anyone heard when Ed Gilbert, who was heavily involved in the Irish Dna Project, will publish the results concerning analysis of approximately 500 mainland Scotland autosomal Dna samples and 250 samples from the Western Isles of Scotland and the Isle of Man?

I think this project was first mentioned on 9-26-2018 on around page 12 in the “How Celtic is Scotland?” (or named something like that) thread.

I’m very interested in seeing how these new Scotland and Manx clusters compare to the 10 Irish clusters the IDA project found.

Kind Regards
Fred

Jessie
04-24-2019, 05:23 AM
Has anyone heard when Ed Gilbert, who was heavily involved in the Irish Dna Project, will publish the results concerning analysis of approximately 500 mainland Scotland autosomal Dna samples and 250 samples from the Western Isles of Scotland and the Isle of Man?

I think this project was first mentioned on 9-26-2018 on around page 12 in the “How Celtic is Scotland?” (or named something like that) thread.

I’m very interested in seeing how these new Scotland and Manx clusters compare to the 10 Irish clusters the IDA project found.

Kind Regards
Fred

I'm sure if there is anything that Heber will post.

fridurich
04-29-2019, 03:12 AM
I'm sure if there is anything that Heber will post.

Thanks Jessie! I have a tendency to get impatient sometimes. Seems like it took about 5 years for the IDA project to conclude. Surely, this project shouldn’t take near that long.

Kind Regards
Fred

Heber
09-03-2019, 11:05 AM
A study led by experts in human genetics at RCSI (Royal College of Surgeons in Ireland) and the University of Edinburgh has created the first comprehensive genomic analysis of Scotland.

The study, published in the current edition of the journal Proceedings of the National Academy of Science, has found strong genetic connections between the Scots and Norse Vikings, and sheds light on the Gaelic component to the Icelandic gene pool.

Researchers investigated the DNA of more than 2,500 individuals with extended ancestry from specific regions across Great Britain and Ireland, with a specific focus on Scotland. The new data from Scotland means this is the first time the genetic map of the UK and the Republic of Ireland can be seen in its entirety, researchers say.

The map reveals that Scotland is divided into at least six clusters of genetically similar individuals, who cluster together geographically – the Borders, the south-west, the north-east, the Hebrides, Orkney and Shetland. Some of these clusters, notably those linked with the south-west and Hebrides share particularly strong affinity for clusters of Irish ancestry.

These Scottish clusters show remarkably similar locations to Dark Age kingdoms such as Strathclyde in the south-west, Pictland in the north-east, and Gododdin in the south-east. The results suggest that these kingdoms may have maintained regional identities that extend to the present. The modern genetic landscape of Britain and Ireland described by the researchers also reflects splits in the early languages of the Isles: Q-Celtic (Scottish, Irish and Manx Gaelic) and P-Celtic (Welsh, Cumbric, Cornish, Old Brythonic and Pictish).

Shetland, an archipelago of approximately 100 islands, located between Norway and mainland Scotland, was found to harbour the largest proportion of Norwegian-related ancestry, a consequence of the Norse Viking migrations that began in the eighth century.

The study compared the genomes of ancient Gaels buried in Iceland to the modern genetic diversity of Britain and Ireland. The comparison showed that these ancient settlers in Iceland shared the greatest genetic affinity with those on the western Isles of Scotland and the North-West of Ireland.

The researchers were also able to analyse the county of Donegal in more detail than before, revealing it as the most genetically isolated region of Ireland observed to date. This isolation shows little evidence of the migrations that have impacted the rest of Ulster.

The study, ‘The Genetic Landscape of Scotland and the Isles’, was completed in collaboration with the University of Edinburgh, University of Bristol and the Genealogical Society of Ireland. Funding was provided by Science Foundation Ireland, the Scottish Funding Council, Wellcome Trust and the Medical Research Council UK.

Scottish DNA map

Commenting on the study, Professor of Human Genetics at the RCSI School of Pharmacy and Bimolecular Sciences and Deputy Director of FutureNeuro, the SFI Research Centre for Chronic and Rare Neurological Diseases, Gianpiero Cavalleri, said: “The discoveries made in this study illustrate from the perspective of DNA, the shared history of Britain, Ireland and other European regions. People are well aware of historical migrations between Scotland and Ireland but seeing this history come alive in the DNA is nonetheless remarkable.”

Professor Jim Wilson, from the University of Edinburgh’s Usher Institute and MRC Human Genetics Unit, said: “It is remarkable how long the shadows of Scotland’s Dark Age kingdoms are, given the massive increase in movement from the industrial revolution to the modern era. We believe this is largely due to the majority of people marrying locally and preserving their genetic identity.”

Dr Edmund Gilbert, the paper’s lead author from RCSI, said: “This work is important not only from the historical perspective, but also for helping understand the role of genetic variation in human disease. Understanding the fine scale genetic structure of a population helps researchers better separate disease-causing genetic variation from that which occurs naturally in the British and Irish populations, but has little or no impact on disease risk.”

The collaboration between RCSI scientists, their international network of experts, and the Irish Genealogical Society, provides an exciting example of how citizens can contribute to important scientific discoveries. The Irish DNA Atlas is an ongoing study. If you have ancestry from a specific part of Ireland and you are interested in participating, please contact Séamus O’Reilly from the Genealogical Association of Ireland via [email protected]

https://www.rcsi.com/dublin/news-and-events/news/news-article/2019/09/researchers-connect-irish-and-scottish-genetic-maps

Dubhthach
09-03-2019, 11:12 AM
https://www.rcsi.com/dublin/-/media/feature/news/dublin/2019/scottish-dna-map.jpg

Anyone have a link to the study: ‘The Genetic Landscape of Scotland and the Isles’

The bit about Donegal makes alot of sense given the linguistic history of the county over the last 200 years.

Dubhthach
09-03-2019, 11:36 AM
I can't find this on the PNAS site at all. Makes me wonder if it will be published tomorrow in their next issue (well online published?).

Heber
09-03-2019, 04:13 PM
https://www.pnas.org/content/pnas/early/2019/08/27/1904761116.full.pdf

The genetic landscape of Scotland and the Isles
Edmund Gilberta,b, Seamus O’Reillyc, Michael Merriganc, Darren McGettiganc, Veronique Vitartd, Peter K. Joshie, David W. Clarke, Harry Campbelle, Caroline Haywardd, Susan M. Ringf,g, Jean Goldingh, Stephanie Goodfellowi, Pau Navarrod, Shona M. Kerrd, Carmen Amadord, Archie Campbellj, Chris S. Haleyd,k, David J. Porteousj, Gianpiero L. Cavalleria,b,1, and James F. Wilsond,e,1,2
aSchool of Pharmacy and Molecular and Cellular Therapeutics, Royal College of Surgeons in Ireland, Dublin D02 YN77, Ireland; bFutureNeuro Research Centre, Royal College of Surgeons in Ireland, Dublin D02 YN77, Ireland; cGenealogical Society of Ireland, Dún Laoghaire, Co. Dublin A96 AD76, Ireland; dMedical Research Council Human Genetics Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland; eCentre for Global Health Research, Usher Institute, University of Edinburgh, Edinburgh EH8 9AG, Scotland; fBristol Bioresource Laboratories, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol BS8 2BN, United Kingdom; gMedical Research Council Integrative Epidemiology Unit at the University of Bristol, Bristol BS8 2BN, United Kingdom; hCentre for Academic Child Health, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol BS8 1NU, United Kingdom; iPrivate address, Isle of Man IM7 2EA, Isle of Man; jCentre for Genomic and Experimental Medicine, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland; and kThe Roslin Institute and Royal (Dick) School of Veterinary Sciences, University of Edinburgh, Edinburgh EH25 9RG, Scotland
Edited by Chris Tyler-Smith, The Wellcome Trust Sanger Institute, Oxford, United Kingdom, and accepted by Editorial Board Member Mary-Claire King July 29, 2019 (received for review March 20, 2019)


Britain and Ireland are known to show population genetic structure; however, large swathes of Scotland, in particular, have yet to be described. Delineating the structure and ancestry of these popula- tions will allow variant discovery efforts to focus efficiently on areas not represented in existing cohorts. Thus, we assembled genotype data for 2,554 individuals from across the entire archipelago with geographically restricted ancestry, and performed population struc- ture analyses and comparisons to ancient DNA. Extensive geographic structuring is revealed, from broad scales such as a NE to SW divide in mainland Scotland, through to the finest scale observed to date: across 3 km in the Northern Isles. Many genetic boundaries are consistent with Dark Age kingdoms of Gaels, Picts, Britons, and Norse. Populations in the Hebrides, the Highlands, Argyll, Donegal, and the Isle of Man show characteristics of isolation. We document a pole of Norwegian ancestry in the north of the archipelago (reaching 23 to 28% in Shetland) which complements previously described poles of Germanic ancestry in the east, and “Celtic” to the west. This modern genetic structure suggests a northwestern British or Irish source population for the ancient Gaels that contributed to the founding of Iceland. As rarer variants, often with larger effect sizes, become the focus of complex trait genetics, more diverse rural co- horts may be required to optimize discoveries in British and Irish populations and their considerable global diaspora.

Sikeliot
09-03-2019, 04:39 PM
Now Munster is similar to Connacht and Leinster, and Donegal is the outlier?

sktibo
09-04-2019, 01:29 AM
Now Munster is similar to Connacht and Leinster, and Donegal is the outlier?

It looks difficult to say with certainty. If we look at the new PCA from the recent study including Scotland (Below the map and spanning tree) we see the Donegal markers on one extremity of the part of the PCA Ireland inhabits and the Munster markers on the other extremity. Since they seem to flip back and fourth but both appear on the extremities on the PCA it looks to me like both are outliers, but one appears to be the stronger of the two depending on the study and perhaps the methods used?

Sikeliot
09-04-2019, 01:40 AM
It looks difficult to say with certainty. If we look at the new PCA from the recent study including Scotland (Below the map and spanning tree) we see the Donegal markers on one extremity of the part of the PCA Ireland inhabits and the Munster markers on the other extremity. Since they seem to flip back and fourth but both appear on the extremities on the PCA it looks to me like both are outliers, but one appears to be the stronger of the two depending on the study and perhaps the methods used?

I can see this being the case. With that said, Donegal (likely representative of pre-Planter period Ulster) definitely has less genetic affinity to England than does Munster, Leinster, or Connacht. But we seem to see, yes, that Donegal and Munster are the "extremes" and Leinster/Connacht on the other hand form a large cluster across central Ireland.

I do remember, though, one of the other 2 studies on Ireland confirming that one of the Ulster clusters, likely similar to Donegal here, having the least affinity to Britain.

So maybe Munster has the least Viking input, and Donegal the least British?

sktibo
09-04-2019, 01:48 AM
I can see this being the case. With that said, Donegal (likely representative of pre-Planter period Ulster) definitely has less genetic affinity to England than does Munster, Leinster, or Connacht. But we seem to see, yes, that Donegal and Munster are the "extremes" and Leinster/Connacht on the other hand form a large cluster across central Ireland.

I do remember, though, one of the other 2 studies on Ireland confirming that one of the Ulster clusters, likely similar to Donegal here, having the least affinity to Britain.

So maybe Munster has the least Viking input, and Donegal the least British?

Seems like a reasonable suggestion, although on this recent study Munster appears to have more Norway admixture than Donegal. It looks like they didn't split the central Munster (green circles on the IDA map) from the rest of the Munster markers in this recent study though. Could have something to do with it...

Sikeliot
09-04-2019, 01:56 AM
Seems like a reasonable suggestion, although on this recent study Munster appears to have more Norway admixture than Donegal.

This would make sense. What I think we are seeing though is that since the main thing differentiating English from Welsh ancestry is Anglo-Saxon input, so the ability to model the Donegal clusters as just Welsh with no English input, but requiring English input for all the other Irish clusters, shows an infusion of indirect Anglo-Saxon input brought by the English to Ireland, and this is what would be higher in Leinster and Connacht.

So the question is, are we seeing Donegal might actually have NO English ancestry at all?

Viking input being higher in Munster actually makes sense since Limerick, Cork, and Waterford were all founded by Vikings, and I was honestly surprised other studies did not come to the same conclusion.

pmokeefe
09-04-2019, 01:58 AM
Seems like a reasonable suggestion, although on this recent study Munster appears to have more Norway admixture than Donegal. It looks like they didn't split the central Munster (green circles on the IDA map) from the rest of the Munster markers in this recent study though. Could have something to do with it...

Here's the Irish section from Dataset S3 (https://www.pnas.org/highwire/filestream/885975/field_highwire_adjunct_files/3/pnas.1904761116.sd03.xlsx) of The genetic landscape of Scotland and the Isles (https://www.pnas.org/content/early/2019/08/27/1904761116).

Mean SOURCEFIND ancestry proportions for each tested cluster form each of the 6 reference populations.


TARGET
England
Wales
Scotland
Denmark
Sweden
Norway


N Ireland
0.490
0.070
0.400
0.001
0.010
0.030


Donegal 1
0.484
0.070
0.405
0.001
0.006
0.035


Donegal 2
0.444
0.058
0.453
0.001
0.012
0.032


C Ireland
0.677
0.109
0.161
0.001
0.013
0.039


Munster
0.606
0.161
0.170
0.003
0.012
0.047

Sikeliot
09-04-2019, 02:02 AM
Here's the Irish section from Dataset S3 (https://www.pnas.org/highwire/filestream/885975/field_highwire_adjunct_files/3/pnas.1904761116.sd03.xlsx) of The genetic landscape of Scotland and the Isles (https://www.pnas.org/content/early/2019/08/27/1904761116).

Mean SOURCEFIND ancestry proportions for each tested cluster form each of the 6 reference populations.


TARGET
England
Wales
Scotland
Denmark
Sweden
Norway


N Ireland
0.490
0.070
0.400
0.001
0.010
0.030


Donegal 1
0.484
0.070
0.405
0.001
0.006
0.035


Donegal 2
0.444
0.058
0.453
0.001
0.012
0.032


C Ireland
0.677
0.109
0.161
0.001
0.013
0.039


Munster
0.606
0.161
0.170
0.003
0.012
0.047




What might explain the Irish being better modeled with England than with Scotland here?

Sikeliot
09-04-2019, 02:07 AM
Seems like a reasonable suggestion, although on this recent study Munster appears to have more Norway admixture than Donegal. It looks like they didn't split the central Munster (green circles on the IDA map) from the rest of the Munster markers in this recent study though. Could have something to do with it...

I thought of an explanation... the part of Munster that is joining, in part, the Connacht/Leinster cluster is Clare. Clare was historically part of Connacht, and only later became part of Munster.

pmokeefe
09-04-2019, 02:08 AM
What might explain the Irish being better modeled with England than with Scotland here?

From that table at least, it looks like the main difference between the northern regions of Ireland (including Donegal) and the center/south is that the center/south is more English, and to a certain extent Welsh, while the north is more Scottish. Which makes a certain amount of sense, geographically speaking, no?

Sikeliot
09-04-2019, 02:13 AM
From that table at least, it looks like the main difference between the northern regions of Ireland (including Donegal) and the center/south is that the center/south is more English, and to a certain extent Welsh, while the north is more Scottish. Which makes a certain amount of sense, geographically speaking, no?

Yes. And what then makes Donegal an outlier, is the English contribution being lower. Most of the Northern Irish Planter-descended population came from Scotland, but Donegal may have some affinity to Scotland because of the constant migration from Ireland to Scotland and vice versa well before that. Norman migration could be responsible for much of the "English" in Connacht, Leinster, and Munster also.

I would personally read the Irish regions with higher Scottish/Welsh as more proportionally 'native' Irish versus of Norman, English, etc. descent.

sktibo
09-04-2019, 02:31 AM
Yes. And what then makes Donegal an outlier, is the English contribution being lower. Most of the Northern Irish Planter-descended population came from Scotland, but Donegal may have some affinity to Scotland because of the constant migration from Ireland to Scotland and vice versa well before that. Norman migration could be responsible for much of the "English" in Connacht, Leinster, and Munster also.

I would personally read the Irish regions with higher Scottish/Welsh as more proportionally 'native' Irish versus of Norman, English, etc. descent.

I can add here that the Northern Ireland and Western Scotland had close ties up until relatively recent times. When I used to study Scottish Gaelic our teacher would tell stories and a lot of the time these stories took place in Northern Ireland, he told us that people were constantly moving back and forth between these regions and that Northern Ireland was considered by some to be the center of the Gaelic speaking world - I believe I was told that was one of the reasons it was selected for the plantation. Quite a few Scottish Gaelic songs take place in Northern Ireland instead of Scotland, or at least that's how it appears to me and my experiences with this matter.

Sikeliot
09-04-2019, 02:35 AM
I can add here that the Northern Ireland and Western Scotland had close ties up until relatively recent times. When I used to study Scottish Gaelic our teacher would tell stories and a lot of the time these stories took place in Northern Ireland, he told us that people were constantly moving back and forth between these regions and that Northern Ireland was considered by some to be the center of the Gaelic speaking world - I believe I was told that was one of the reasons it was selected for the plantation. Quite a few Scottish Gaelic songs take place in Northern Ireland instead of Scotland, or at least that's how it appears to me and my experiences with this matter.

So what this could mean is that when Donegal matches with Scotland, it could be the Gaelic Scottish population that was continuous with northern Ireland throughout most of history and not those with higher Saxon input.

What I am seeing now is that when one of the studies described South Munster as an outlier, this does not mean "less mixed" or "more Irish" but could simply mean differing proportions of Viking and English mixture than what is present in Leinster or Connacht. I was interpreting that incorrectly for a while it seems.

