New I1 map

[Maciamo]

I have revised my map of haplogroup I1, adding a new category of colour for regions with 1 to 5% of I1, so as to distinguish them clearly from those with under 1% of I1. I have also added the hotspots around Slovenia, Macedonia, Central Ukraine, and northern Sicily.

I was able to give a more accurate representation for Sicily and Crete thanks to the more detailed studies of these islands by Di Gaetano et al. 2009 (Sicily) and Martinez et al. 2007 (Crete). The eastern part of both islands completely lacked I1.

Iberia is one of the most difficult part of the map. Whereas the Asturias, Galicia, Portugal and Extremadura have undeniably more than 1% of I1, and the Basque country, northern Castilla, Murcia and Andalusia less than 1%, I have conflicting data for Castilla-La-Mancha, Valencia and Aragon. Adams et al. give about 5% of I1 in Aragon (n=34), but Iberianroots gives 0% (n=74). In Valencia, Adams et al. reported 2.2% (n=73) of I1, but Iberianroots still has 0% (n=77). In Castilla La Mancha, Adams et al. has 0% (n=63), while Iberianroots has 2.6% (n=77).

Please let me know if you notice any possible errors (and please provide link to studies that contradict the data, not just hearsay or personal opinions). Keep in mind that the map remains quite approximate.

[Frank]

Well, Northern Portugal has 4-5% and Asturias about 4% :

http://iberianroots.com/statistics/i...peninsula.html

[Maciamo]

Well, Northern Portugal has 4-5% and Asturias about 4% :

http://iberianroots.com/statistics/i...peninsula.html

That's what the map says too. +1% means "more than 1%" (and as much as 5%).

[Frank]

That's what the map says too. +1% means "more than 1%" (and as much as 5%).

Some spot areas of Northern Portugal and Galicia have reported 6-12% of I1, i'll try to find later.

[sparkey]

Where are you getting your data for I1 frequencies in different regions of Britain? It seems unexpected, for example, that Kent is as low as Devon in its I1 levels, considering that Devonians cluster closely with the Cornish and Kent not so much in studies like Oppenheimer. I would expect Kent to be consistent with the rest of East England, especially because they are thought to be descendants of Jutes, but I don't have any studies on hand that would contradict you right now.

[LeBrok]

The 5% by Black Sea might be footprint of Goths, same as on I2b map.
Interesting is that I1 doesn't show on east coast of Greece like on I2b map, supposedly associated with Goths.
Another interesting thing is the trail and a spot in Macedonia.
Maybe Macedonians were mostly I1 proto Germanic tribe?
I wonder if we get more precise map of R1b, if R1b will correlates with Macedonian I1 hot spot?

[Maciamo]

Where are you getting your data for I1 frequencies in different regions of Britain? It seems unexpected, for example, that Kent is as low as Devon in its I1 levels, considering that Devonians cluster closely with the Cornish and Kent not so much in studies like Oppenheimer. I would expect Kent to be consistent with the rest of East England, especially because they are thought to be descendants of Jutes, but I don't have any studies on hand that would contradict you right now.

Sorry for the late reply. The lower frequency of I1 in Kent comes from Capelli et al. 2003

[LeBrok]

It is an interesting situation for Poland, specially for a region of Pomerania, north-west Poland. After WW2 most Germans from Pomerania were resettled to Germany, most new population in this region came from Poles from west Ukraine, west Belarus, and south Lithuania. Looking at the map one wouldn't be able to conclude this.
Is your map showing situation before WW2 or recent?

[Maciamo]

It is an interesting situation for Poland, specially for a region of Pomerania, north-west Poland. After WW2 most Germans from Pomerania were resettled to Germany, most new population in this region came from Poles from west Ukraine, west Belarus, and south Lithuania. Looking at the map one wouldn't be able to conclude this.
Is your map showing situation before WW2 or recent?

That's a good question. I didn't collect the data nor meet the people whose DNA was tested, so I don't know whether they were chosen because they lived there now or because their ancestors came from there. As there are many studies involved, it might be a bit of both. Then if you look at the situation 500 years ago (as mentioned on the maps on 23andMe), or 1000 years ago, it wouldn't be that different than after the Germans left Poland after WWII, as the German Duchy of Pomerania was only founded in the 12th century and Germans moved little by little to the region over the centuries.

Even if German speaking people left Poland and the Baltic, their genetic influence is still surely around. Hybrid couples (German + local) would have had children who could have remained German speakers or turned native. If they married locals, there is a good chance that they would have become locals themselves, especially after many centuries. That's probably why many Poles, Lithuanians and Latvians have a minority of German haplogroups among them, even if autosomally they are not really German.

