Scots, how Celtic are they?

[Wilhelm]

SW Scots is an interesting choice for a sample, and could help us understand the admixture we expect from Brythonic populations, as SW Scotland is the traditionally Brythonic bit. Not surprisingly, they seem to be closest to the (also Brythonic) Cornish using this dataset (are you using a different one, Wilhelm?):

Southeast Baltic/Northern European/North Atlantic/East or North Eurasian/Sub-Saharan African/Southern European/Western European
Irish: 2/27/52/0/0/1/18
SW Scots: 4/30/48/0/0/0/17
Cornish: 2/29/48/0/0/4/17
Kentish: 4/40/35/0/0/5/16
Dutch: 5/45/29/0/0/10/10

I was using the latest EU7c, yours is EU7b, but anyways the patterns are pretty much the same.

[barbarian]

Middle Eastern haplogroups (G2a, J2, E1b1b, T)

is G2a middle eastern or caucasian?

[Taranis]

is G2a middle eastern or caucasian?

"Middle Eastern" is such a foggy term in my opinion. G2a probably originated in either Anatolia or the Caucasus, however. It's clear though that it did not originate in the Fertile Crescent or the Levante.

[DavidCoutts]

You might want to check FTDNA's Scotland Y-DNA Project. It is so far the largest and most detailed database for Scottish Y-DNA.

There are about 9% of Germanic haplogroup I1. The rest is less clear. Most of the R1a (8.5%) and I2b (4%) is probably Germanic too, although some of it might be Celtic. Haplogroup Q (0.5%) is surely of Scandinavian origin.

R1b is mixed Celtic and Germanic. Here is a small analysis.

Not all R1b members tested for subclades, but among those who did I counted 66 Germanic R1b1a2a1a1a, aka R1b-U106 (12.5%) and 470 predominantly Celtic R1b1a2a1a1b (87.5%). The latter includes :

- 60 Irish-Scottish R1b-M222 (12%)
- 31 purely Scottish R1b-S68 (6%)
- 31 Italo-Gaulish R1b-U152 (6%)
- 11 mostly Franco-Iberian R1b-SRY2627 (2%)

The others are undefined. There are 205 R1b-L21 (38%), which is the most common kind of R1b in Britain. It is found all along the Atlantic coast from Iberia to Norway, as well as in Germany. It could be just as well Celtic or Germanic. In the Netherlands and Scandinavia, L21 is found is approximately the same proportions as U106. So it is fair to assess that 12.5% of Scottish L21 is Germanic and 25.5% is Celtic.

Within R1b, 12.5% is Germanic R1b-U106 and 12.5% is Germanic R1b-L21. One fourth of R1b is Germanic, and three fourth Celtic. As 72.5% of Scots are R1b, it means that about 18% of all Scottish haplogroups are Germanic R1b.

The total for Germanic lineages (I1, I2b, Q, R1a and Germanic R1b) is therefore about 40%.

Middle Eastern haplogroups (G2a, J2, E1b1b, T), which account for 4.5% of Scottish lineages, might have come to Britain during the Neolithic, or through continental Celts, Romans and Germanic tribes. It's probably a bit of everything, though nobody knows in which proportion. Let's say that 2.5% is of Germanic origin to keep the proportions with average haplogroup frequencies in the Netherlands and Norway, the source countries of the Anglo-Saxons and Vikings in Scotland.

This gives us 42.5% of lineages of Germanic origin. The rest (57.5%) can be considered Celtic.

My bold. So if one is an L21+ Scot, is there anyway to determine whether one's Y-DNA is Germanic or Celt?

[Knovas]

The best option is to participate in admixture projects. As somebody noted above, Eurogenes gives a good idea about the Celtic background taking the North Atlantic cluster as reference.

It's the best you can do, since we can't know who was the first ancestor reflected in the Y-DNA marker (going back thousands of years ago).

[DavidCoutts]

Eurogenes -forgive my ignorance, but is that a company like 23andMe or FTDNA?

[Knovas]

No, it's a genetic project managed by Davidski (Polish). He uses the raw data provided, for example, at 23andme, and interprets your allele frequencies into different clusters. The same as Dodecad.

Here is the blog where the results appear (different analyses): http://bga101.blogspot.com/