Hence why the Insular Celtic paper simultaneously found South Munster to be an outlier, but it was the Gaelic Ulster cluster with the lowest genetic affinity to Britain.

Tomenable
09-04-2019, 03:15 AM
https://www.pnas.org/content/pnas/early/2019/08/27/1904761116.full.pdf

The genetic landscape of Scotland and the Isles
Edmund Gilberta,b, Seamus O’Reillyc, Michael Merriganc, Darren McGettiganc, Veronique Vitartd, Peter K. Joshie, David W. Clarke, Harry Campbelle, Caroline Haywardd, Susan M. Ringf,g, Jean Goldingh, Stephanie Goodfellowi, Pau Navarrod, Shona M. Kerrd, Carmen Amadord, Archie Campbellj, Chris S. Haleyd,k, David J. Porteousj, Gianpiero L. Cavalleria,b,1, and James F. Wilsond,e,1,2
aSchool of Pharmacy and Molecular and Cellular Therapeutics, Royal College of Surgeons in Ireland, Dublin D02 YN77, Ireland; bFutureNeuro Research Centre, Royal College of Surgeons in Ireland, Dublin D02 YN77, Ireland; cGenealogical Society of Ireland, Dún Laoghaire, Co. Dublin A96 AD76, Ireland; dMedical Research Council Human Genetics Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland; eCentre for Global Health Research, Usher Institute, University of Edinburgh, Edinburgh EH8 9AG, Scotland; fBristol Bioresource Laboratories, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol BS8 2BN, United Kingdom; gMedical Research Council Integrative Epidemiology Unit at the University of Bristol, Bristol BS8 2BN, United Kingdom; hCentre for Academic Child Health, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol BS8 1NU, United Kingdom; iPrivate address, Isle of Man IM7 2EA, Isle of Man; jCentre for Genomic and Experimental Medicine, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland; and kThe Roslin Institute and Royal (Dick) School of Veterinary Sciences, University of Edinburgh, Edinburgh EH25 9RG, Scotland
Edited by Chris Tyler-Smith, The Wellcome Trust Sanger Institute, Oxford, United Kingdom, and accepted by Editorial Board Member Mary-Claire King July 29, 2019 (received for review March 20, 2019)

Britain and Ireland are known to show population genetic structure; however, large swathes of Scotland, in particular, have yet to be described. Delineating the structure and ancestry of these popula- tions will allow variant discovery efforts to focus efficiently on areas not represented in existing cohorts. Thus, we assembled genotype data for 2,554 individuals from across the entire archipelago with geographically restricted ancestry, and performed population struc- ture analyses and comparisons to ancient DNA. Extensive geographic structuring is revealed, from broad scales such as a NE to SW divide in mainland Scotland, through to the finest scale observed to date: across 3 km in the Northern Isles. Many genetic boundaries are consistent with Dark Age kingdoms of Gaels, Picts, Britons, and Norse. Populations in the Hebrides, the Highlands, Argyll, Donegal, and the Isle of Man show characteristics of isolation. We document a pole of Norwegian ancestry in the north of the archipelago (reaching 23 to 28% in Shetland) which complements previously described poles of Germanic ancestry in the east, and “Celtic” to the west. This modern genetic structure suggests a northwestern British or Irish source population for the ancient Gaels that contributed to the founding of Iceland. As rarer variants, often with larger effect sizes, become the focus of complex trait genetics, more diverse rural co- horts may be required to optimize discoveries in British and Irish populations and their considerable global diaspora.

In this study they tested old people with all 4 grandparents born in the same area? So they reconstructed old genetic landscape.

If you test randomly modern Scottish people without checking their genealogy, most will have mixed ancestry from many regions.

There were large migrations of people between various regions of Scotland during the last 250 years.

The authors of "Scottish Population Statistics", Edinburgh 1952, divided Scotland into three areas:

Highland = 5/7 of Scotland (21330 square miles)
Central Belt = 1/7 of Scotland (4269 square miles)
Lowland = 1/7 of Scotland (4196 square miles)

Central = counties Ayr, Dunbarton, Lanark, Renfrew, Clackmannan, Stirling, the Lothians, Fife, the City of Dundee
Highland = everything north of Central
Lowland = everything south of Central

Population by region in 1755 (in thousands):

Total - 1265
Highland - 652 (51%)
Central - 464 (37%)
Lowland - 149 (11%)

Population by region in 1861 (in thousands):

Total - 3062
Highland - 1020 (33%)
Central - 1769 (58%)
Lowland - 273 (9%)

Population by region in 1871 (in thousands):

Total - 3360
Highland - 1041 (31%)
Central - 2047 (61%)
Lowland - 272 (8%)

Population by region in 1931 (in thousands):

Total - 4843
Highland - 980 (20%)
Central - 3612 (75%)
Lowland - 251 (5%)

Population by region in 1951 (in thousands):

Total - 5096
Highland - 1000 (20%)
Central - 3840 (75%)
Lowland - 256 (5%)

As you can see the share of Highlands in the total population declined from over 50% to just 20% in 200 years.

The share of Lowlands also declined. People both from the north and the south were migrating to Central Belt.

Tomenable
09-04-2019, 04:08 AM
So what this could mean is that when Donegal matches with Scotland, it could be the Gaelic Scottish population that was continuous with northern Ireland throughout most of history

In Eurogenes K15 an Irish person with all ancestry from County Donegal scores "West_Scottish" before Irish. From the new study, about County Donegal:

"The researchers were also able to analyse the county of Donegal in more detail than before, revealing it as the most genetically isolated region of Ireland observed to date. This isolation shows little evidence of the migrations that have impacted the rest of Ulster."

^^^
Here is an Irish with 100% ancestry from County Donegal in K36 Similtude map, very isolated indeed (no similarities over 85 except for Irish average itself):

https://i.imgur.com/Q78kwHE.png

^^^
And for comparison Irish 100% from County Donegal in Eurogenes K15:

Admix Results (sorted):

# Population Percent
1 North_Sea 36.44
2 Atlantic 35.98
3 Eastern_Euro 9.61
4 Baltic 7.79
5 West_Med 7.19
6 West_Asian 2.34
7 Amerindian 0.61
8 South_Asian 0.04

Single Population Sharing:

# Population (source) Distance
1 West_Scottish 5.52
2 Irish 5.78
3 Southeast_English 6.35
4 Orcadian 6.97
5 Southwest_English 7.69
6 Danish 8.21
7 North_Dutch 8.87
8 North_German 9.92
9 South_Dutch 11.49
10 West_Norwegian 12.02
11 Norwegian 12.23
12 Swedish 13.77
13 West_German 14.34
14 French 15.14
15 North_Swedish 15.91
16 East_German 18.17
17 Southwest_Finnish 19.77
18 Spanish_Cataluna 20.75
19 Austrian 21.18
20 Spanish_Castilla_Y_Leon 21.91

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 88.9% West_Scottish + 11.1% French_Basque @ 4.55
2 90.7% Irish + 9.3% French_Basque @ 5.18
3 93.8% West_Scottish + 6.2% Spanish_Aragon @ 5.31
4 84.7% Orcadian + 15.3% French_Basque @ 5.31
5 94.4% West_Scottish + 5.6% Southwest_French @ 5.37
6 95% West_Scottish + 5% Spanish_Castilla_La_Mancha @ 5.39
7 95.1% West_Scottish + 4.9% Spanish_Cantabria @ 5.41
8 95.7% West_Scottish + 4.3% Spanish_Valencia @ 5.43
9 97% West_Scottish + 3% Spanish_Andalucia @ 5.47
10 97.6% West_Scottish + 2.4% Spanish_Murcia @ 5.5
11 98% West_Scottish + 2% Spanish_Cataluna @ 5.51
12 98.2% West_Scottish + 1.8% Spanish_Castilla_Y_Leon @ 5.51
13 98.7% West_Scottish + 1.3% Spanish_Extremadura @ 5.51
14 100% West_Scottish + 0% Abhkasian @ 5.52
15 100% West_Scottish + 0% Adygei @ 5.52
16 100% West_Scottish + 0% Afghan_Hazara @ 5.52
17 100% West_Scottish + 0% Afghan_Pashtun @ 5.52
18 100% West_Scottish + 0% Afghan_Tadjik @ 5.52
19 100% West_Scottish + 0% Afghan_Turkmen @ 5.52
20 100% West_Scottish + 0% Afghan_Uzbeki @ 5.52

Sikeliot
09-04-2019, 04:14 AM
In Eurogenes K15 an Irish person with all ancestry from County Donegal scores "West_Scottish" before Irish. From the new study, about County Donegal:

"The researchers were also able to analyse the county of Donegal in more detail than before, revealing it as the most genetically isolated region of Ireland observed to date. This isolation shows little evidence of the migrations that have impacted the rest of Ulster."

^^^
Here is an Irish with 100% ancestry from County Donegal in K36 Similtude map, very isolated indeed (no similarities over 85 except for Irish average itself):

^^^
And for comparison Irish 100% from County Donegal in Eurogenes K15:



Compare to this person from Galway.. see the affinity to Netherlands, Denmark is much higher as is affinity to England.


# Population Percent
1 North_Sea 42.43
2 Atlantic 28.13
3 West_Med 8.91
4 Baltic 8.38
5 Eastern_Euro 6.82
6 West_Asian 2.78
7 South_Asian 1.23
8 Amerindian 1.2
9 Oceanian 0.08
10 Red_Sea 0.03

Single Population Sharing:

# Population (source) Distance
1 Orcadian 3.54
2 West_Scottish 5.97
3 North_Dutch 6.48
4 West_Norwegian 6.49
5 Irish 6.63
6 Southeast_English 6.82
7 Southwest_English 7.08
8 Danish 7.46
9 Norwegian 7.87
10 Swedish 9.74
11 North_German 10.37
12 West_German 11.11
13 South_Dutch 12.13
14 North_Swedish 13.58
15 French 15.11
16 East_German 18.09
17 Southwest_Finnish 20
18 Spanish_Galicia 22.35
19 Hungarian 22.45
20 Spanish_Cataluna 22.5

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 85.4% Orcadian + 14.6% West_Norwegian @ 3.41
2 99.5% Orcadian + 0.5% Karitiana @ 3.49
3 99.5% Orcadian + 0.5% Mayan @ 3.5
4 99.5% Orcadian + 0.5% Anzick-1 @ 3.5
5 99.5% Orcadian + 0.5% Pima @ 3.5
6 99.7% Orcadian + 0.3% North_Amerindian @ 3.53
7 99% Orcadian + 1% Norwegian @ 3.54
8 100% Orcadian + 0% Abhkasian @ 3.54
9 100% Orcadian + 0% Adygei @ 3.54
10 100% Orcadian + 0% Afghan_Hazara @ 3.54
11 100% Orcadian + 0% Afghan_Pashtun @ 3.54
12 100% Orcadian + 0% Afghan_Tadjik @ 3.54
13 100% Orcadian + 0% Afghan_Turkmen @ 3.54
14 100% Orcadian + 0% Afghan_Uzbeki @ 3.54
15 100% Orcadian + 0% Algerian @ 3.54
16 100% Orcadian + 0% Algerian_Jewish @ 3.54
17 100% Orcadian + 0% Altaian @ 3.54
18 100% Orcadian + 0% Armenian @ 3.54
19 100% Orcadian + 0% Ashkenazi @ 3.54
20 100% Orcadian + 0% Assyrian @ 3.54



https://i.imgur.com/CxU2YLb.png

Jessie
09-04-2019, 04:30 AM
I can see this being the case. With that said, Donegal (likely representative of pre-Planter period Ulster) definitely has less genetic affinity to England than does Munster, Leinster, or Connacht. But we seem to see, yes, that Donegal and Munster are the "extremes" and Leinster/Connacht on the other hand form a large cluster across central Ireland.

I do remember, though, one of the other 2 studies on Ireland confirming that one of the Ulster clusters, likely similar to Donegal here, having the least affinity to Britain.

So maybe Munster has the least Viking input, and Donegal the least British?

It would make sense to me that Munster has more Norse than Donegal as Munster had the Viking city of Limerick and Cork also had some Viking settlement. Munster also had some Plantations although lesser than Ulster and also Normans. Possibly Dubthach or some other Irish poster could add to this discussion. Donegal was always a bit remote but did get some Scottish settlement and Gallowglass.

Sikeliot
09-04-2019, 04:32 AM
It would make sense to me that Munster has more Norse than Donegal as Munster had the Viking city of Limerick and Cork also had some Viking settlement. Munster also had some Plantations although lesser than Ulster and also Normans. Possibly Dubthach or some other Irish poster could add to this discussion. Donegal was always a bit remote but did get some Scottish settlement and Gallowglass.

I would also imagine much of the ancestry being picked up as English or Anglo Saxon in both Connacht and Munster is from the Normans.

But as I mentioned I think we should interpret Munster as a possible outlier to not mean "less foreign admixture" but different ratios of it from Leinster/Connacht, while Donegal may actually be more natively "Gaelic" Irish.

Tomenable
09-04-2019, 04:43 AM
I would also imagine much of the ancestry being picked up as English or Anglo Saxon in both Connacht and Munster is from the Normans.

Not necessarily. This can be a signal of Pre-Germanic English-like ancestry (Non-Pictish).

They might be matching Pre-Anglo-Saxon alleles that are common in England but rare in Scotland (due to different types of Pre-Germanic populations).

Who says that in year, say - 500 BCE - populations living in Ireland were more closely related to those in Scotland than to those in England and Wales?

Elevated Anglo-Saxon ancestry can be confirmed if they also show similar affinity to Denmark, Schleswig, Netherlands, etc. In addition to modern England.

BTW, from this new study:

"We estimate that Norwegian (as well as Danish/Swedish) ancestry is also markedly low in Ireland (average 7%) compared with previous estimates (8, 9) (we explore this further in Discussion)." - this is much lower than some previous estimates.

Jessie
09-04-2019, 04:47 AM
In Eurogenes K15 an Irish person with all ancestry from County Donegal scores "West_Scottish" before Irish. From the new study, about County Donegal:

"The researchers were also able to analyse the county of Donegal in more detail than before, revealing it as the most genetically isolated region of Ireland observed to date. This isolation shows little evidence of the migrations that have impacted the rest of Ulster."

^^^
Here is an Irish with 100% ancestry from County Donegal in K36 Similtude map, very isolated indeed (no similarities over 85 except for Irish average itself):

https://i.imgur.com/Q78kwHE.png

^^^
And for comparison Irish 100% from County Donegal in Eurogenes K15:

Admix Results (sorted):

# Population Percent
1 North_Sea 36.44
2 Atlantic 35.98
3 Eastern_Euro 9.61
4 Baltic 7.79
5 West_Med 7.19
6 West_Asian 2.34
7 Amerindian 0.61
8 South_Asian 0.04

Single Population Sharing:

# Population (source) Distance
1 West_Scottish 5.52
2 Irish 5.78
3 Southeast_English 6.35
4 Orcadian 6.97
5 Southwest_English 7.69
6 Danish 8.21
7 North_Dutch 8.87
8 North_German 9.92
9 South_Dutch 11.49
10 West_Norwegian 12.02
11 Norwegian 12.23
12 Swedish 13.77
13 West_German 14.34
14 French 15.14
15 North_Swedish 15.91
16 East_German 18.17
17 Southwest_Finnish 19.77
18 Spanish_Cataluna 20.75
19 Austrian 21.18
20 Spanish_Castilla_Y_Leon 21.91

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 88.9% West_Scottish + 11.1% French_Basque @ 4.55
2 90.7% Irish + 9.3% French_Basque @ 5.18
3 93.8% West_Scottish + 6.2% Spanish_Aragon @ 5.31
4 84.7% Orcadian + 15.3% French_Basque @ 5.31
5 94.4% West_Scottish + 5.6% Southwest_French @ 5.37
6 95% West_Scottish + 5% Spanish_Castilla_La_Mancha @ 5.39
7 95.1% West_Scottish + 4.9% Spanish_Cantabria @ 5.41
8 95.7% West_Scottish + 4.3% Spanish_Valencia @ 5.43
9 97% West_Scottish + 3% Spanish_Andalucia @ 5.47
10 97.6% West_Scottish + 2.4% Spanish_Murcia @ 5.5
11 98% West_Scottish + 2% Spanish_Cataluna @ 5.51
12 98.2% West_Scottish + 1.8% Spanish_Castilla_Y_Leon @ 5.51
13 98.7% West_Scottish + 1.3% Spanish_Extremadura @ 5.51
14 100% West_Scottish + 0% Abhkasian @ 5.52
15 100% West_Scottish + 0% Adygei @ 5.52
16 100% West_Scottish + 0% Afghan_Hazara @ 5.52
17 100% West_Scottish + 0% Afghan_Pashtun @ 5.52
18 100% West_Scottish + 0% Afghan_Tadjik @ 5.52
19 100% West_Scottish + 0% Afghan_Turkmen @ 5.52
20 100% West_Scottish + 0% Afghan_Uzbeki @ 5.52

Interesting as the Baltic is about 4% lower than mine and there Atlantic is about 5% higher. As most people would know my ancestry is North Munster and Sligo/Roscommon.