[Emi]

My father's an I1*, and he is from the Black Sea Coast of Turkey. We always thought he was Pontus, aka "Rum" (Roman), even though my grandfather never spoke Greek or the Rum dialect, and he was a Muslim. Now I'm thinking the rare occurrence of I1 in Turkey might be related to Alexander the Great. Any thoughts?

[sparkey]

My father's an I1*, and he is from the Black Sea Coast of Turkey. We always thought he was Pontus, aka "Rum" (Roman), even though my grandfather never spoke Greek or the Rum dialect, and he was a Muslim. Now I'm thinking the rare occurrence of I1 in Turkey might be related to Alexander the Great. Any thoughts?

I1 is a relatively young haplogroup in terms of its MRCA (about 4200 years old) and probably didn't spread significantly throughout Europe until the Migration Period. Its highest concentrations are definitely among North Germanic peoples, followed by West Germanic peoples and descendants of East Germanic peoples. You'll probably see a pattern there, and it's not Macedonian. There is an interesting concentration of I1 in northern Greece (more than 5%) but my guess is that most of it got there around the same time as most of it got to Anatolia. There are multiple ways your particular line could have gotten to Anatolia, but the most common case is probably descent from the Goths.

[Taranis]

I1 is a relatively young haplogroup in terms of its MRCA (about 4200 years old) and probably didn't spread significantly throughout Europe until the Migration Period. Its highest concentrations are definitely among North Germanic peoples, followed by West Germanic peoples and descendants of East Germanic peoples. You'll probably see a pattern there, and it's not Macedonian. There is an interesting concentration of I1 in northern Greece (more than 5%) but my guess is that most of it got there around the same time as most of it got to Anatolia. There are multiple ways your particular line could have gotten to Anatolia, but the most common case is probably descent from the Goths.

Yes, Haplogroup I1 itself appears to be Chalcolithic in age, even though Haplogroup I as a whole is probably Mesolithic in age (we know that I2a was present in the Neolithic in the Atlantic region, thereby making Haplogoup I a probably candidate for a Mesolithic European haplogroup). From this perspective, it would seem most plausible that I1 is essentially a single surviving male lineage that coincidentially survived inside a population of otherwise Chalcolithic newcomers. The case for this discontinuity can also be made elsewhere.

The question is, can the spread of Haplogroup I be solely explained by the migrations period and by the spread of the Germanic peoples? Even though without a doubt the Haplogroup was spread by the Germanic peoples, I'm not quite convinced that the distribution pattern we see can be solely explained by this. In particular, there is a significant percentage amongst the Finnic peoples which in my opinion can not be explained solely through interaction with the Vikings (we are talking about some of the highest frequencies of the Haplogroup here, after all). Another issue is the surprisingly high concentrations in the Atlantic region, especially in Ireland, Wales and on the Iberian pennsula. While it would be possible to attribute this to the Vikings (in Ireland) and the Suebi (in Galicia), I would ask the question if it is possible that there a small Celtic component to Haplogroup I1 as well. Specifically, if Haplogroup I1 was already present in Central Europe by the start of the iron age, and subsequently spread with the Hallstatt and La-Tene Cultures, this would better explain some of the distribution patterns we see in the Atlantic region (as well as in some other areas, such as along the Danube), rather than if we solely assume a spread by the Germanic peoples during the Migration Period.

The reason I ask this is because the distribution in Iberia corresponds better with the maximum extend of Celtic influence on the Iberian penninsula (3rd century BC) than it does with Germanic influence (the Suebi are often cited as the source of I1 in Iberia). However, what about the Visigoths, why were they not such significant bearers of I1 if the Suebi were?

[sparkey]

Good post as usual, Taranis. Naturally, I'm not going to defend a 100% Migration Period spread of I1, I think that even though it is very young and its spread is quite recent in the grand scheme of things, it must have had some sort of an earlier spread as well. 4200 years ago predates the Nordic Bronze Age, which I suspect is where it was confined to at one point, but you obviously can't keep it all in one place. I don't think that it was a major haplogroup among proto-Celtic peoples, although it must have been a minor one after some point; similarly, it looks like it could have been a significant contributing population to modern Finnic peoples. Probably, whatever original I1 among them is representative of the contribution of the Nordic Bronze Age diaspora within them (I really have trouble imagining that the contribution could have come earlier than the Nordic Bronze Age--although I suppose that it is possible).

Still, the correlation between I1 and Germanic peoples remains one of the better correlations between Y-DNA haplogroups and language families. And it still seems to me that Anatolian and Black Sea I1 maps best to East Germanic peoples. There could be earlier I1 there as well, but if we're talking the most likely origin of a given I1 sample, that seems to me to be it.

[LeBrok]

The reason I ask this is because the distribution in Iberia corresponds better with the maximum extend of Celtic influence on the Iberian penninsula (3rd century BC) than it does with Germanic influence (the Suebi are often cited as the source of I1 in Iberia). However, what about the Visigoths, why were they not such significant bearers of I1 if the Suebi were?

