An Irish Sampler

Born a Duffy, and extremely proud of her Irish heritage, it was no surprise when FTDNA’s MyOrigins returned a 99% British Isles result for my grandmother Gladys.  After seeing she also returned a 1% Asia Minor result, I decided to compare her ancestry results to her group of Irish cousins to see what other ancestry they might or -might not- have in common.

Featured image

FTDNA is the go-to company for international testers, and Gladys has many matches living in Ireland, the UK, New Zealand and Australia.  Some have also uploaded their raw data to Gedmatch, which provides the opportunity to use advanced admixture tools to compare cousins who are either completely or predominately Irish.  Two of the most popular admixture tools for Europeans, Eurogenes K13 and Dodecad V3 have also published the average results of their Irish samples.  Those results are entered in the last column of each table*.

Our Cousins

  • Gladys and CK are known second cousins through the Kane family from Castlebar, County Mayo. Gedmatch predicts: 2.9 generations to their common ancestor
  • Bridget is a cousin with an ancestor upstream of those shared by Gladys and CK, through the Kane/Forrestal families from Castlebar, Co Mayo. Gedmatch predicts a shared ancestor at 4.8.
  • Sean is a Mackin cousin from Northern Ireland. Gedmatch predicts a shared ancestor at 4.3 generations
  • Dennis belongs to a group of Northern Irish cousins. Their common ancestor is estimated at 4.8 generations
  • SK is a cousin with a shared ancestor upstream of the Mayo Kane ancestors shared by Gladys and CK. Gedmatch predicts a shared ancestor at 4.8 generations.
  • Maire is a Northern Irish X match cousin, and is probably shares an upstream ancestor of our Mackin or McCloskey ancestors. A common ancestor is estimated at 4.6 generations

Eurogenes K13

Detailed information about the Eurogenes K13 tool can be found here.  Information about all the Eurogenes tools can be found here.  A spreadsheet detailing the average results by population, including Irish samples, can be found here.

Featured image

Figure 1

In addition to the broad breakdowns illustrated in Figure 1, Eurogenes K13 also provides an ‘oracle’.  This utility breaks down ancestry even further.  The ‘oracle’ attempts to match the tester’s DNA with the DNA of testers from various populations.  The closer the genetic distance (example @3.699) the better the DNA matches the population.

For each of the cousins, the Oracle 4 utility was run and the top 5 populations are listed below:

Gladys

Using 1 population approximation:
1 Irish @ 3.699319
2 West_Scottish @ 4.227885
3 Orcadian @ 5.009747
4 Southwest_English @ 5.606747
5 North_Dutch @ 6.093086

CK

Using 1 population approximation:
1 South_Dutch @ 3.740911
2 Southeast_English @ 4.757324
3 West_German @ 4.948592
4 Southwest_English @ 6.509983
5 Orcadian @ 6.739860

Sean

Using 1 population approximation:
1 Irish @ 4.181266
2 Southwest_English @ 4.201335
3 West_Scottish @ 4.536220
4 Orcadian @ 4.843127
5 Southeast_English @ 5.027599

Maire

Using 1 population approximation:
1 Southeast_English @ 4.672073
2 West_Scottish @ 4.773833
3 Southwest_English @ 4.905573
4 Orcadian @ 5.280790
5 Irish @ 5.639902

Dennis

Using 1 population approximation:
1 Southeast_English @ 3.840825
2 Orcadian @ 4.607929
3 West_Scottish @ 4.867234
4 Southwest_English @ 5.397524
5 Irish @ 5.514109

Bridget

Using 1 population approximation:
1 Irish @ 4.226087
2 West_Scottish @ 4.787307
3 Southwest_English @ 5.084749
4 Orcadian @ 5.592003
5 Southeast_English @ 7.030124

SK

Using 1 population approximation:
1 Irish @ 2.728588
2 West_Scottish @ 3.401917
3 Orcadian @ 4.520518
4 Southwest_English @ 5.304957
5 North_Dutch @ 5.565923

The cousins with Kane ancestry from the port towns of Castlebar and Westport return nearly identical North Atlantic values that are lower than the Irish norm.  Their West Asian score is also almost identical and higher than the Irish average. Only two cousins, both from Castlebar return Sub Saharan results and they are slightly higher than the Irish average.  3 of the 4 Kane cousins return South Asian ancestry and it is also higher than the Irish norm.  Only one of the Northern Irish cousins returns Northeast African ancestry and it is at less than half of the Irish average. When looking at the oracle results, the Kane cousin who differs has a German ancestor, and the oracle matches him most closely to a South Dutch population, while the rest of the cousins most closely resemble to Irish population samples.