Admix Results (sorted):

# Population Percent
1 North_Sea 36.96
2 Atlantic 29.87
3 Baltic 11.92
4 Eastern_Euro 8.78
5 West_Med 5.03
6 West_Asian 4.95
7 Red_Sea 1.35
8 Amerindian 1.09
9 Siberian 0.05

Single Population Sharing:

# Population (source) Distance
1 Irish 2.93
2 West_Scottish 3.09
3 North_Dutch 3.63
4 Danish 3.85
5 Orcadian 4.93
6 Southeast_English 4.97
7 North_German 5.04
8 Southwest_English 6.62
9 Norwegian 7.2
10 West_Norwegian 7.49
11 Swedish 7.98
12 South_Dutch 9.66
13 West_German 10.77
14 North_Swedish 10.9
15 East_German 14.14
16 French 14.36
17 Southwest_Finnish 15.47
18 Austrian 18.02
19 Hungarian 18.72
20 Finnish 18.85

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 85.3% Irish + 14.7% Swedish @ 2.63
2 89.6% Irish + 10.4% North_Swedish @ 2.67
3 67.9% Irish + 32.1% North_Dutch @ 2.7
4 74.2% West_Scottish + 25.8% North_German @ 2.71
5 97.1% West_Scottish + 2.9% Chechen @ 2.71
6 97.7% Irish + 2.3% Tabassaran @ 2.72
7 97.3% West_Scottish + 2.7% North_Ossetian @ 2.72
8 97.9% Irish + 2.1% Chechen @ 2.73
9 94.6% Irish + 5.4% Finnish @ 2.73
10 86.7% Irish + 13.3% West_Norwegian @ 2.73
11 95.8% Irish + 4.2% Estonian @ 2.73
12 86.4% Irish + 13.6% Norwegian @ 2.74
13 95.5% West_Scottish + 4.5% Lithuanian @ 2.74
14 94.6% West_Scottish + 5.4% Estonian @ 2.75
15 97.9% Irish + 2.1% Lezgin @ 2.75
16 95.3% West_Scottish + 4.7% Ukrainian_Belgorod @ 2.75
17 93.9% Irish + 6.1% Southwest_Finnish @ 2.76
18 97.3% West_Scottish + 2.7% Kabardin @ 2.77
19 98.2% Irish + 1.8% North_Ossetian @ 2.77
20 97.5% West_Scottish + 2.5% Adygei @ 2.77

Sikeliot
09-04-2019, 04:47 AM
Not necessarily.

They might be matching Pre-Anglo-Saxon alleles that are common in England but rare in Scotland (due to different types of Pre-Germanic populations).

Who says that in year, say - 500 BCE - populations living in Ireland were more closely related to those in Scotland than to those in England and Wales?

Elevated Anglo-Saxon ancestry can be confirmed if they also show similar affinity to Denmark, Schleswig, Netherlands, etc. In addition to modern England.

BTW, from this new study:

"We estimate that Norwegian (as well as Danish/Swedish) ancestry is also markedly low in Ireland (average 7%) compared with previous estimates (8, 9) (we explore this further in Discussion)." - this is much lower than some previous estimates.

You could be correct, as the only reason Scots speak Gaelic is because it arrived from Ireland. So we have no way of knowing for sure if the original Gaels of Ireland were closer to Celts in Scotland or to people in England/Wales... but the only way to determine whether that is the reason for elevated English affinity or if it's Anglo-Saxon input is if we took England out of the equation and modeled the Irish as a mixture of Scots, Welsh, and Scandinavians. And they didn't do this.

If we are modeling Leinster/Connacht as 67% "English" then isn't that English cluster masking a lot of Anglo-Saxon, which is about 40% or so in England? Therefore, a calculation would arrive at around 27% "AngloSaxon" for those two regions of Ireland. For comparison, Donegal was modeled with 40% "English", which means their "Anglo-Saxon" is only 16% if we assume 40% Anglo-Saxon for England.

Sikeliot
09-04-2019, 04:47 AM
Interesting as the Baltic is about 4% lower than mine and there Atlantic is about 5% higher. As most people would know my ancestry is North Munster and Sligo/Roscommon.

Tipperary, no?

Jessie
09-04-2019, 04:49 AM
Not necessarily. This can be a signal of Pre-Germanic English-like ancestry (Non-Pictish).

They might be matching Pre-Anglo-Saxon alleles that are common in England but rare in Scotland (due to different types of Pre-Germanic populations).

Who says that in year, say - 500 BCE - populations living in Ireland were more closely related to those in Scotland than to those in England and Wales?

Elevated Anglo-Saxon ancestry can be confirmed if they also show similar affinity to Denmark, Schleswig, Netherlands, etc. In addition to modern England.

BTW, from this new study:

"We estimate that Norwegian (as well as Danish/Swedish) ancestry is also markedly low in Ireland (average 7%) compared with previous estimates (8, 9) (we explore this further in Discussion)." - this is much lower than some previous estimates.

I'd say 7% appears about correct. I think that's what you had estimated previously?

Sikeliot
09-04-2019, 04:52 AM
I'd say 7% appears about correct. I think that's what you had estimated previously?

I still think Scandinavian input appears so low because it is eaten up by what they call "England."

I actually don't think this study was well conducted with regards to trying to measure Scandinavian ancestry.

It'd be like trying to measure Levantine DNA in Malta and but modeling them with "Cyprus" and then saying they have limited Levantine DNA, when it's all sucked into the "Cyprus" cluster.

Jessie
09-04-2019, 04:53 AM
Tipperary, no?

Yes North Tipperary. Ancestry was remarkably correct with their GCs for me and my daughter.

Jessie
09-04-2019, 04:56 AM
I still think Scandinavian input appears so low because it is eaten up by what they call "England."

I actually don't think this study was well conducted with regards to trying to measure Scandinavian ancestry.

It'd be like trying to measure Levantine DNA in Malta and but modeling them with "Cyprus" and then saying they have limited Levantine DNA, when it's all sucked into the "Cyprus" cluster.

I do agree with this. It seems odd as modelling other British is going to mask a lot of other admixture. I think the only way to really solve this is with ancient genomes.

Sikeliot
09-04-2019, 05:26 AM
I do agree with this. It seems odd as modelling other British is going to mask a lot of other admixture. I think the only way to really solve this is with ancient genomes.

All we were able to figure out this time is which Irish regions have more English ancestry, really. Why, historically, Donegal has comparatively less is interesting to hypothesize about though.

sktibo
09-04-2019, 05:39 AM
I'd say 7% appears about correct. I think that's what you had estimated previously?

7% is IIRC the estimate they gave in the early Y DNA analysis of Ireland, I think it mentioned in the paper that their new estimate is in line with this

fridurich
09-04-2019, 05:40 PM
I am very excited that at last we have the results of this test! Seems like the authors of the report said something about the Hebrideans stand out from the rest of Scotland, but, on a PCA chart on one axis the Hebrideans show genetic similarity with North Scotland, and on the other axis the Hebrideans show genetic similarity to the Irish.

So where are these PCA charts? As I scanned the article last night, it sounded like there was at least one PCA chart to be shown. However, I downloaded all that could be downloaded and saw no PCA chart. I did see a T-sne chart though.

Another article I scanned today which was talking about this latest test, said that the Hebrides and Southwest Scotland showed strong links to Ireland. But are they they talking about the Hebrides and Southwest Scotland (Galloway included?) show strong links to the Gaelic Irish (that is what it could easily be taken for), or show strong links to the people of the island of Ireland in general, both the Planter clusters and the native Gaelic Irish ones?

Kind Regards
Fred

sktibo
09-04-2019, 07:09 PM
I am very excited that at last we have the results of this test! Seems like the authors of the report said something about the Hebrideans stand out from the rest of Scotland, but, on a PCA chart on one axis the Hebrideans show genetic similarity with North Scotland, and on the other axis the Hebrideans show genetic similarity to the Irish.

So where are these PCA charts? As I scanned the article last night, it sounded like there was at least one PCA chart to be shown. However, I downloaded all that could be downloaded and saw no PCA chart. I did see a T-sne chart though.

Another article I scanned today which was talking about this latest test, said that the Hebrides and Southwest Scotland showed strong links to Ireland. But are they they talking about the Hebrides and Southwest Scotland (Galloway included?) show strong links to the Gaelic Irish (that is what it could easily be taken for), or show strong links to the people of the island of Ireland in general, both the Planter clusters and the native Gaelic Irish ones?

Kind Regards
Fred

All I found was the t-SNE as well. The only group in Scotland that appears to have exceptionally strong links (because it clusters with Ireland instead of Scotland) is N-Ire-Sco, which is around the west central belt of Scotland. Quite a few markers from the N Ireland cluster in the central lowlands as well. Aside from that, going off the t-SNE chart it looks like the planter cluster (Sco-Ire) which it looks like most of the Galloway samples belong to is the closest Scottish cluster to Ireland with Argyll looking like the next closest one. I'm personally quite surprised that the lowland SW Scottish groups are closer to the Irish than the Hebrideans but I suppose they are geographically closer?

pmokeefe
09-04-2019, 07:16 PM
The PCA for The genetic landscape of Scotland and the Isles (https://www.pnas.org/content/early/2019/08/27/1904761116) was in the appendix (https://www.pnas.org/highwire/filestream/885975/field_highwire_adjunct_files/0/pnas.1904761116.sapp.pdf) to the paper:https://anthrogenica.com/attachment.php?attachmentid=32912&stc=1

Fig. S7. Broad Ancestry Estimates of Ancient Icelanders. (A) First and second principal components of 4,828 Europeans calculated using smartpca11,12 with 209,028 common markers. Twenty-seven ancient Icelanders are projected onto this variation, whose sample IDs are shown. (The D-statistic estimates of each ancient Icelander testing D(Yoruban, Icelander; Gael, Norse) where ‘Gael’ is all modern Scottish and Irish samples and ‘Norse’ is all modern Swedish and Norwegian samples. Points are colour coded according to general group used for subsequent analyses; green is ‘Gael’, orange is ‘Other’, and purple is ‘Norse’. (C) The first and second principal components as in panel A with colour scheme of ancient individuals same as panel B. All plots were created using the statistical computing language R2 and the packages ggplots2.

FionnSneachta
09-04-2019, 09:24 PM
Compare to this person from Galway.. see the affinity to Netherlands, Denmark is much higher as is affinity to England.

Looking at my own family, I get Irish as my top result with Eurogenes K15 (Roscommon, Galway and Mayo) with the mixed mode either being West Scottish or Irish.
My mum (Roscommon and Mayo) gets Danish as her top result with mixed mode giving West Scottish or Danish.
My dad (Roscommon and Galway) gets Orcadian with that being his only suggested mixed mode population.
My dad's aunt (Roscommon) gets Southeast English with mixed mode being Orcadian, Southwest English, Southeast English or West Scottish.

Me
Admix Results (sorted):

# Population Percent
1 North_Sea 36.14
2 Atlantic 34.34
3 Baltic 10.47
4 Eastern_Euro 9.34
5 West_Asian 5.79
6 West_Med 2.89
7 Amerindian 0.8
8 South_Asian 0.24

Single Population Sharing:

# Population (source) Distance
1 Irish 5.57
2 West_Scottish 5.58
3 Danish 7.09
4 Southeast_English 7.31
5 Orcadian 7.42
6 North_Dutch 7.81
7 North_German 7.91
8 Southwest_English 9.31
9 West_Norwegian 11.18
10 Norwegian 11.18
11 Swedish 11.99
12 South_Dutch 12.05
13 North_Swedish 14.16
14 West_German 14.31
15 French 16.64
16 East_German 17.04
17 Southwest_Finnish 17.78
18 Austrian 19.82
19 Finnish 21.4
20 Hungarian 21.65

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 98.1% West_Scottish + 1.9% North_Ossetian @ 5.48
2 98% West_Scottish + 2% Chechen @ 5.48
3 98.3% West_Scottish + 1.7% Adygei @ 5.51
4 98.3% West_Scottish + 1.7% Kabardin @ 5.51
5 98.5% West_Scottish + 1.5% Ossetian @ 5.52
6 98.4% West_Scottish + 1.6% Lezgin @ 5.52
7 98.7% West_Scottish + 1.3% Abhkasian @ 5.52
8 98.4% West_Scottish + 1.6% Tabassaran @ 5.52
9 98.5% West_Scottish + 1.5% Balkar @ 5.52
10 98.7% Irish + 1.3% Chechen @ 5.53
11 53.4% Irish + 46.6% West_Scottish @ 5.53
12 98.8% West_Scottish + 1.2% Georgian @ 5.54
13 98.8% Irish + 1.2% Tabassaran @ 5.54
14 99% Irish + 1% North_Ossetian @ 5.54
15 99% Irish + 1% Lezgin @ 5.54
16 99.2% Irish + 0.8% Adygei @ 5.55
17 99.2% Irish + 0.8% Kabardin @ 5.55
18 99.2% Irish + 0.8% Ossetian @ 5.55
19 98.9% West_Scottish + 1.1% Kumyk @ 5.55
20 99.5% Irish + 0.5% Abhkasian @ 5.56

Mum
Admix Results (sorted):

# Population Percent
1 North_Sea 36.53
2 Atlantic 29.09
3 Eastern_Euro 13.13
4 Baltic 11.23
5 West_Med 5.11
6 West_Asian 3.92
7 Oceanian 0.35
8 South_Asian 0.31
9 Amerindian 0.17
10 East_Med 0.15

Single Population Sharing:

# Population (source) Distance
1 Danish 3.12
2 West_Scottish 4.4
3 North_Dutch 4.49
4 Irish 4.97
5 North_German 5.32
6 Southeast_English 5.85
7 Orcadian 6
8 Norwegian 6.34
9 West_Norwegian 7.16
10 Southwest_English 7.37
11 Swedish 7.55
12 North_Swedish 8.94
13 South_Dutch 9.8
14 West_German 11
15 East_German 13.28
16 Southwest_Finnish 13.66
17 French 14.96
18 Finnish 17.14
19 Austrian 17.34
20 Hungarian 18.12

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 87% West_Scottish + 13% East_Finnish @ 3.01
2 90.6% West_Scottish + 9.4% Erzya @ 3.01
3 89.4% West_Scottish + 10.6% Kargopol_Russian @ 3.03
4 98.4% Danish + 1.6% Chuvash @ 3.05
5 98.7% Danish + 1.3% Mari @ 3.06
6 97.5% Danish + 2.5% East_Finnish @ 3.07
7 96.2% Danish + 3.8% Southwest_Finnish @ 3.07
8 98.2% Danish + 1.8% Kargopol_Russian @ 3.08
9 98.6% Danish + 1.4% Erzya @ 3.08
10 72.9% West_Scottish + 27.1% North_Swedish @ 3.1
11 96.1% Danish + 3.9% North_Swedish @ 3.1
12 98.9% Danish + 1.1% Estonian_Polish @ 3.1
13 98.3% Danish + 1.7% Finnish @ 3.1
14 98.8% Danish + 1.2% Estonian @ 3.1
15 99.1% Danish + 0.9% Southwest_Russian @ 3.11
16 98.9% Danish + 1.1% La_Brana-1 @ 3.11
17 99.2% Danish + 0.8% Belorussian @ 3.11
18 99.3% Danish + 0.7% Ukrainian_Belgorod @ 3.11
19 99.4% Danish + 0.6% Tatar @ 3.11
20 99.2% Danish + 0.8% Polish @ 3.11


Dad
Admix Results (sorted):

# Population Percent
1 North_Sea 39.96
2 Atlantic 32.78
3 Eastern_Euro 6.65
4 Baltic 6.43
5 West_Med 6.02
6 West_Asian 5.61
7 Amerindian 1.89
8 South_Asian 0.53
9 Sub-Saharan 0.13

Single Population Sharing:

# Population (source) Distance
1 Orcadian 5.01
2 West_Scottish 5.85
3 Irish 6.31
4 Southeast_English 7.07
5 North_Dutch 8.28
6 Southwest_English 8.3
7 Danish 8.39
8 West_Norwegian 10.35
9 North_German 10.47
10 Norwegian 11.29
11 South_Dutch 12.64
12 Swedish 12.85
13 West_German 13.49
14 French 15.91
15 North_Swedish 16.03
16 East_German 19.51
17 Southwest_Finnish 21.33
18 Spanish_Cataluna 22.58
19 Austrian 23.01
20 Spanish_Galicia 23.39

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 98.9% Orcadian + 1.1% Karitiana @ 4.82
2 98.9% Orcadian + 1.1% Anzick-1 @ 4.85
3 98.9% Orcadian + 1.1% Pima @ 4.85
4 98.9% Orcadian + 1.1% Mayan @ 4.86
5 99% Orcadian + 1% North_Amerindian @ 4.93
6 96.9% Orcadian + 3.1% French_Basque @ 4.93
7 95.1% Orcadian + 4.9% Irish @ 5.01
8 95.7% Orcadian + 4.3% West_Scottish @ 5.01
9 99.9% Orcadian + 0.1% East_Greenlander @ 5.01
10 99.9% Orcadian + 0.1% Abhkasian @ 5.01
11 100% Orcadian + 0% Ossetian @ 5.01
12 100% Orcadian + 0% Georgian @ 5.01
13 100% Orcadian + 0% Adygei @ 5.01
14 100% Orcadian + 0% Afghan_Hazara @ 5.01
15 100% Orcadian + 0% Afghan_Pashtun @ 5.01
16 100% Orcadian + 0% Afghan_Tadjik @ 5.01
17 100% Orcadian + 0% Afghan_Turkmen @ 5.01
18 100% Orcadian + 0% Afghan_Uzbeki @ 5.01
19 100% Orcadian + 0% Algerian @ 5.01
20 100% Orcadian + 0% Algerian_Jewish @ 5.01


Dad's aunt
Admix Results (sorted):