I've asked myself the same question too. One of the options is that Suebi came directly and quickly from Germania to Portugal, and Goths were wondering around Europe for almost one thousand years. Possibly long enough to get their I1 substantially deluded. Also most Visigoths left for Africa, but Suebi stayed in Portugal for ever, completely giving their I1 to locals.

[Taranis]

Good post as usual, Taranis. Naturally, I'm not going to defend a 100% Migration Period spread of I1, I think that even though it is very young and its spread is quite recent in the grand scheme of things, it must have had some sort of an earlier spread as well. 4200 years ago predates the Nordic Bronze Age, which I suspect is where it was confined to at one point, but you obviously can't keep it all in one place. I don't think that it was a major haplogroup among proto-Celtic peoples, although it must have been a minor one after some point; similarly, it looks like it could have been a significant contributing population to modern Finnic peoples. Probably, whatever original I1 among them is representative of the contribution of the Nordic Bronze Age diaspora within them (I really have trouble imagining that the contribution could have come earlier than the Nordic Bronze Age--although I suppose that it is possible).

Thank you.

My point with the Atlantic region is that it would not have needed to ever have been any major component, nor that it would necessarily have arrived with the Proto-Celts in the Atlantic region, it would have been sufficient if I1 arrived in Central Europe by the Bronze Age and spread from there subsequently. I just noted that there may be a small Celtic component to I1, which would explain the approximately 1-5% of I1 we find in certain areas that saw considerable Celtic but negligible Germanic influence.

Still, the correlation between I1 and Germanic peoples remains one of the better correlations between Y-DNA haplogroups and language families. And it still seems to me that Anatolian and Black Sea I1 maps best to East Germanic peoples. There could be earlier I1 there as well, but if we're talking the most likely origin of a given I1 sample, that seems to me to be it.

Otherwise, I absolutely I agree that the main correlation is between I1 and the Germanic peoples, and that this explains the bulk of the patterns we see.

I've asked myself the same question too. One of the options is that Suebi came directly and quickly from Germania to Portugal, and Goths were wondering around Europe for almost one thousand years. Possibly long enough to get their I1 substantially deluded. Also most Visigoths left for Africa, but Suebi stayed in Portugal for ever, completely giving their I1 to locals.

This is also a possibility, especially considering the intermingling of the Goths already before the migration period. Just a small nitpick though, the Visigoths didn't leave for Africa, it was the Vandals who did that. The Visigoths were the ones who ended up controlling virtually the entire Iberian penninsula at the eve of the invasion by the Umayyad Caliphate.

[Emi]

I don't think I'm convinced yet about the extent of long-term Goth influence in the Black Sea area of Turkey. Just because a certain group of people has conducted campaigns in a region, does it necessarily indicate they settled in the area for long enough to pass on their DNAs, for thousands of more years? I suppose it depends on whether they wanted to "populate" the local population, just expand territory, or reach another destination. So, their admixture would really depend on their original intent to spread. Please bear in mind, I'm by no means well versed in this subject.

As far as the Black Sea goes, I do not consider the region as Anatolia, ethnically nor historically. The region has a distinctive climate, geography [little arable/grazing land], making it more suitable to seafarers. Any major port on the Black Sea, you'll find Genoan (Genoese?) forts for instance. Here's a map I found that shows where they spread and established commercial/military relationships. Looks like they did settle in Crimea.

http://www.turkcebilgi.com/cenevizli...i-genova#resim

[Cambrius (The Red)]

I have revised my map of haplogroup I1, adding a new category of colour for regions with 1 to 5% of I1, so as to distinguish them clearly from those with under 1% of I1. I have also added the hotspots around Slovenia, Macedonia,Central Ukraine, and northern Sicily.

I was able to give a more accurate representation for Sicily and Crete thanks to the more detailed studies of these islands by Di Gaetano et al. 2009 (Sicily) and Martinez et al. 2007 (Crete). The eastern part of both islands completely lacked I1.

Iberia is one of the most difficult part of the map. Whereas the Asturias, Galicia, Portugal and Extremadura have undeniably more than 1% of I1, and the Basque country, northern Castilla, Murcia and Andalusia less than 1%, I have conflicting data for Castilla-La-Mancha, Valencia and Aragon. Adams et al. give about 5% of I1 in Aragon (n=34), but Iberianroots gives 0% (n=74). In Valencia, Adams et al. reported 2.2% (n=73) of I1, but Iberianroots still has 0% (n=77). In Castilla La Mancha, Adams et al. has 0% (n=63), while Iberianroots has 2.6% (n=77).

Please let me know if you notice any possible errors (and please provide link to studies that contradict the data, not just hearsay or personal opinions). Keep in mind that the map remains quite approximate.