The Northern Irish cousin group shows more variety in general.  As seen in Figure 2, their Baltic scores are lower than the Irish average, and their West Asian scores are higher.

Dodecad V3

Detailed information about the Dodecad project can be found here.  A spreadsheet detailing the average results by population, including Irish samples, can be found here.

Featured image

Figure 2

Gladys

Using 1 population approximation:
1 Argyll @ 2.802603
2 N._European @ 3.152682
3 CEU @ 3.180986
4 Orcadian @ 3.840527
5 Orkney @ 4.039654

Bridget

Using 1 population approximation:
1 Argyll @ 1.383295
2 Orcadian @ 1.830747
3 Orkney @ 2.134289
4 CEU @ 2.153667
5 N._European @ 2.927236

Sean

Using 1 population approximation:
1 Argyll @ 5.047738
2 Orcadian @ 5.080379
3 N._European @ 5.158332
4 Orkney @ 5.187712
5 CEU @ 5.332207

CK

Using 1 population approximation:
1 CEU @ 6.219530
2 N._European @ 7.462104
3 Argyll @ 7.730925
4 Orcadian @ 8.337189
5 German @ 8.603835

Dennis

Using 1 population approximation:
1 Orcadian @ 0.850356
2 Orkney @ 1.155132
3 Argyll @ 2.134491
4 CEU @ 2.274293
5 N._European @ 3.791664

Maire

Using 1 population approximation:
1 Orkney @ 1.008123
2 Argyll @ 1.034154
3 Orcadian @ 1.225209
4 CEU @ 3.116245
5 N._European @ 3.219170

SK

Using 1 population approximation:
1 Orkney @ 2.687200
2 Orcadian @ 2.962321
3 Argyll @ 3.461854
4 N._European @ 4.474788
5 CEU @ 4.554245

Much like the Eurogenes K13 tool, all of the cousins were most closely matched to British Isles populations, except for the one cousin with a German ancestor.  Oddly, all of our Irish cousins have much more Eastern European and much less Western European than the Irish samples reported by Dodecad.

Mixed Modes and Calculator Effect

To produce ‘mixed mode’ results, the tools compare your DNA to sample populations. The algorithm then calculates the differences or similarities, or ‘genetic distance’ between your DNA and the population in question. Mixed mode results are the estimate of the 1-4 most similar populations to your DNA.   The algorithms also attempt to estimate which combined populations, or ‘admixtures’ most closely resemble your DNA.  This is often referred to as a ‘best fit’, or best estimate.

The Eurogenes K13 oracle 4 populations gives a 1st choice guess of Gladys having 4 Irish grandparents.  Its next best guess is an Irish parent and a parent 1/2 Irish and 1/2 Scottish.  Both guesses are very good considering her known ancestry. Dodecad’s algorithm didn’t do quite so well!

Gladys Eurogenes K13

Using 4 populations approximation:
1 Irish + Irish + Irish + Irish @ 3.699319
2 Irish + Irish + Irish + West_Scottish @ 3.780499
3 Irish + Irish + West_Scottish + West_Scottish @ 3.893456
4 Irish + Irish + Irish + Southwest_English @ 3.898322
5 Irish + Irish + Irish + Orcadian @ 3.937483

Gladys Dodecad V3

Using 4 populations approximation:
1 British_Isles + French + Hungarians + Argyll @ 0.643529
2 British + French + Hungarians + Argyll @ 0.673814
3 British_Isles + French + Hungarians + N._European @ 0.727025
4 British_Isles + French + Hungarians + Argyll @ 0.735014
5 French + Hungarians + Kent + Argyll @ 0.758637

This is a young, growing science, and it is important to realize results will improve as more samples are taken and compared. 

No tool is 100% accurate and results must be taken as part of a work in progress. Also, different tools work better with different populations.  None the less, they are another tool at your disposal to use as a general guideline in your research.  Eurogene’s Polako points out, “users from the UK often come out much more continental European than they should. Some of them actually believe that this is because they’re genetically more Norman or Saxon than the average Brit. Nope, the real reason is what I call the ‘calculator effect’.”  Dodecad’s  explains, “This is when the algorithm produces different results for people who are part of the original ADMIXTURE runs that set up the allele frequencies used by the calculators, than those who aren’t, even though both sets of users are of exactly the same origin, and should expect basically identical results.”   More information about the calculator effect can be found here.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s