# Population Percent
1 North_Sea 36.7
2 Atlantic 30.88
3 West_Med 9.66
4 Baltic 8.62
5 Eastern_Euro 7.74
6 West_Asian 4.06
7 South_Asian 1.33
8 Amerindian 0.79
9 Sub-Saharan 0.22

Single Population Sharing:

# Population (source) Distance
1 Southeast_English 3.13
2 Southwest_English 3.37
3 Irish 3.68
4 West_Scottish 3.71
5 Orcadian 3.83
6 North_Dutch 5.78
7 Danish 5.95
8 North_German 7.5
9 South_Dutch 8.28
10 Norwegian 9.27
11 West_Norwegian 9.33
12 West_German 10.02
13 Swedish 10.93
14 French 11.78
15 North_Swedish 13.83
16 East_German 15.31
17 Southwest_Finnish 18.44
18 Spanish_Cataluna 18.66
19 Austrian 19.2
20 Spanish_Galicia 19.3

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 88.1% Orcadian + 11.9% French_Basque @ 1.57
2 86.6% Orcadian + 13.4% Southwest_French @ 2
3 87.8% Orcadian + 12.2% Spanish_Aragon @ 2.21
4 86.8% Orcadian + 13.2% Spanish_Cantabria @ 2.22
5 87.9% Orcadian + 12.1% Spanish_Castilla_La_Mancha @ 2.34
6 55.9% Southwest_English + 44.1% Orcadian @ 2.4
7 88.2% Orcadian + 11.8% Spanish_Valencia @ 2.42
8 89.3% Orcadian + 10.7% Spanish_Andalucia @ 2.49
9 86.8% Orcadian + 13.2% Spanish_Cataluna @ 2.61
10 55.3% Southwest_English + 44.7% West_Scottish @ 2.61
11 87.6% Orcadian + 12.4% Spanish_Castilla_Y_Leon @ 2.62
12 63% Southeast_English + 37% Orcadian @ 2.69
13 55.5% Southwest_English + 44.5% Irish @ 2.73
14 88.6% Orcadian + 11.4% Spanish_Murcia @ 2.74
15 88.9% Orcadian + 11.1% Spanish_Extremadura @ 2.77
16 94.8% Orcadian + 5.2% Sardinian @ 2.82
17 91.6% West_Scottish + 8.4% French_Basque @ 2.84
18 89.8% West_Scottish + 10.2% Spanish_Cantabria @ 2.88
19 88.8% Orcadian + 11.2% Portuguese @ 2.88
20 95.5% West_Scottish + 4.5% Sardinian @ 2.95


In terms of the K36 similarity map, I get Irish as my top similarity (94) but other places in Britain as well as Brittany, Denmark and the Netherlands get over 85.
My mum gets Ireland (89) as her top similarity with other places in Britain and Brittany getting over 85.
My dad's highest result (92) is in Wales and England with over 85 being in Ireland, the rest of Britain, Brittany, Denmark and the Netherlands.
My great aunt's only result over 85 is Brittany (87).

32917

I don't find it unusual that Donegal is the most isolated. It's probably still the most isolated county.

sktibo
09-05-2019, 02:02 AM
I am very excited that at last we have the results of this test! Seems like the authors of the report said something about the Hebrideans stand out from the rest of Scotland, but, on a PCA chart on one axis the Hebrideans show genetic similarity with North Scotland, and on the other axis the Hebrideans show genetic similarity to the Irish.

So where are these PCA charts? As I scanned the article last night, it sounded like there was at least one PCA chart to be shown. However, I downloaded all that could be downloaded and saw no PCA chart. I did see a T-sne chart though.


Think I may have found what you are talking about:

32921329223292332924

fridurich
09-05-2019, 02:57 AM
The PCA for The genetic landscape of Scotland and the Isles (https://www.pnas.org/content/early/2019/08/27/1904761116) was in the appendix (https://www.pnas.org/highwire/filestream/885975/field_highwire_adjunct_files/0/pnas.1904761116.sapp.pdf) to the paper:https://anthrogenica.com/attachment.php?attachmentid=32912&stc=1

Fig. S7. Broad Ancestry Estimates of Ancient Icelanders. (A) First and second principal components of 4,828 Europeans calculated using smartpca11,12 with 209,028 common markers. Twenty-seven ancient Icelanders are projected onto this variation, whose sample IDs are shown. (The D-statistic estimates of each ancient Icelander testing D(Yoruban, Icelander; Gael, Norse) where ‘Gael’ is all modern Scottish and Irish samples and ‘Norse’ is all modern Swedish and Norwegian samples. Points are colour coded according to general group used for subsequent analyses; green is ‘Gael’, orange is ‘Other’, and purple is ‘Norse’. (C) The first and second principal components as in panel A with colour scheme of ancient individuals same as panel B. All plots were created using the statistical computing language R2 and the packages ggplots2.

Hi and thanks so much for your reply. But I think the PCA charts they are referring to are on page 8 of the appendix. I just found them. On one PCA it looks like the Hebrideans are close to the Irish, and on another one, close to other N. Scotland clusters.

Kind Regards
Fred

fridurich
09-05-2019, 03:12 AM
Think I may have found what you are talking about:

32921329223292332924

Yes, those are them! I just found them myself a little while ago. So is Principal Component 2 the Y axis and Principal Component 1 the horizontal axis? If I remember right, it seems like on one of these charts you can show relationship to other clusters on both axis. So is a rough way of putting it is that everything below or above a cluster is also to some degree genetically close to another cluster, but also, everything to either side of them is somewhat close to them also? I guess with the clusters that are physically closest on the same axis being more closely related? I may have that a little off.

So, how does a T-sne chart show who is the most closely related to a cluster? Sorry for so many questions.

King Regards
Fred

sktibo
09-05-2019, 03:37 AM
Yes, those are them! I just found them myself a little while ago. So is Principal Component 2 the Y axis and Principal Component 1 the horizontal axis? If I remember right, it seems like on one of these charts you can show relationship to other clusters on both axis. So is a rough way of putting it is that everything below or above a cluster is also to some degree genetically close to another cluster, but also, everything to either side of them is somewhat close to them also? I guess with the clusters that are physically closest on the same axis being more closely related? I may have that a little off.

So, how does a T-sne chart show who is the most closely related to a cluster? Sorry for so many questions.

King Regards
Fred

I don't understand the technicalities of it either, my understanding is that the PCA graphs demonstrate similarity by distance but are limited to two components. The relationships between the clusters can often change depending on which components are selected but I think I read that 1 and 2 are more indicative of the nature of the components than the later numbers. I'm not exactly sure what t-SNE stands for but it looks to me like they selected it for the main paper because it shows distinctions between the clusters much more clearly than the PCAs. I think you just look at the distance between one cluster to the next on the t-SNE graph like the PCA, but there could be more to it than that. I believe the spanning trees are the best method to examine the relationship between these clusters but the graphs must also be examined to get a better look.. Don't want to pass anything up with this stuff...

fridurich
09-08-2019, 05:22 AM
In the Scottish/Manx autosomal DNA article it says that the Hebrides show a genetic affinity for Ireland on Principal Component 2, which is what it looks like to me, with the Hebrides showing a particularly strong affinity to the N. Ireland Cluster. Also, it looks like most of Argyll (blue boxes with brown edges) shows genetic affinity to Ireland, as well as the Hebrides, and most of Argyll appears to be positioned between the Hebrides and Ireland on PC1.

It also looks like on Principal Component 1, that the clusters of Ireland stand like a tower over, or close to directly over the Sco-Ire cluster (which I believe represents the 17th and 18th Century Planters in Ulster), an Isle of Man cluster (with its blue boxes with black edges), other clusters, and the Eng-Sco-Ire cluster!! To me this appears to show on Principal Component 1, that there is some genetic affinity between Ireland, and the Sco-Ire, Isle of Man, Eng-Sco-Ire cluster, and others that I haven't taken the time to identify yet. Surprisingly, on PC1, it appears that most of the Eng-Sco-Ire cluster is directly under much of the Irish clusters.

Has anyone else come up with my observations here? I saw a video on youtube where it said the axes on PCA charts are ranked according to importance. So, Principal Component 1 would be most important. I also read somewhere that the first two or three PCA's are the most important (it might have been the first two).

Kind Regards
Fred

sktibo
09-08-2019, 07:03 AM
In the Scottish/Manx autosomal DNA article it says that the Hebrides show a genetic affinity for Ireland on Principal Component 2, which is what it looks like to me, with the Hebrides showing a particularly strong affinity to the N. Ireland Cluster. Also, it looks like most of Argyll (blue boxes with brown edges) shows genetic affinity to Ireland, as well as the Hebrides, and most of Argyll appears to be positioned between the Hebrides and Ireland on PC1.

It also looks like on Principal Component 1, that the clusters of Ireland stand like a tower over, or close to directly over the Sco-Ire cluster (which I believe represents the 17th and 18th Century Planters in Ulster), an Isle of Man cluster (with its blue boxes with black edges), other clusters, and the Eng-Sco-Ire cluster!! To me this appears to show on Principal Component 1, that there is some genetic affinity between Ireland, and the Sco-Ire, Isle of Man, Eng-Sco-Ire cluster, and others that I haven't taken the time to identify yet. Surprisingly, on PC1, it appears that most of the Eng-Sco-Ire cluster is directly under much of the Irish clusters.

Has anyone else come up with my observations here? I saw a video on youtube where it said the axes on PCA charts are ranked according to importance. So, Principal Component 1 would be most important. I also read somewhere that the first two or three PCA's are the most important (it might have been the first two).

Kind Regards
Fred

I think I heard the first two for most important as well. Even if the others are significant I find them difficult to analyze but thankfully 1 & 2 plot in a way where you can see the relationship between the clusters despite how crowded the plot is. On PCA 1/2 the Hebridean clusters have definitely migrated out of Scotland and up with the Irish clusters, with a bit of a pull towards the northern Isles. That's a bit of a different relationship than the one we see in the spanning tree but Scotland and Ireland are quite closely related. The spanning tree has them grouped together, separated from England and Wales, but on that the Hebrides still cluster with Scotland despite forming their own outer branch. It's incredibly surprising that the N-Sco-Ire cluster found primarily in Scotland actually clusters with Ireland and is a lowland Scottish cluster! As for the Eng-Sco-Ire cluster I think the paper speculated that many of these were individuals of mixed Irish and English descent, the result of people who moved from Ireland for work and married locals. What looks really interesting is that the markers from around Caithness (either called Highland or N Scotland) have pulled away from the main clusters towards the northern Isles - even further than the Hebrides or Aberdeen clusters.

fridurich
09-10-2019, 05:29 AM
I think I heard the first two for most important as well. Even if the others are significant I find them difficult to analyze but thankfully 1 & 2 plot in a way where you can see the relationship between the clusters despite how crowded the plot is. On PCA 1/2 the Hebridean clusters have definitely migrated out of Scotland and up with the Irish clusters, with a bit of a pull towards the northern Isles. That's a bit of a different relationship than the one we see in the spanning tree but Scotland and Ireland are quite closely related. The spanning tree has them grouped together, separated from England and Wales, but on that the Hebrides still cluster with Scotland despite forming their own outer branch. It's incredibly surprising that the N-Sco-Ire cluster found primarily in Scotland actually clusters with Ireland and is a lowland Scottish cluster! As for the Eng-Sco-Ire cluster I think the paper speculated that many of these were individuals of mixed Irish and English descent, the result of people who moved from Ireland for work and married locals. What looks really interesting is that the markers from around Caithness (either called Highland or N Scotland) have pulled away from the main clusters towards the northern Isles - even further than the Hebrides or Aberdeen clusters.

I agree, it is really surprising that the N-Sco-Ire cluster found primarily in Scotland clusters with the Irish! I think I have seen only 3 of it's circles in Ireland, 2 in Ulster and 1 in Dublin? A large portion of N-Sco-Ire is in the Lowlands although to me some of them appear to be close to the Highland line, or even a bit into the Highlands. On PC 1, the N-Sco-Ire cluster is sandwiched right between the N. Ireland cluster and the Munster cluster. You are right the N Scotland cluster does appear to be pulling away from mainland Scotland and kind of towards the Shetlands and Orkneys, interesting.

Also, intriguing is the one lonely N. Ireland circle in Perthshire. This kind of reminds me that YDNA wise, my O'Hair/O'Hare paternal line from County Down, in what is now Northern Ireland, is M222>…S588>S603. I think time to most recent common ancestor for all S588 men is about 1500 years ago. S588 has a lot of Irish families in it and also has a lot of Scottish families in it. That one lone N. Ireland circle doesn't appear real far away from Struan in Perthshire, where the Struan Robertsons were from (Clann Donnachaidh).

A lady that has a lot to do with the Robertson or Clann Donnachaidh YDNA Project said they have identified the chiefly Robertson line of Struan and the cadet Robertson line of Lude (also in Perthshire). These can be seen on the Big Tree for S603 which has 15 Robertsons who have taken the Big Y or other NGS YDNA test! At least a few of these Robertsons have tested for S603. I think all of those have been finalized positive for S603. There are also quite a number of both Irish and other Scottish families who are S603.

https://www.ytree.net/DisplayTree.php?blockID=2344

I know that that one lone N. Ireland circle in Perthshire only represents autosomal DNA (would be nice to know what their surname and YDNa is), but interesting that they are so close to where the Struan Robertson M222>...S588>S603 line was from and in Ulster, where there are so many of the N. Ireland circles, there are so many M222>...S588 men.

Would be interesting to know where the migration route originated from and went to for these S588 guys in Ulster and Perthshire.

Kind Regards
Fred

sktibo
09-10-2019, 02:46 PM
I agree, it is really surprising that the N-Sco-Ire cluster found primarily in Scotland clusters with the Irish! I think I have seen only 3 of it's circles in Ireland, 2 in Ulster and 1 in Dublin? A large portion of N-Sco-Ire is in the Lowlands although to me some of them appear to be close to the Highland line, or even a bit into the Highlands. On PC 1, the N-Sco-Ire cluster is sandwiched right between the N. Ireland cluster and the Munster cluster. You are right the N Scotland cluster does appear to be pulling away from mainland Scotland and kind of towards the Shetlands and Orkneys, interesting.

Also, intriguing is the one lonely N. Ireland circle in Perthshire. This kind of reminds me that YDNA wise, my O'Hair/O'Hare paternal line from County Down, in what is now Northern Ireland, is M222>…S588>S603. I think time to most recent common ancestor for all S588 men is about 1500 years ago. S588 has a lot of Irish families in it and also has a lot of Scottish families in it. That one lone N. Ireland circle doesn't appear real far away from Struan in Perthshire, where the Struan Robertsons were from (Clann Donnachaidh).

A lady that has a lot to do with the Robertson or Clann Donnachaidh YDNA Project said they have identified the chiefly Robertson line of Struan and the cadet Robertson line of Lude (also in Perthshire). These can be seen on the Big Tree for S603 which has 15 Robertsons who have taken the Big Y or other NGS YDNA test! At least a few of these Robertsons have tested for S603. I think all of those have been finalized positive for S603. There are also quite a number of both Irish and other Scottish families who are S603.

https://www.ytree.net/DisplayTree.php?blockID=2344

I know that that one lone N. Ireland circle in Perthshire only represents autosomal DNA (would be nice to know what their surname and YDNa is), but interesting that they are so close to where the Struan Robertson M222>...S588>S603 line was from and in Ulster, where there are so many of the N. Ireland circles, there are so many M222>...S588 men.

Would be interesting to know where the migration route originated from and went to for these S588 guys in Ulster and Perthshire.

Kind Regards
Fred

The yellow N Scotland marker is not near Struan nor is it in the Perthshire group, it's on top of one of the Tayside-Fife clusters but actually it appears to be near Aberlemno. It's in line with the Aberdeenshire markers, well out of Perthshire.
The colours can be hard to see with the N Scotland markers so it's good to get verification that these look like the N Scotland (Caithness) markers to you as well!

fridurich
09-11-2019, 03:45 AM
The yellow N Scotland marker is not near Struan nor is it in the Perthshire group, it's on top of one of the Tayside-Fife clusters but actually it appears to be near Aberlemno. It's in line with the Aberdeenshire markers, well out of Perthshire.
The colours can be hard to see with the N Scotland markers so it's good to get verification that these look like the N Scotland (Caithness) markers to you as well!

Well, I really got the location of the lone solid yellow circle (which I thought was a N Ireland circle) on top of the Tayside-Fife markers wrong!! lol So, that lone solid yellow circle on top of the cluster of Tayside-Fife markers is a symbol for the N. Scotland cluster? To me the symbol for the N Scotland marker looks like either a brown, or purple diamond with a brown or yellow border (kind of hard to tell), but not solid yellow like the lone circle I'm talking about.

Maybe I need to get my eyeglass prescription changed! lol

Kind Regards
Fred

sktibo
09-11-2019, 04:15 AM
Well, I really got the location of the lone solid yellow circle (which I thought was a N Ireland circle) on top of the Tayside-Fife markers wrong!! lol So, that lone solid yellow circle on top of the cluster of Tayside-Fife markers is a symbol for the N. Scotland cluster? To me the symbol for the N Scotland marker looks like either a brown, or purple diamond with a brown or yellow border (kind of hard to tell), but not solid yellow like the lone circle I'm talking about.

Maybe I need to get my eyeglass prescription changed! lol

Kind Regards
Fred

My bad Fred! I meant to say: "The yellow N Ireland marker is not near Struan nor is it in the Perthshire group, it's on top of one of the Tayside-Fife clusters but actually it appears to be near Aberlemno. It's in line with the Aberdeenshire markers, well out of Perthshire."
And in the same post, I went on to talk about PCA 1 and 2 in regards to the N Scotland markers without specifying a change of topic (yes, the blue/purple diamond with the brownish border.) I can see how that would be rather confusing.. I'm even confused re-reading my previous post!

Sikeliot
09-11-2019, 12:39 PM
Something I am wondering is why people find the new study's estimate of only 7% Scandinavian in Ireland, more convincing than past studies which estimated it in the 15-20% range, with an upper limit of 20%.

What I found most convincing in the new study is they had Munster and then southern Connacht/Leinster (Central Ireland) as the areas with the most of this Viking input, since we know the majority of Viking settlements were in this area. But is 7% really the right estimate? To me that seems quite low.

sktibo
09-11-2019, 04:16 PM
Something I am wondering is why people find the new study's estimate of only 7% Scandinavian in Ireland, more convincing than past studies which estimated it in the 15-20% range, with an upper limit of 20%.

What I found most convincing in the new study is they had Munster and then southern Connacht/Leinster (Central Ireland) as the areas with the most of this Viking input, since we know the majority of Viking settlements were in this area. But is 7% really the right estimate? To me that seems quite low.

It's convincing because its in line with the estimate done using Y DNA. 20% is also way too high because it would mean that Ireland has around about as much Scandinavian admixture as the Orkneys.
IMO, the 15-20% Scandinavian estimate is great for article headlines and press but doesn't seem realistic.

fridurich
09-12-2019, 04:02 AM
My bad Fred! I meant to say: "The yellow N Ireland marker is not near Struan nor is it in the Perthshire group, it's on top of one of the Tayside-Fife clusters but actually it appears to be near Aberlemno. It's in line with the Aberdeenshire markers, well out of Perthshire."
And in the same post, I went on to talk about PCA 1 and 2 in regards to the N Scotland markers without specifying a change of topic (yes, the blue/purple diamond with the brownish border.) I can see how that would be rather confusing.. I'm even confused re-reading my previous post!

That's o.k., no worries! There is no telling how many times I make mistakes, that could be avoided, everyday!

Kind Regards
Fred

Nqp15hhu
09-14-2019, 11:27 AM
All I found was the t-SNE as well. The only group in Scotland that appears to have exceptionally strong links (because it clusters with Ireland instead of Scotland) is N-Ire-Sco, which is around the west central belt of Scotland. Quite a few markers from the N Ireland cluster in the central lowlands as well. Aside from that, going off the t-SNE chart it looks like the planter cluster (Sco-Ire) which it looks like most of the Galloway samples belong to is the closest Scottish cluster to Ireland with Argyll looking like the next closest one. I'm personally quite surprised that the lowland SW Scottish groups are closer to the Irish than the Hebrideans but I suppose they are geographically closer?

Isn’t that a region with a lot of Irish migration? So surely it wouldn’t be representative of Scottish DNA?

Nqp15hhu
09-14-2019, 11:30 AM
I am still trying to get my head around the Northern Irish clusters.

Is the Northern Ireland cluster for “natives” with maybe a scattering or Ulster Scots ancestors or is it for someone with majority Irish but still some Ulster Scots? What about the NI-Scotland Cluster?

They need to explain this.

I don’t know what cluster I fit into. The majority of my ancestry is Ulster Irish but I still have a decent chunk of Ulster Scots, as in 30%.

sktibo
09-14-2019, 03:38 PM
Isn’t that a region with a lot of Irish migration? So surely it wouldn’t be representative of Scottish DNA?

While they would have obviously tried their best to choose individuals without backgrounds from multiple nations or countries, I think it must be a possibility that some of these individuals may have Irish ancestry. The cluster called "Eng-Sco-Ire" was noted as probably being the result of this if I remember correctly. I don't recall reading anything that discussed this possibility for the N-Sco-Ire cluster. The areas this cluster includes do seem to be representative of many areas with 19th Century Irish immigration - except there are none near Dundee which apparently was a major location that people from Ireland would immigrate to.
I think it is possible looking at this cluster that it could have been affected by 19th C immigration but it does not appear to be completely certain.


I am still trying to get my head around the Northern Irish clusters.

Is the Northern Ireland cluster for “natives” with maybe a scattering or Ulster Scots ancestors or is it for someone with majority Irish but still some Ulster Scots? What about the NI-Scotland Cluster?


Looking at it from a personal perspective is going to confuse you. They took the samples from all over Ireland and grouped them by genetic similarity. The names to the clusters are given after they are grouped by genetic similarity. "N Ireland" is a really broad cluster and it spans North Connacht to Leinster to Ulster. The admixture charts included in this study indicate that Donegal is the group which is least mixed, then N Ireland, then Munster, then Central Ireland. The cluster which corresponds to the Ulster Scots is called "Sco-Ire" and is a cluster which groups with Scotland, not Ireland. While "N-Sco-Ire" clusters with Ireland, it is found only in Scotland with the exception of two markers found in Ireland.
Donegal, N Ireland, C Ireland, and Munster are all "Native Irish" clusters - I'm including the old English and Scandinavian who have lived there for hundreds of years as part of the native mixture in this specific instance. "Sco-Ire" is not a native Irish cluster as it is pretty clearly sourced from Scotland.

fridurich
09-14-2019, 07:24 PM
I am still trying to get my head around the Northern Irish clusters.

Is the Northern Ireland cluster for “natives” with maybe a scattering or Ulster Scots ancestors or is it for someone with majority Irish but still some Ulster Scots? What about the NI-Scotland Cluster?

They need to explain this.

I don’t know what cluster I fit into. The majority of my ancestry is Ulster Irish but I still have a decent chunk of Ulster Scots, as in 30%.

I wish that the authors of the Scotland /Isle of Man dna study had explained more which Irish clusters they thought were native Irish, etc. Although I thought it was a very interesting study, I thought it was way too brief, especially considering the length of the Irish DNA Project results paper.

Also, basically I agree with Sktibo on which clusters are native Irish. I think they are the Donegal clusters, N Ireland, C Ireland, and Munster.

I also include the Vikings and Old English/Normans as part of the native Irish mix. After all they intermarried extensively with the Gaelic Irish, and had been there for a very, very long time. I think certainly someone whose ancestors who arrived in Ireland about 900 A. D. (such as some of the Vikings) or any time between about 1169 through up to about 1400 A. D. (such as the Normans and other people such as English, Scot and Welsh) can be regarded as native Irish since they were in Ireland so long and almost all had intermarried with the Gaelic Irish anyway.

In the Scot and Isle of Man DNA paper the authors did say that the Hebrides showed an affinity to Ireland along Principal Component 2, which is the y axis. I agree with that, but also much of the Argyll cluster is between the Hebrides and the N Ireland and Munster clusters on Principal Component 1 and also Argyll is in about the same place as those Irish clusters on PC2. So, I think that shows that Argyll also shows genetic similarity to those Irish clusters, yet this isn't mentioned in the paper.

The N Ire Sco cluster really clusters close to the native Irish being sandwiched between the Munster and N Ireland clusters, yet there are only 3 of their green circular symbols in Ireland/Northern Ireland. Since they are mainly in Scotland, many of them in the Lowlands, I would have to call it a Scot cluster. Perhaps many of the N Ire Sco cluster are descended from 19th Century Irish immigrants to Scotland, but if so, then would almost the whole group have left Ireland? As there is only 3 of their symbols shown in Ireland. Maybe it is a mix of some who came to Scotland anciently and others in the 19th Century.

There are quite a number of N. Ireland clusters also in about mid to Southwest Scotland. I don't remember that being mentioned in the paper either.

Kind Regards
Fred

sktibo
09-14-2019, 07:30 PM
Perhaps many of the N Ire Sco cluster are descended from 19th Century Irish immigrants to Scotland, but if so, then would almost the whole group have left Ireland? As there is only 3 of their symbols shown in Ireland. Maybe it is a mix of some who came to Scotland anciently and others in the 19th Century.

There are quite a number of N. Ireland clusters also in about mid to Southwest Scotland. I don't remember that being mentioned in the paper either.

Kind Regards
Fred

Good point Fred, perhaps the N Ireland circles found in Scotland are more indicative of Irish immigration than N-Sco-Ire, and that one yellow circle on top of the eastern Tayside-Fife cluster isn't that far away from Dundee... mind you, if the samples did make it into the study they would surely be well mixed with the locals.

fridurich
09-14-2019, 08:49 PM
Good point Fred, perhaps the N Ireland circles found in Scotland are more indicative of Irish immigration than N-Sco-Ire, and that one yellow circle on top of the eastern Tayside-Fife cluster isn't that far away from Dundee... mind you, if the samples did make it into the study they would surely be well mixed with the locals.

Thanks Sktibo. I also agree with you that the Sco Ire cluster represents the Planters in Northern Ireland. Where they are shown to in Scotland corresponds to where most or a big percentage of the Planters came from and where they are shown in Ireland is in about the same place as where the Planters settled.

There was a couple of other things I would like to get your ideas on. Looking at the first PCA chart, could we say that C Ireland shows genetic affinity with the Sco Ire cluster? I see where both groups are in about the same place on Principal Component One, and are close on Principal Component 2. Could we say that the Munster sample shows genetic affinity with the Sco Ire cluster? Munster is also in about the same position on Principal Component 1 as the Sco Ire cluster is, and on PC2, it looks to me that Munster is in about the same position, or not far from, the Sco Ire cluster, although I don't know if some of the Sco Ire cluster is hidden behind some of the other clusters.

I see where the N. Ireland cluster is in about the same position as the Sco Ire cluster on Principal Component 1, but it is higher above the Sco Ire cluster on PC2 than the C Ireland and Munster cluster are. So, is the N. Ireland cluster too high to say it shows genetic affinity to the Sco Ire cluster? If we can't say the N Ireland cluster has genetic affinity with the Sco Ire cluster, maybe we can say that the N Ireland cluster is closer genetically to the Sco Ire cluster than the N Ireland cluster is to the Borders (Scotland England border region) cluster and some other clusters.

I watched some more YouTube video on PCA charts. For those of you that aren't familiar with how Principal Components are ranked, the PCAs are ranked in order of importance, that is Principal Component 1 is most important and shows more variation that any of the other PCs, PC2 shows the second most variation, PC3 shows the next most variation, etc. And if two objects or symbols on PC1 show the same distance from each other as another pair of objects or symbols show distance from each other on PC2, that even though the distance is the same on the chart, the two objects on PC1 are more different from each other than the two objects on PC2.

However, I'm not quite sure how the height of the N. Ireland cluster over the Sco Ire cluster affects it's relationship with the Sco Ire cluster on PC2.

I was disappointed that there weren't many samples from Galloway shown on the Scotland map. I don't know how many people each symbol represents. Also, I know that Galloway isn't heavily populated and most of the population is on the coast, but, a lot of the Planters came from there and it would be nice if they had gotten a lot more samples. Maybe a large scale Galloway autosomal DNA project should be done, and then compared to Ireland, Scotland, England, Wales and the Isle of Man.

Kind Regards
Fred

sktibo
09-14-2019, 11:08 PM
Thanks Sktibo. I also agree with you that the Sco Ire cluster represents the Planters in Northern Ireland. Where they are shown to in Scotland corresponds to where most or a big percentage of the Planters came from and where they are shown in Ireland is in about the same place as where the Planters settled.

There was a couple of other things I would like to get your ideas on. Looking at the first PCA chart, could we say that C Ireland shows genetic affinity with the Sco Ire cluster? I see where both groups are in about the same place on Principal Component One, and are close on Principal Component 2. Could we say that the Munster sample shows genetic affinity with the Sco Ire cluster? Munster is also in about the same position on Principal Component 1 as the Sco Ire cluster is, and on PC2, it looks to me that Munster is in about the same position, or not far from, the Sco Ire cluster, although I don't know if some of the Sco Ire cluster is hidden behind some of the other clusters.

I see where the N. Ireland cluster is in about the same position as the Sco Ire cluster on Principal Component 1, but it is higher above the Sco Ire cluster on PC2 than the C Ireland and Munster cluster are. So, is the N. Ireland cluster too high to say it shows genetic affinity to the Sco Ire cluster? If we can't say the N Ireland cluster has genetic affinity with the Sco Ire cluster, maybe we can say that the N Ireland cluster is closer genetically to the Sco Ire cluster than the N Ireland cluster is to the Borders (Scotland England border region) cluster and some other clusters.

I watched some more YouTube video on PCA charts. For those of you that aren't familiar with how Principal Components are ranked, the PCAs are ranked in order of importance, that is Principal Component 1 is most important and shows more variation that any of the other PCs, PC2 shows the second most variation, PC3 shows the next most variation, etc. And if two objects or symbols on PC1 show the same distance from each other as another pair of objects or symbols show distance from each other on PC2, that even though the distance is the same on the chart, the two objects on PC1 are more different from each other than the two objects on PC2.

However, I'm not quite sure how the height of the N. Ireland cluster over the Sco Ire cluster affects it's relationship with the Sco Ire cluster on PC2.

I was disappointed that there weren't many samples from Galloway shown on the Scotland map. I don't know how many people each symbol represents. Also, I know that Galloway isn't heavily populated and most of the population is on the coast, but, a lot of the Planters came from there and it would be nice if they had gotten a lot more samples. Maybe a large scale Galloway autosomal DNA project should be done, and then compared to Ireland, Scotland, England, Wales and the Isle of Man.

Kind Regards
Fred

I don't think you can judge importance based on Principal Component One alone because if you are going by PC 1, then Donegal has a relationship with England as they're on the same line for PC 1. It's showing distance from one cluster to the next - England to Donegal are all basically in the same line, with the exception of some of the Scottish clusters leaning a bit towards what I assume would be a Scandinavian direction. So if we are comparing everything except for the Welsh, Orcadian, and Shetlandic clusters, then PC 1 is not significant in the comparison. I don't see this as C Ireland having an affinity or anything with Sco-Ire, more than say, the Isle of Man cluster. It shows that the furthest, most separated cluster from England (if you choose to look at England as the opposing end) is Donegal, and from closest to furthest from Donegal we have: N Ireland, then N-Ire-Sco and it looks to me like the Hebrides, then Munster, then Central Ireland, then Sco-Ire, Man, and it gets really hard to tell which is the next in line at this point. I don't know exactly how to gauge the differences in PC 1 which appear slight to me, like with Tayside-Fife or the Hebrides.

The clusters that have an affinity for one another can be seen on the dendrogram, I believe that's what that indicates. This PCA looks to me like it reflects the results we see in the admixture charts, notably, the Wales-Norway-England one, where if we look at the clusters which more or less align with PC 1 as they move away from Donegal so does their "England" admixture component increase.
The clusters that have an affinity to Sco-Ire appear to be Argyll and Man.

fridurich
09-16-2019, 02:35 AM
I don't think you can judge importance based on Principal Component One alone because if you are going by PC 1, then Donegal has a relationship with England as they're on the same line for PC 1. It's showing distance from one cluster to the next - England to Donegal are all basically in the same line, with the exception of some of the Scottish clusters leaning a bit towards what I assume would be a Scandinavian direction. So if we are comparing everything except for the Welsh, Orcadian, and Shetlandic clusters, then PC 1 is not significant in the comparison. I don't see this as C Ireland having an affinity or anything with Sco-Ire, more than say, the Isle of Man cluster. It shows that the furthest, most separated cluster from England (if you choose to look at England as the opposing end) is Donegal, and from closest to furthest from Donegal we have: N Ireland, then N-Ire-Sco and it looks to me like the Hebrides, then Munster, then Central Ireland, then Sco-Ire, Man, and it gets really hard to tell which is the next in line at this point. I don't know exactly how to gauge the differences in PC 1 which appear slight to me, like with Tayside-Fife or the Hebrides.

The clusters that have an affinity for one another can be seen on the dendrogram, I believe that's what that indicates. This PCA looks to me like it reflects the results we see in the admixture charts, notably, the Wales-Norway-England one, where if we look at the clusters which more or less align with PC 1 as they move away from Donegal so does their "England" admixture component increase.
The clusters that have an affinity to Sco-Ire appear to be Argyll and Man.

Thanks Sktibo for your reply. I agree that you can't just judge importance on Principal Component 1 alone. I hope I didn't come across as thinking that was the only PC that mattered, because I believe that both do, just PC1 is considered more important. I agree that the dendogram is important, but somewhere back in the many pages of this IDA thread I believe someone said that the dendogram isn't as important as the PCA charts in showing relationship. I don't know how much expertise they had on reading different charts.

The authors of the paper, in paragraph 7 of Genetic Landscape of Ireland, say the Hebridean people show genetic affinities to the North of Scotland and Ireland. I can see where along PC2 the Hebrides are right across from the N Ireland cluster. You can see the distance between the two horizontally on PC1, but apparently it isn't enough to nullify that the Hebrides and N. Ireland show genetic affinity on PC2. Additionally, the authors say that the samples from the Hebrides have a higher proportion of Admixture ancestral components that are the most frequent in Ireland.

The majority of the blue with brown border Argyll squares appear between the Hebrides and the N Ireland cluster, the majority are about the same height as N Ireland and the Hebrides on PC2, and I believe this shows that all 3 have genetic affinity on PC2. Since the horizontal distance on PC1 between most of Argyll and N Ireland is about half the horizontal PC1 distance between the Hebrides and N Ireland, to me it seems inescapable that Argyll shows genetic affinity with the N Ireland cluster. Also, I remember where the authors say that Argyll is intermediate between the Hebrides and Sco Ire.

I looked at it hard again, and it looks like part of the C Ireland cluster is in about the same position on both PC1 and PC2 as some of the Sco Ire cluster. Also, it looks like about half of the Isle of Man symbols have the same PC1 and PC2 positions as the Sco Ire cluster. And it looks like a small number of the Munster cluster circles are close to the PC1 and PC2 positions of some of the Sco Ire circles.

It's kind of like almost each cluster has a continuum. All of the symbols in each cluster are similar to each other. However some of the symbols in Cluster A can be quite close to, or among some of the symbols of Cluster B, while other Cluster A symbols can be quite a distance from even the closest symbols of Cluster B.

Thanks again Sktibo. Hey, I have been wrong on interpreting charts before, so if anyone sees something I missed, or want to add any other constructive comments, your input is welcome.

Kind Regards
Fred

sktibo
09-16-2019, 03:56 AM
Thanks Sktibo for your reply. I agree that you can't just judge importance on Principal Component 1 alone. I hope I didn't come across as thinking that was the only PC that mattered, because I believe that both do, just PC1 is considered more important. I agree that the dendogram is important, but somewhere back in the many pages of this IDA thread I believe someone said that the dendogram isn't as important as the PCA charts in showing relationship. I don't know how much expertise they had on reading different charts.

The authors of the paper, in paragraph 7 of Genetic Landscape of Ireland, say the Hebridean people show genetic affinities to the North of Scotland and Ireland. I can see where along PC2 the Hebrides are right across from the N Ireland cluster. You can see the distance between the two horizontally on PC1, but apparently it isn't enough to nullify that the Hebrides and N. Ireland show genetic affinity on PC2. Additionally, the authors say that the samples from the Hebrides have a higher proportion of Admixture ancestral components that are the most frequent in Ireland.

The majority of the blue with brown border Argyll squares appear between the Hebrides and the N Ireland cluster, the majority are about the same height as N Ireland and the Hebrides on PC2, and I believe this shows that all 3 have genetic affinity on PC2. Since the horizontal distance on PC1 between most of Argyll and N Ireland is about half the horizontal PC1 distance between the Hebrides and N Ireland, to me it seems inescapable that Argyll shows genetic affinity with the N Ireland cluster. Also, I remember where the authors say that Argyll is intermediate between the Hebrides and Sco Ire.

I looked at it hard again, and it looks like part of the C Ireland cluster is in about the same position on both PC1 and PC2 as some of the Sco Ire cluster. Also, it looks like about half of the Isle of Man symbols have the same PC1 and PC2 positions as the Sco Ire cluster. And it looks like a small number of the Munster cluster circles are close to the PC1 and PC2 positions of some of the Sco Ire circles.

It's kind of like almost each cluster has a continuum. All of the symbols in each cluster are similar to each other. However some of the symbols in Cluster A can be quite close to, or among some of the symbols of Cluster B, while other Cluster A symbols can be quite a distance from even the closest symbols of Cluster B.

Thanks again Sktibo. Hey, I have been wrong on interpreting charts before, so if anyone sees something I missed, or want to add any other constructive comments, your input is welcome.

Kind Regards
Fred

Hey Fred, another long-winded response for you.. (perhaps you can tell I also enjoy staring at these little circles a bit too much?)

Just to reiterate in this case because we are not including the clusters that branch off significantly along PC1 I was saying that as all the clusters we are examining fall more or less along the same line PC 1 is not going to inform us of much for the purposes of what we are looking at here, so we can get that out of the way and start looking at the differences along PC 2. Surely the PCA is reflecting some kind of relationship between the Hebrides, and as I believe you have correctly pointed out, those Argyll markers, and the Northern Irish (not Sco-Ire) clusters. However, that is truly unsurprising, given these regions have been interconnected for thousands of years, well, at least two-thousand I think - but that's conservative IMO. I'm surprised that the dendrogram gives a different perspective on this and that all the tools which gauge the relationships between the clusters don't agree upon these relationships. I would say that each cluster does exist in a continuum, or a range, absolutely. Ultimately I see this as another tool to analyze the relationship between these clusters. Analysis from Donegal down to England looks pretty straight forward to me, and could be representative of Insular isolation to continental influence of some kind, with continental influence increasing as it moves towards England.
Currently I'm quite interested in the charts found in the appendix which show the relationship of these clusters to seven ancient Icelandic Gaels. A very strong indication of a relationship between the Hebrides, Donegal, and Argyll, can be seen there as these regions all match two of the samples particularly strongly. It's all very much in line with the migratory history of people coming from the north of Ireland to the west and north-west of Scotland, but interesting to see it verified in multiple ways.
The t-SNE chart appears to show a very similar picture to PCA 1 and 2 with the exception that the axis shown by PC 1 appears to be more exaggerated. It looks to me like if you moved those more into one single line it would look very similar to PCA 1/2, what do you think about that?

fridurich
09-18-2019, 12:58 AM
Hey Fred, another long-winded response for you.. (perhaps you can tell I also enjoy staring at these little circles a bit too much?)

Just to reiterate in this case because we are not including the clusters that branch off significantly along PC1 I was saying that as all the clusters we are examining fall more or less along the same line PC 1 is not going to inform us of much for the purposes of what we are looking at here, so we can get that out of the way and start looking at the differences along PC 2. Surely the PCA is reflecting some kind of relationship between the Hebrides, and as I believe you have correctly pointed out, those Argyll markers, and the Northern Irish (not Sco-Ire) clusters. However, that is truly unsurprising, given these regions have been interconnected for thousands of years, well, at least two-thousand I think - but that's conservative IMO. I'm surprised that the dendrogram gives a different perspective on this and that all the tools which gauge the relationships between the clusters don't agree upon these relationships. I would say that each cluster does exist in a continuum, or a range, absolutely. Ultimately I see this as another tool to analyze the relationship between these clusters. Analysis from Donegal down to England looks pretty straight forward to me, and could be representative of Insular isolation to continental influence of some kind, with continental influence increasing as it moves towards England.
Currently I'm quite interested in the charts found in the appendix which show the relationship of these clusters to seven ancient Icelandic Gaels. A very strong indication of a relationship between the Hebrides, Donegal, and Argyll, can be seen there as these regions all match two of the samples particularly strongly. It's all very much in line with the migratory history of people coming from the north of Ireland to the west and north-west of Scotland, but interesting to see it verified in multiple ways.
The t-SNE chart appears to show a very similar picture to PCA 1 and 2 with the exception that the axis shown by PC 1 appears to be more exaggerated. It looks to me like if you moved those more into one single line it would look very similar to PCA 1/2, what do you think about that?

Sktibo, I looked at the data on the Icelandic ancient Gaels, and it was very interesting. What you are saying about it seems very likely. To me, the T-sne chart and PCA 1 and 2 show some similarities, and in some ways look different. I tried to imaging moving the axes of the t-sne chart into a single line, and it is hard for me to imagine what the results would be like.

Kind Regards
Fred

fridurich
09-27-2019, 02:08 AM
I would like some input about the N Ire-Sco cluster. The majority of the green/yellow circular symbols for this group are close to Central/Southwest Scotland. There are only 3 that I see in Ireland/Northern Ireland and one of those seems to be close to Dublin. Do you all think this cluster originated in Ireland and came to Scotland during the 19th Century? But if so, would almost the whole group immigrate, leaving such a small number behind in Ireland? Could this be an ancient immigration of Irish into Scotland? Or, could it be a mixture of Irish 19th Century immigration and ancient immigration into Scotland?

This may seem like less likely, but could it represent an ancient immigration of Scots into Ireland? I know the dendogram has the N Ire-Sco cluster rooted in Ireland, but why are there so few of them in Ireland/Northern Ireland today? The N Ire-Sco cluster does show affinity with the Irish being sandwiched between the N Ireland and the Munster cluster on the first PCA chart on page 8 of the Appendix. However, the N Ire-Sco cluster is close to many of the Sco-Ire cluster symbols in Scotland. Is it possible that the N Ire-Sco cluster could also be Planters that came to Ulster in the 17th and 18th Centuries, but not near as many of the N Ire-Sco group came over as the Sco-Ire group?

Kind Regards
Fred

Rufus191
10-07-2019, 03:26 PM
I would like some input about the N Ire-Sco cluster. The majority of the green/yellow circular symbols for this group are close to Central/Southwest Scotland. There are only 3 that I see in Ireland/Northern Ireland and one of those seems to be close to Dublin. Do you all think this cluster originated in Ireland and came to Scotland during the 19th Century? But if so, would almost the whole group immigrate, leaving such a small number behind in Ireland? Could this be an ancient immigration of Irish into Scotland? Or, could it be a mixture of Irish 19th Century immigration and ancient immigration into Scotland?

This may seem like less likely, but could it represent an ancient immigration of Scots into Ireland? I know the dendogram has the N Ire-Sco cluster rooted in Ireland, but why are there so few of them in Ireland/Northern Ireland today? The N Ire-Sco cluster does show affinity with the Irish being sandwiched between the N Ireland and the Munster cluster on the first PCA chart on page 8 of the Appendix. However, the N Ire-Sco cluster is close to many of the Sco-Ire cluster symbols in Scotland. Is it possible that the N Ire-Sco cluster could also be Planters that came to Ulster in the 17th and 18th Centuries, but not near as many of the N Ire-Sco group came over as the Sco-Ire group?

Kind Regards
Fred
I presume you mean blue/yellow circular symbols rather than green. It definitely looks like a cluster originating in Scotland to me. Dublin had had enough Scottish settlement to have two largish Presbyterian congregations, the first, originally known by the name of Capel Street, later St. Mary's Abbey or the Abbey Presbyterian Church which began in 1672

https://archive.org/details/historyofcongreg00kill/page/128

https://en.wikipedia.org/wiki/Abbey_Presbyterian_Church,_Dublin

and the slightly later congregation at Usher's Quay, later Ormond Quay, which began in 1717

https://archive.org/details/historyofcongreg00kill/page/130

https://en.wikipedia.org/wiki/Ormond_Quay_Presbyterian_church

If you zoom into Dublin, there is also a (255) yellow-blue square hiding there which probably is also a result of these long standing Scottish congregations there.

fridurich
10-08-2019, 12:03 AM
I presume you mean blue/yellow circular symbols rather than green. It definitely looks like a cluster originating in Scotland to me. Dublin had had enough Scottish settlement to have two largish Presbyterian congregations, the first, originally known by the name of Capel Street, later St. Mary's Abbey or the Abbey Presbyterian Church which began in 1672

https://archive.org/details/historyofcongreg00kill/page/128

https://en.wikipedia.org/wiki/Abbey_Presbyterian_Church,_Dublin

and the slightly later congregation at Usher's Quay, later Ormond Quay, which began in 1717

https://archive.org/details/historyofcongreg00kill/page/130

https://en.wikipedia.org/wiki/Ormond_Quay_Presbyterian_church

If you zoom into Dublin, there is also a (255) yellow-blue square hiding there which probably is also a result of these long standing Scottish congregations there.

Thanks so much for answering! Yes, I mean the yellow circles with the blue stroke/border around them. I just looked at the T-sne chart on the PDF and it has the best graphic quality of all. On the first PCA chart on P. 8 in the Index, it looks kind of like the
N Ire-Sco cluster has kind of greenish circles, perhaps the blue borders had blended with the yellow circles, making them appear green.

What you are saying about the N Ire-Sco cluster being Scots that immigrated to Ireland, such as in Dublin sounds like a possibility to me. Something interesting is the N Ire-Sco cluster shows strong affinity with at least some of the Irish clusters, like on PCA 1 of the first PCA chart on page 8 of the Index. Do you think this immigration to Ireland happened in the 17th Century, or before then?

On the T-sne chart, the N Ire-Sco cluster is right by the N Ireland cluster and appears to be reaching out to the Sco-Ire cluster, almost touching it. I think the Sco-Ire cluster are descendants of the Planters of Ulster of the 17th and 18th Century because of where their blue (with yellow border) squares appear in Southwest Scotland and in Northern Ireland.

Kind Regards
Fred

fridurich
10-11-2019, 02:53 AM
Here is an interesting video, Nov. 27, 2018, of Ed Gilbert apparently talking about the results of the Scotland part of the study that would later include more results for the Isle of Man when the Scotland/Isle of Man autosomal DNA study was published.

https://www.youtube.com/watch?v=DerzHJW_Fy0

Great talk, but I wish Ed had gone into more detail on some things, such as the Scotland clusters that had the most genetic relatedness to Native Irish clusters. Also, would have been nice to hear him say when he thought the N Ire-Sco cluster entered Scotland, if it was a movement from Ireland to Scotland, and not vice versa. The dendogram has the N Ire-Sco cluster rooted with the Irish. But is it possible it could be a movement of Scots into Ireland? If it was a movement from Ireland into Scotland is it likely they would pour almost their whole population into Scotland, leaving few behind in Ireland? The N Ireland cluster had a number of people that immigrated close to where the N Ire-Sco cluster was in Scotland, but the N Ireland cluster didn't empty out their whole population into Scotland.

Kind Regards
Fred

Sikeliot
10-11-2019, 03:12 AM
I figured out in the Insular Celtic paper, btw, which regions in Ulster the "North Leinster/Ulster" cluster occupies. It has a presence in Meath, Longford, and to some extent around Dublin and Louth, but other clusters are also present in all of these regions. I wonder why some of Leinster falls into this cluster to begin with but I never questioned it. Does anyone know whether that region experienced isolation?

sktibo
10-11-2019, 05:09 AM
Here is an interesting video, Nov. 27, 2018, of Ed Gilbert apparently talking about the results of the Scotland part of the study that would later include more results for the Isle of Man when the Scotland/Isle of Man autosomal DNA study was published.

https://www.youtube.com/watch?v=DerzHJW_Fy0

Great talk, but I wish Ed had gone into more detail on some things, such as the Scotland clusters that had the most genetic relatedness to Native Irish clusters.


In regards to this question it appears with certainty that in the Scottish paper N-Ire-Sco has the most genetic relatedness to the modern native Irish clusters, in every form of analysis it clusters with the Irish clusters, and even resembles them in the admixture chart. After this, it looks to me like Argyll could be the next closest due to it's position in PCA 1 and 2, the t-SNE chart, and the admixture chart. The Hebrides appear to be a bit more distant probably due to increased Scandinavian admixture, and I wouldn't be surprised if a bit of isolation added to that distance as well.
After that it starts to look less clear.. in my estimation, considering the admixture charts, dendrogram, t-SNE and PCA 1/2, I would say it's likely that the Inner Hebrides are next and then the Outer Hebrides are the next closest. After those, I'm not sure if it's N Scotland (Inverness), Sco-Ire, or Isle of Man.. maybe Sco-Ire? Back to squinting at the charts for that one...

On the topic of the origins of N-Ire-Sco, (is it Scottish or Irish in origin?) I don't think it is necessarily the case that what we are seeing is reflective of some migration. I'm not saying it isn't either.
It seems to be that our position or outlook of interpretation of these charts coming from a genealogical background, as nearly all of us (or maybe all of us) are, we tend to impose our ideas of migrations or origin points upon these markers, wanting to know how they got there. However, I personally don't think this is always the case, because what is really going on here is that we are seeing the markers which are grouped based on similarity to one another, and named based upon geographic position of the samples afterwards.
It might be that there was no migration, event, or movement that led to these markers, and it just happens that they form this cluster together with the data-set that is presented. Previously I think the ones that appeared in Ireland in the IDA were considered to be part of the "Dublin" cluster. Now that we have the Scottish samples there isn't a Dublin cluster any longer. What would happen if even more samples were gathered from all over Ireland and Scotland? the whole thing could change again. However, we see some clusters that have "survived" all four studies, such as Cornwall or North Wales.
We want to analyze this sort of data based on our ideas about history and geography but it doesn't appear to hold up with every cluster. I think it could be that this cluster falls somewhere between Ireland and Scotland, currently more samples of it's type are found in Scotland, and that might be all we get to know about it. I'm not sure our ideas about history and geography will tell us about this one.

Nqp15hhu
10-11-2019, 06:10 AM
Fred, everybody in Northern Ireland has some percentage of Scottish ancestry. It just depends on the individual as to the total percentage.

His bias on that subject will determine his interest in it.

fridurich
10-12-2019, 09:00 PM
In regards to this question it appears with certainty that in the Scottish paper N-Ire-Sco has the most genetic relatedness to the modern native Irish clusters, in every form of analysis it clusters with the Irish clusters, and even resembles them in the admixture chart. After this, it looks to me like Argyll could be the next closest due to it's position in PCA 1 and 2, the t-SNE chart, and the admixture chart. The Hebrides appear to be a bit more distant probably due to increased Scandinavian admixture, and I wouldn't be surprised if a bit of isolation added to that distance as well.
After that it starts to look less clear.. in my estimation, considering the admixture charts, dendrogram, t-SNE and PCA 1/2, I would say it's likely that the Inner Hebrides are next and then the Outer Hebrides are the next closest. After those, I'm not sure if it's N Scotland (Inverness), Sco-Ire, or Isle of Man.. maybe Sco-Ire? Back to squinting at the charts for that one...

On the topic of the origins of N-Ire-Sco, (is it Scottish or Irish in origin?) I don't think it is necessarily the case that what we are seeing is reflective of some migration. I'm not saying it isn't either.
It seems to be that our position or outlook of interpretation of these charts coming from a genealogical background, as nearly all of us (or maybe all of us) are, we tend to impose our ideas of migrations or origin points upon these markers, wanting to know how they got there. However, I personally don't think this is always the case, because what is really going on here is that we are seeing the markers which are grouped based on similarity to one another, and named based upon geographic position of the samples afterwards.
It might be that there was no migration, event, or movement that led to these markers, and it just happens that they form this cluster together with the data-set that is presented. Previously I think the ones that appeared in Ireland in the IDA were considered to be part of the "Dublin" cluster. Now that we have the Scottish samples there isn't a Dublin cluster any longer. What would happen if even more samples were gathered from all over Ireland and Scotland? the whole thing could change again. However, we see some clusters that have "survived" all four studies, such as Cornwall or North Wales.
We want to analyze this sort of data based on our ideas about history and geography but it doesn't appear to hold up with every cluster. I think it could be that this cluster falls somewhere between Ireland and Scotland, currently more samples of it's type are found in Scotland, and that might be all we get to know about it. I'm not sure our ideas about history and geography will tell us about this one.

Sktibo, I pretty much agree with your assessment of which Scottish clusters are closest to the modern native Irish clusters. I also concur that the N Ire-Sco cluster is very close to modern native Irish clusters. I hadn't noticed that some of the N Scotland (Inverness) cluster were somewhat close to some of the Irish clusters. Thanks for pointing that out. When you say"... After those, I'm not sure if it's N Scotland (Inverness), Sco-Ire, or Isle of Man.. maybe Sco-Ire? Back to squinting at the charts for that one...", I think it may be the Sco-Ire cluster that is next closest to the modern native Irish clusters after the ones you had already mentioned in the sequence. But as you say, it's back to squinting on that one!

What makes it kind of hard to compare on a PCA chart, as we have discussed earlier, each cluster is in a blob or envelope, and while at one end of that blob, the cluster's symbols inside it may be close to the closest cluster symbols of another cluster's blob, at the opposite end of the first cluster's blob, those symbols may be far from the second cluster's blob!!! A continuum of sorts.

That is very interesting what you said about the N Ire-Sco cluster may not have immigrated and the reasons you give for that. I hadn't thought about that and I think that what you said is possible.

Kind Regards
Fred

fridurich
10-12-2019, 09:17 PM
Fred, everybody in Northern Ireland has some percentage of Scottish ancestry. It just depends on the individual as to the total percentage.

His bias on that subject will determine his interest in it.

Thanks for your reply. In addition to Gaelic Irish ancestry from Northern Ireland (O'Hair/O'Hare from County Down), I also have some Ulster Scot ancestry. I have some Armstrong ancestors who are said to have come from County Fermanagh and I have other lines that were probably Ulster Scot. Also, I have some Scottish ancestry through my Burnett (Aberdeenshire) and Small lines, and other lines that are either Ulster Scot or Scottish, as well as other lines that are English, etc.

Great to see that you have done some YDNA testing and interesting that you have found some Scottish and Border surnames that are kin to you!

Kind Regards
Fred

Nqp15hhu
10-13-2019, 12:42 AM
Thanks for your reply. In addition to Gaelic Irish ancestry from Northern Ireland (O'Hair/O'Hare from County Down), I also have some Ulster Scot ancestry. I have some Armstrong ancestors who are said to have come from County Fermanagh and I have other lines that were probably Ulster Scot. Also, I have some Scottish ancestry through my Burnett (Aberdeenshire) and Small lines, and other lines that are either Ulster Scot or Scottish, as well as other lines that are English, etc.

Great to see that you have done some YDNA testing and interesting that you have found some Scottish and Border surnames that are kin to you!

Kind Regards
Fred

Thank you. I am still trying to determine a place of origin.

My match surnames are from all over in GB, so there is no clear place. Very unfortunate!!

fridurich
10-14-2019, 02:34 AM
On the video I mentioned earlier at almost exactly 22:03 on the timeline, Ed Gilbert makes an interesting statement about what must be the N Ire-Sco cluster.

https://www.youtube.com/watch?v=DerzHJW_Fy0

Some of his words sounded indistinct to me on the computer, so I watched it on T. V., turning the volume way up and listed to it about 4 times to try to make sure I heard all that he said correctly. The video showed the T-sne chart about or during the time Ed was making his comment about the N Ire-Sco cluster:

33873

The following is exactly, or extremely close to what Ed said about the N Ire-Sco cluster:

"...and you got these individuals that are kind of a mixture between Irish and Scottish genetics kind of bridging those two islands together."

When you look at the t-sne chart, it does look kind of like the N Ire-Sco cluster (which is very close to most of the Irish clusters) is reaching out to the Sco-Ire cluster, almost touching it.

Kind Regards
Fred

Heber
03-15-2020, 11:12 AM
Update on ‘The Irish DNA Atlas – Revealing Irish History through Genetics' lecture by Dr. Edmund Gilbert of the Royal College of Surgeons in Ireland. Includes studies on Scotland DNA and Iceland Ancient DNA.

https://youtu.be/qKXIiyQJTJY

sktibo
03-15-2020, 07:33 PM
Thank you for the link Heber

The video cleared up why that admixture chart didn't make much sense to us (Chart B of the GLSI) because they were essentially trying to measure Scandinavian ancestry in Ireland, and more or less ignored the England/Wales/Scotland Admixture levels in Ireland - This begins around the 26 minute mark in the video. It was exciting to hear him talk about re-running this analysis once they have appropriate ancient samples to use! So they will get around to it!

Heber
03-15-2020, 10:16 PM
I have posted some of the slides from the video here:

https://pin.it/1Ku6daZ

Heber
05-03-2020, 10:57 PM
The genetic landscape of Scotland and the Isles

Britain and Ireland are known to show population genetic structure; however, large swathes of Scotland, in particular, have yet to be described. Delineating the structure and ancestry of these populations will allow variant discovery efforts to focus efficiently on areas not represented in existing cohorts. Thus, we assembled genotype data for 2,554 individuals from across the entire archipelago with geographically restricted ancestry, and performed population structure analyses and comparisons to ancient DNA. Extensive geographic structuring is revealed, from broad scales such as a NE to SW divide in mainland Scotland, through to the finest scale observed to date: across 3 km in the Northern Isles. Many genetic boundaries are consistent with Dark Age kingdoms of Gaels, Picts, Britons, and Norse. Populations in the Hebrides, the Highlands, Argyll, Donegal, and the Isle of Man show characteristics of isolation. We document a pole of Norwegian ancestry in the north of the archipelago (reaching 23 to 28% in Shetland) which complements previously described poles of Germanic ancestry in the east, and “Celtic” to the west. This modern genetic structure suggests a northwestern British or Irish source population for the ancient Gaels that contributed to the founding of Iceland. As rarer variants, often with larger effect sizes, become the focus of complex trait genetics, more diverse rural cohorts may be required to optimize discoveries in British and Irish populations and their considerable global diaspora.

https://www.researchgate.net/publication/335597095_The_genetic_landscape_of_Scotland_and_th e_Isles

fridurich
05-21-2020, 12:57 AM
So all of the findings in Dr. Lara Cassidy's thesis will be available to read on May 23, 2020, including any M222 YDNA found in ancient Irish remains?

http://www.tara.tcd.ie/handle/2262/82960

Kind Regards
Fred

Dubhthach
05-21-2020, 10:41 AM
So all of the findings in Dr. Lara Cassidy's thesis will be available to read on May 23, 2020, including any M222 YDNA found in ancient Irish remains?

http://www.tara.tcd.ie/handle/2262/82960

Kind Regards
Fred

We'll have to wait and see what Trinity College do, I do note there is older thesis in that system that are open access. So it will be interesting to see what happens, worth remembering though that the 23rd is a Saturday so we might just have to wait until Monday.

sktibo
05-21-2020, 05:26 PM
It's been years for many of us who want to know about the nature of the early medieval Irish and it's hard to believe the answer might finally be here.

razyn
05-21-2020, 05:57 PM
Maybe the listed end date for the thesis embargo coincides with scheduled publication of a paper in Nature or somewhere, on or by that date (day after tomorrow)?

CillKenny
05-21-2020, 08:52 PM
So all of the findings in Dr. Lara Cassidy's thesis will be available to read on May 23, 2020, including any M222 YDNA found in ancient Irish remains?

http://www.tara.tcd.ie/handle/2262/82960

Kind Regards
Fred

I hope so but we did not hear comforting indications on this at Lara's talk last October at Genetic Genealogy Ireland. I seem to remember it being said that this date might be pushed back. Perhaps someone else who was there can show I am confused on this, which would be great actually.

fridurich
05-22-2020, 03:39 AM
We'll have to wait and see what Trinity College do, I do note there is older thesis in that system that are open access. So it will be interesting to see what happens, worth remembering though that the 23rd is a Saturday so we might just have to wait until Monday.

Thanks Dubhthach. I didn't think about the 23rd being a Saturday, so it could very well be Monday before we are able to see what the thesis consists of. I'm sure it will be interesting regardless, but since I'm confirmed YDNA - M222...S588>S603>BY3347...BY18204, I'm really interested in seeing if the thesis mentions any M222 in ancient/Medieval Irish remains, and if the DNA was good enough quality that several subclades downstream from M222 can be detected. If isotopic info is included, maybe some good hypothesis on where some of these clades originated can be arrived at.

Kind Regards
Fred

Jessie
05-22-2020, 04:00 AM
Thanks Dubhthach. I didn't think about the 23rd being a Saturday, so it could very well be Monday before we are able to see what the thesis consists of. I'm sure it will be interesting regardless, but since I'm confirmed YDNA - M222...S588>S603>BY3347...BY18204, I'm really interested in seeing if the thesis mentions any M222 in ancient/Medieval Irish remains, and if the DNA was good enough quality that several subclades downstream from M222 can be detected. If isotopic info is included, maybe some good hypothesis on where some of these clades originated can be arrived at.

Kind Regards
Fred

Yes I'm very interested as well as my father is M222 - S588 and from Roscommon which is I think where this ancient M222 was found in remains.

Dubhthach
05-22-2020, 05:53 PM
I'm not sure if the Ranelagh remains from Roscommon are in Lara's thesis. She mentioned them at the event in the RDS in Dublin last autumn and mentioned the research was been done out of Queen's in Belfast. If I recall right team been led by same people involved in Ballyhanna cemetery site.

Ranelagh site had about 800 remains covering 6th century right up to some later remains in late Medieval period. Like Ballyhanna it was discovered due to road construction. As was mentioned there was M222 (at least one S588) and DF21 (P314.2) in sampleset. I also believe in some of later set there was alsoa t least one U106 individual. I believe that dataset was made up of 40 individuals out of 800 remains, which have been also subjected to C-14 dating and isotope analyshttps://www.tii.ie/_internal/cimg!0/nvu0zm67h76ijhb2z595smtizrdvgfe

https://www.irishexaminer.com/remote/media.central.ie/media/images/z/zzzRingfortSiteRanelagh310317b_large.jpg

https://www.irishexaminer.com/remote/media.central.ie/media/images/r/RingfortSiteRanelagh310317_large.jpg

The site was reported in the media back in 2017, I believe the dig finished in October 2016
https://www.irishexaminer.com/viewpoints/analysis/800-medieval-bodies-found-during-hillfort-dig-446670.html#

alan
05-22-2020, 08:22 PM
I'm not sure if the Ranelagh remains from Roscommon are in Lara's thesis. She mentioned them at the event in the RDS in Dublin last autumn and mentioned the research was been done out of Queen's in Belfast. If I recall right team been led by same people involved in Ballyhanna cemetery site.

Ranelagh site had about 800 remains covering 6th century right up to some later remains in late Medieval period. Like Ballyhanna it was discovered due to road construction. As was mentioned there was M222 (at least one S588) and DF21 (P314.2) in sampleset. I also believe in some of later set there was alsoa t least one U106 individual. I believe that dataset was made up of 40 individuals out of 800 remains, which have been also subjected to C-14 dating and isotope analyshttps://www.tii.ie/_internal/cimg!0/nvu0zm67h76ijhb2z595smtizrdvgfe

https://www.irishexaminer.com/remote/media.central.ie/media/images/z/zzzRingfortSiteRanelagh310317b_large.jpg

https://www.irishexaminer.com/remote/media.central.ie/media/images/r/RingfortSiteRanelagh310317_large.jpg

The site was reported in the media back in 2017, I believe the dig finished in October 2016
https://www.irishexaminer.com/viewpoints/analysis/800-medieval-bodies-found-during-hillfort-dig-446670.html#

Where did you get that info about the yDNA? Its of personal interest as I am a DF21-P314.2

Dubhthach
05-22-2020, 10:05 PM
Where did you get that info about the yDNA? Its of personal interest as I am a DF21-P314.2

I was at the talk, I took some notes, just had a look back at them today and see mention of P314.2 with regards to Ranelagh.

Dubhthach
05-23-2020, 12:53 AM
The Thesis has dropped by the way. There are at least three M222+ samples mentioned in it


Within Ireland a large diversity in R1b-DF13 subclades is seen in both the Bronze and Iron Ages, with an expectedly higher number of downstream mutations observed for Iron Age samples (See Electronic Data Table S5). Notably, all northwestern Irish Iron Age individuals sampled (Ballyglass44, Derrynamanagh08 and Derrynamanagh09) were seen to belong to the R1b-M222 subclade or a lineage leading directly to it. This haplogroup peaks in northwestern Ireland today and has been previously associated with the early Medieval Uí Néill dynasty of the region (Moore et al. 2006). Intriguingly, the two southern Iron Age individuals sampled, Courtmacsherry37 and Ballybunnion54, also both share a subclade, R1b-CTS3087. Given the known emphasis placed on patrilineal descent in Gaelic Ireland, denser surveys of Iron Age Y chromosomal variation on the island may contribute greatly to the understanding of territorial boundaries and patronymic surname distributions that were recorded during the early historical period.



Derrynamanagh, Co. Galway
Site Description
The remains of several late Iron Age burials were discovered during farm works at a large rath (GA085- 045), in Derrynamanagh, Co. Galway and excavated by Professor Etienne Rynne in 1969. The site is approximately 12.5km west of Athenry and only 4km from the rath of Feerwore, the proposed original site of the Turoe Stone. It is situated within a steep bend of the Raford river and surrounded by bog. The rath consists of a bank and external earthwork, which form a subcircular enclosure. The burials included two single inhumations in pits and one multiple burial in a stone lined cist. An unpublished report of the skeletal material, written by Professor Stephen Shea of the NUIG Anatomy Dept. (1970), identified the remains within each of the single burials to be of young children, approximately 2-3 years in age, while the multiple burial contained the remains of at least 13 adults.

Sampling
Nine petrous bones, DM01-DM09, were located and sampled from the site (Fig. I.15)
Further Notes
Extensive kinship was found among individuals at the site (Appendix Two; Fig. II.9). A pair of sisters were identified, DM02 and DM05, to whom DM09 is a second degree male relative. Given all three individuals share the same mitochondrial haplogroup, it is likely DM09 is related to the sisters via the female line. The possible scenarios include a maternal uncle, a nephew (another sister’s child), a half- sibling who shares the same mother, but not a grandfather or grandchild. All three individuals share a fourth degree relative, DM04, while DM09 shares a separate fifth degree relative, DM08.


--


Ballyglass Middle , Co Mayo
Site Description
The site most likely represents an Iron Age ringbarrow re-used for burial approximately 400 to 600 AD. The skeletal remains were disturbed during quarrying and original burial positions cannot be determined. The remains of an adult male and female, as well as a juvenile of 3-5 years were identified. Radiocarbon determinations for the adult male and female returned dates of 1765±70 BP (80-420 AD; OxA-3871) and 1590±60 (330-610 AD; OxA-3872) respectively.



The most likely candidate migrants from an archaeological perspective are two eastern individuals from Knowth (175-50 cal BC; 86-252 cal AD), whose burial rites are common in Britain and almost unknown in Ireland during the period (McGarry 2010). This interpretation is supported by their placement away from the main distributions of Irish Iron Age and modern variation. However, it must be noted that a number of other Irish Iron Age individuals place even further towards the British cluster, including three unrelated samples from Ballyglass Middle, the only Irish Iron Age site sampled that shows clear ancestral homogeneity among burials. The site itself is unusual in the relatively early date retrieved from unburnt bone (80-420 cal AD), at a time cremation was ubiquitous in Ireland. Two individuals from Derrynamanagh also show increased British affinities, but again the site shows wide differentiation on PCs 2 and 4, despite the close kinship among a number of samples (not included here). Previously undetected relatedness is also seen here between Derrynamanagh04 and Derrynamanagh05, later confirmed through IBD kinship analysis, which revealed the pair to be fourth degree relatives (Appendix II).


http://www.tara.tcd.ie/handle/2262/82960

fridurich
05-23-2020, 12:56 AM
I'm not sure if the Ranelagh remains from Roscommon are in Lara's thesis. She mentioned them at the event in the RDS in Dublin last autumn and mentioned the research was been done out of Queen's in Belfast. If I recall right team been led by same people involved in Ballyhanna cemetery site.

Ranelagh site had about 800 remains covering 6th century right up to some later remains in late Medieval period. Like Ballyhanna it was discovered due to road construction. As was mentioned there was M222 (at least one S588) and DF21 (P314.2) in sampleset. I also believe in some of later set there was alsoa t least one U106 individual. I believe that dataset was made up of 40 individuals out of 800 remains, which have been also subjected to C-14 dating and isotope analyshttps://www.tii.ie/_internal/cimg!0/nvu0zm67h76ijhb2z595smtizrdvgfe

https://www.irishexaminer.com/remote/media.central.ie/media/images/z/zzzRingfortSiteRanelagh310317b_large.jpg

https://www.irishexaminer.com/remote/media.central.ie/media/images/r/RingfortSiteRanelagh310317_large.jpg

The site was reported in the media back in 2017, I believe the dig finished in October 2016
https://www.irishexaminer.com/viewpoints/analysis/800-medieval-bodies-found-during-hillfort-dig-446670.html#

Very interesting and great photos!!

Kind Regards
Fred

razyn
05-23-2020, 03:48 AM
Maybe the listed end date for the thesis embargo coincides with scheduled publication of a paper in Nature or somewhere, on or by that date (day after tomorrow)?

It was released at midnight. Dubhthach posted the link in New Papers.

MitchellSince1893
05-23-2020, 05:37 AM
No surprise that all the L151 subclades were L21 (in case anyone was looking for ancient U106, DF27, U152 etc).

FionnSneachta
05-23-2020, 10:43 AM
The most recent SNP in my dad's line is Z2961. The samples positive for this SNP are the following:
DM08 Derrynamanagh, Co. Galway (Late Iron Age)
DM09 Derrynamanagh, Co. Galway (Late Iron Age)

My dad's paternal anestors were from around the Galway-Roscommon border. However, both these samples are also positive for M222 with both being DF104 so they're not potential ancestors anyway! The data can be accessed here: https://docs.google.com/spreadsheets/d/1mk9pMMUbChzyW8CwVUYgokVL4iv83WBAKdIf3pWXJnw/edit#gid=1656941323

Webb
05-23-2020, 01:42 PM
No surprise that all the L151 subclades were L21 (in case anyone was looking for ancient U106, DF27, U152 etc).

There were two samples that were P312 and P311 but it doesn't state in the paper whether this was because of a poor sample or testing limitations.

Nqp15hhu
05-23-2020, 03:30 PM
No L513’s or L193’s?

Caledonian
05-23-2020, 03:37 PM
No L513’s or L193’s?

I may be wrong but L193 is mainly a Scottish subclade isn't it? that said I was curious to see if any L1335 or L1065 would show up as it may have answered questions as to whether it is of Dalriadan origin, but it's not there either from the looks of it.

Dubhthach
05-23-2020, 05:18 PM
No L513’s or L193’s?

The paper mentions at least two L513+ from Munster:

Intriguingly, the two southern Iron Age individuals sampled, Courtmacsherry37 and Ballybunnion54, also both share a subclade, R1b-CTS3087.

https://www.ytree.net/DisplayTree.php?blockID=222



Courtmacsherry, Co. Cork
Site Description
The site consists of a lintel grave, within which the extended inhumation of an adult male (1966:38) was found. There were no accompanying artefacts. A sample of bone yielded a radiocarbon determination of 1445±45 BP (GrA-24500), or 540-662 AD (Cahill & Sikora 2011).
Sampling
A single petrous bone with some attached temporal was identified as belonging to the site. However, it is stored under the registration number 1966:37, rather than 1966:38. As there are no other human remains associated with this site we presumed this to be the adult male individual and took the petrous bone for aDNA analysis (CM37; Fig I.13).




Ballybunnion, Co. Kerry
Site Description
The site consists of a long cist containing an inhumation burial. The excavation of this cist revealed two further inhumation burials, one within a lintel grave, and the other unprotected. Grave 2 contained the inhumed remains of an adult male (1987:54), which dated to 1400±35 BP, or 583-674 AD (Cahill & Sikora 2011).


For Geographic reference ~140km apart:

https://www.google.com/maps/dir/Ballybunion,+County+Kerry/Meelmane,+Courtmacsherry,+Co.+Cork/@52.0720069,-9.7488797,9z/data=!3m1!4b1!4m14!4m13!1m5!1m1!1s0x485ac5d0328099 c3:0xa00c7a99731d880!2m2!1d-9.67097!2d52.51108!1m5!1m1!1s0x4844594b253080cf:0x a00c7a99731f960!2m2!1d-8.7059339!2d51.6338554!3e2

Jon
10-19-2020, 06:57 PM
This is great for L513: I believe this is the first aDNA L513 found in Ireland or Scotland?

Munster is still a hotspot for L513 in present-day Ireland I think. Like the O'Sheas. I wonder what this might mean, if anything, for the folks that were L513 back in the Iron Age, in terms of clans, groups, etc?

Jon
10-20-2020, 05:41 AM
Also: 140km apart, sharing a HG. That seems pretty big news, no? Like this was a big regional HG in Munster in the 6th-7th centuries? She says "intriguingly"...which is science speak for WOW!!??

fridurich
10-29-2020, 01:51 AM
Does anyone know if the Irish DNA Atlas people are going to do any projects or have any joint projects with anybody else anytime soon? I really liked the study published about a year ago that included Scotland and the Isle of Man.

Kind Regards
Fridurich

Heber
11-10-2020, 09:36 PM
Update on the Irish DNA Atlas
Dr. Ed. Gilbert
Genetic Landscape between Ireland and Scotland

In our 2017 research paper, the Irish DNA Atlas revealed the fine-scale details of the genetic landscape of Ireland, demonstrating subtle regional differences in Irish genetics which were reflective of historical boundaries. Furthermore, we demonstrated considerable ‘British’ ancestry in the north-east of Ulster, but due to the limits of our British reference samples were unable to decisively differentiate between a northern English affinity or a southern Scottish affinity. Working with researchers from the University of Edinburgh and other Scottish academic institutions, we were able to apply the research methodology developed at RCSI to Scottish DNA samples with regional genealogies akin to the Irish DNA Atlas. This allowed us to investigate the genetic landscape of Scotland and its Isles, and Ireland together1 (please find attached to this email). We showed that with additional Scottish references, Irish individuals with high ‘British’ ancestry within the north-east of Ulster show more affinity to references from the south-west of Scotland, not with the north of England (see below). This work also demonstrated the genetic distinctiveness of the various isles around Scotland to its west and the north-east. The results from these Scottish Isles highlights the need for further, focussed, recruitment from the Irish Isles – something that the Irish DNA Atlas is keen to pursue. Lastly the Irish DNA Atlas helped show that, DNA extracted from the remains of the first Gaelic settlers of Iceland, when compared to modern individuals from across Britain and Ireland, showed most affinity to inhabitants of the Scottish western Isles and/or the north-west of Ireland in Donegal.

Legend (above)

Genetic clustering of 2,544 Irish and British individuals based on their genetic affinities.

(A) A tree showing 43 groups, or ‘clusters’, of individuals grouped by genetic affinity. The tree shows each cluster grouped with other similar clusters on shared branches.

(B) A map of Ireland and Britain, with each point showing one individual placed at the average geographic location of their recent ancestors. Each point is colour and shape coded according to clusters listed in (A).


https://drive.google.com/file/d/1jwxHAXv1YJhm1O6Ychg5lUzGmfnH8Zq-/view?usp=drivesdk


The update letter was issued to participants in the Irish DNA Atlas study on 9th November. Here is the rest of the update.

Doctoral Defence

In addition to continuing to provide research materials for new scientific publications, the Irish DNA Atlas has provided the basis of doctoral thesis research at the Royal College of Surgeons in Ireland. In June 2019 Edmund Gilbert successfully defended his research thesis, largely based on his work with the Irish DNA Atlas. As well as this, your participation in the Irish DNA Atlas has also provided the basis for other post- and under-graduate research projects and will continue to provide a resource for future research, study, and education.

Future Research

The Irish DNA Atlas has helped shape our understanding of the current, common, genetic landscape of Ireland. With this research established we are now progressing to further research questions. The first is investigating rare genetic variation. Up until now the Irish DNA Atlas has studied common variation within Ireland. These mutations are ‘common’ in that they are present in more than 5% of the individuals studied. In conjunction with the analyses we have performed, this common genetic variation has been invaluable in understanding the impact of history on the Irish genetic landscape (which we have reported in the first Irish DNA Atlas paper2). However, by its frequent nature, common mutations are generally shared between broad groups of individuals, i.e. across continents, and across Ireland and Britain. It is only by studying hundreds of thousands of such genetic variants together that we can infer different genetic regions. In contrast, rare variation is only found in limited number of individuals because rarer mutations are typically more recent in age, and therefore act as footprints of recent genetic processes, such as those specific to Ireland. Furthermore, rare genetic mutations tend to be more geographically localised, thus specific to specific regions. Rare variation is therefore more informative about recent and local genetic differences and would reveal novel insights into the genetic landscape of Ireland.

Working with collaborators at deCode Genetics, experts of this type of variation in Iceland, we aim to study geographically localised rare genetic variation across the island of Ireland. Furthermore, analysis of rare variation shared between Ireland and neighbouring populations may inform on recent subtle migratory links. This rare variation will be generated across the genome, with high resolution genetic data also generated on Irish DNA Atlas Y-chromosomes and mitochondrial genomes, which we hope will be of interest. Lastly, whilst we have an accurate picture of the fine-scale genetic landscape of Ireland, we have a less clear picture of the demographic history of Ireland, i.e. how large was the population through different periods and if there is evidence of population contraction. These demographic characteristics impact a population’s genetic variation; therefore, this research will further our understanding of both history and disease risk within Ireland.





As the project evolves, and when new data and studies from the Irish DNA Atlas are reported, we will continue to keep you updated. In addition to our work with deCODE Genetics, the Irish DNA Atlas is also collaborating with academic research institutions in Ireland, providing a valuable reference of Irish genetics. Currently the project is working with researchers at Letterkenny Institute of Technology, and Trinity College Dublin.

Furthermore, we wish to again thank you for your ongoing participation and hope you have found this newsletter of interest. Please note that you are free to withdraw from this study at any point, without giving a reason. Please contact us below if you wish to do so and we will destroy your sample and associated data. If you have questions about the study or the points discussed above, please do not hesitate to contact either Prof. Gianpiero Cavalleri or myself.

This research has been funded through a Career Development Award from Science Foundation Ireland. RCSI is ranked among the top 250 (top 2%) of universities worldwide in the Times Higher Education World University Rankings (2018) and its research is ranked first in Ireland for citations. It is an international not-for-profit health sciences institution, with its headquarters in Dublin, focused on education and research to drive improvements in human health worldwide. RCSI is a signatory of the Athena SWAN Charter.

Rufus191
11-12-2020, 05:58 PM
Ed did a talk about some of the updates about 7 months ago

https://www.youtube.com/watch?v=qKXIiyQJTJY

Nqp15hhu
04-12-2021, 01:46 AM
A lot of fuss is made about the English contribution to the gene pool here in Northern Ireland, however, I am not seeing much evidence of it in my ancestrydna matches.

In my Northern Irish matches most people have no English Dna whatsoever, the highest I can find is 7%.

Given the largely English Plantation in Fermanagh, I wonder if a Protestant from Fermanagh would score a high English percentage?

Chaz
05-17-2021, 04:38 PM
Just found this thread from another Y-DNA thread.

I've just been put into a new marker called FGC15275 under R-Y64389 which falls heavily under the Irish names of Durkin, Mulvihill, O Connel and a few others. Problem for me, I dont know my paternal line as I'm adopted but know my birth mom. Smith has also now joined this group of surnames. Was wonder if anyone could take a look at my Big Y 700 data and give a steer? I'm happy to pay for anyone's time in this regard if needed.

I'm the unknown 776917 kit here https://www.ytree.net/DisplayTree.php?blockID=626

Thanks

Chaz
05-17-2021, 04:39 PM
Well I imagine that like way we have a mix of L21 (eg. we have men with obvious post-1169 surnames that are L21+) it's probably the same case with DF27. I will admit I'm not familiar with DF17 and it's current status but when I filter in the Ireland project for men who have a DF17+ result I see a mix for example:

Joyce -- Welsh origin surname in an Irish context




Durkin:



Mulvihill:



What's interesting is that if the Mulvhill family were indeed members of Síol Muireadhaigh (of Uí Briúin Aí of the Connachta) ye'd expect a M222+ and probably A259+ result. So there might be more than one family there. Of course given that Durkin is also a North Connacht surname it's interesting seeing DF17+ show up in it as well.

It's interesting actually seeing the spilt between those names and Joyce here:
http://www.ytree.net/DisplayTree.php?blockID=624&star=false

That's fairly old branching point.

Hi, just posted above, these names seem to be very common in my Y-DNA, was hoping you might take a look and comment on my Y-DNA placement?

https://www.ytree.net/DisplayTree.php?blockID=626

I'm Unknown / Kit 776917.

Thank you.

Chaz
05-17-2021, 07:52 PM
For info, my baptism cert marks my father's name as Chris Christian however we can find no trace of such a man. We also have an example of a Sullivan surname person, marrying a protestant female and changing their surname to Christian. My Y-DNA also doesnt match Christian from Ilse of Man, a contingent went to South Africa but no match.

J1 DYS388=13
05-18-2021, 04:08 AM
What denomination and country is that baptism certificate from?

David Mc
05-18-2021, 06:51 AM
Hi, just posted above, these names seem to be very common in my Y-DNA, was hoping you might take a look and comment on my Y-DNA placement?

https://www.ytree.net/DisplayTree.php?blockID=626

I'm Unknown / Kit 776917.

Thank you.

Your terminal SNP (Y64389) has an estimated TMRCA of 340 BC. That makes it difficult to say anything with any degree of certainty. A lot of time has passed since then, and a lot has happened vis a vis population movements.

Having said that, the fact that your fellow Y64389 is of Irish extraction, and given that the SNP immediately above (R-FGC14114, with a TMRCA of 710 BC) has three matches from Irish background who also have clearly Irish surnames, makes me think that you can be reasonably confident that your father was also of Irish extraction. I'm not speaking of certitude, mind you, just of probability.

As hard as it is, sometimes-- and it is hard-- the kinds of answers you're looking for often come only with patience and time. Eventually the right person will do a test and provide you the match you're looking for.

Chaz
05-18-2021, 09:07 AM
What denomination and country is that baptism certificate from?

South Africa.

Chaz
05-18-2021, 09:09 AM
Your terminal SNP (Y64389) has an estimated TMRCA of 340 BC. That makes it difficult to say anything with any degree of certainty. A lot of time has passed since then, and a lot has happened vis a vis population movements.

Having said that, the fact that your fellow Y64389 is of Irish extraction, and given that the SNP immediately above (R-FGC14114, with a TMRCA of 710 BC) has three matches from Irish background who also have clearly Irish surnames, makes me think that you can be reasonably confident that your father was also of Irish extraction. I'm not speaking of certitude, mind you, just of probability.

As hard as it is, sometimes-- and it is hard-- the kinds of answers you're looking for often come only with patience and time. Eventually the right person will do a test and provide you the match you're looking for.

Thanks and not an unexpected answer. Just looking for further validation / clues. My father (we have a phased DNA kit for him) certainly shows links to South Africa on his maternal side. So the estimation is that his line entered the SA genepool probably 1800-1900. The '1820 Settlers' is when most British (UK) people went to SA. What we cant find is certainty of name.

Heber
02-13-2022, 09:14 AM
Interesting lecture this week at the GSI by Dr. Ed Gilbert on the Irish DNA Atlas and the Newfoundland Ancestry Project

“Newfoundland and Labrador: A mosaic founder population of an Irish and British diaspora from 300 years ago, will be given by Dr Edmund Gilbert.”

Irish DNA Atlas Project Update
New PhD student, Ashwini Shanmugam
Expanded dataset of 7,000 individuals (Britain and Ireland)
Half with Irish Ancestry
Collaboration with TCD Smurfit Lab
New Dataset structure including Isle of Man and Orkney’s
New Analysis tools including t-SNE
New Structure across Britain and Ireland
Irish and Britain Demography

Newfoundland Ancestry Project
Covers Newfoundland and Labrador
Settlement in 18th and 19th Century, Fishermen and Families
SW Britain (Devon and Cornwall) and SE Ireland (Wexford and Waterford)
Protestant and and Catholic Communities
English and Gaelic Speakers
Combined Irish DNA Atlas and Britain DNA Analysis
Collaboration with local Industry partner
Analysis complete, submitting for publication

Heber
04-13-2022, 12:06 PM
https://www.biorxiv.org/content/10.1101/2022.04.01.486593v2

Newfoundland and Labrador: A mosaic founder population of an Irish and British diaspora from 300 years ago
View ORCID ProfileEdmund Gilbert, Heather Zurel, Margaret E. MacMillan, Sedat Demiriz, Sadra Mirhendi, Michael Merrigan, Seamus O’Reilly, Anne M. Molloy, Lawrence C. Brody, Walter Bodmer, Richard A. Leach, Gerald Mugford, Ranjit Randhawa, J. Claiborne Stephens, Alison L. Symington, Gianpiero L. Cavalleri, Michael S. Phillips
doi: https://doi.org/10.1101/2022.04.01.486593

This article is a preprint and has not been certified by peer review [what does this mean?].
0000006

AbstractFull TextInfo/HistoryMetrics Preview PDF
Abstract

The founder population of Newfoundland and Labrador (NL) is a unique genetic resource, in part due to geographic and cultural isolation, where historical records describe a migration of European settlers to NL in the 18th and 19th centuries. However, the fine-scale genetic structure and ancestry of the population has not been well described. Here, we leverage dense genome-wide SNP data on 1,807 NL individuals to reveal fine-scale genetic structure in NL that is clustered around coastal communities and correlated with Christian denomination. We show that the majority of NL European ancestry can be traced back to the south-east and south-west of Ireland and England. We date a substantial population size bottleneck approximately 10-15 generations ago in NL, associated with increased haplotype sharing and autozygosity. Our results elucidate novel insights into the population history of NL and demonstrate evidence of a population conducive to further genetic studies and biomarker discovery.