According to Beleza et al. (2005-2006) your I1 averges for Portugal appear substantially off.

Here are the Beleza et al. figures based on over 650 samples:

AVERAGE NATIONWIDE I1b = 6.1%

AVERAGE NATIONWIDE I1b2 = 1.5%

The regional figures are:

NORTH (Minho and Entre Douro) I1b = 7.5% and I1b2 = 0.9%.

NORTH (Tras-os-Montes) I1b = 9.4% and I1b2 = 4.7%.

CENTRAL-NORTH (Beira Litoral) I1b = 3.4% and I1b2 = 0.9%.

CENTRAL-NORTH (Beira Interior) I1b = 3.5% and I1b2 = 0.

CENTRAL (Extremadura) I1b = 9.3 and I1b2 = 4.7%

CENTRAL-SOUTH (Lisbon and Setubal) I1b = 1.6% and I1b2 = 0

SOUTH (Alentejo) I1b = 7.7% and I1b2 = 3.1%.

SOUTH (Algarve) I1b = 4.8% and I1b2 = 0%.

Based on this research - which I believe is the latest comprehensive Y-DNA halogroup study for Portugal - your Y-DNA tables may require some revision.

The highest frequencies of I1b and I1b2 combined were found in the Braga region (17.7%) and Braganca (16.0%).

Ref:

http://onlinelibrary.wiley.com/doi/1...05.00221.x/pdf

[Maciamo]

According to Beleza et al. (2005-2006) your I1 averges for Portugal appear substantially off.

Here are the Beleza et al. figures based on over 650 samples:

AVERAGE NATIONWIDE I1b = 6.1%

AVERAGE NATIONWIDE I1b2 = 1.5%

The regional figures are:

NORTH (Minho and Entre Douro) I1b = 7.5% and I1b2 = 0.9%.

NORTH (Tras-os-Montes) I1b = 9.4% and I1b2 = 4.7%.

CENTRAL-NORTH (Beira Litoral) I1b = 3.4% and I1b2 = 0.9%.

CENTRAL-NORTH (Beira Interior) I1b = 3.5% and I1b2 = 0.

CENTRAL (Extremadura) I1b = 9.3 and I1b2 = 4.7%

CENTRAL-SOUTH (Lisbon and Setubal) I1b = 1.6% and I1b2 = 0

SOUTH (Alentejo) I1b = 7.7% and I1b2 = 3.1%.

SOUTH (Algarve) I1b = 4.8% and I1b2 = 0%.

Based on this research - which I believe is the latest comprehensive Y-DNA halogroup study for Portugal - your Y-DNA tables may require some revision.

The highest frequencies of I1b and I1b2 combined were found in the Braga region (17.7%) and Braganca (16.0%).

Ref:

http://onlinelibrary.wiley.com/doi/1...05.00221.x/pdf

Don't waste your time, I don't use tiny samples for one province. I make averages for wider regions (e.g. North Portugal) using all data for all studies.

[Knovas]

Good post. Probably Galicia show very similar proportions, it's bad not to have information about it.

[Cambrius (The Red)]

I would say much of north-western, central and northern Spain may have similar frequencies. In fact, there is a I (Y-DNA) hotspot in Castilla.

[Wilhelm]

But isn't I1b the old nomenclature for I2 ?

[Cambrius (The Red)]

Don't know. Still, according to the Beleza et al. results, the I (Y-DNA) frequencies for Portugal on Eupedia are not correct.

I1b is associated with Slavic peoples, among others.

[Knovas]

Sardinians are really rare. Even in admixture analysis there's nothing clear on them, except for the fact they appear to hold the largest Southern European genetic in average (although the proportions are truly unknown).

Acording to Eupedia I2a1 originated in Sardinia. However, I think that the Pyrenees are the most likely place to explain its origin and expansion. The high presence in Sardinia is the result of an ancient bottleneck in my opinion, nothing else. There must be sources talking about this, since the information showed at 23andme clearly opts for the origin in the Pyrenees.

[sparkey]

The type of haplogroup I in Sardinia (I2a1a) would almost certainly have been the most common type of haplogroup I during the Neolithic, considering that it has the oldest TMRCA of any of the major haplogroup I branches, and that it has been found significantly in ancient DNA samples together with G2a. Its apparent correlation with G2a in some cases, combined with the high G2a in Sardinia (15%, the highest in Europe), make me think that it spread as early as the Neolithic into Sardinia alongside G2a with the expansion of farming into the region. A slightly later migration corresponding with the Nuragic civilization is also possible.

[flupke]

I have revised my map of haplogroup I1,

I

Dear Maciamo,
I would have expected Belgium and Northern France (Nord, pas de Calais: Saxons, Franks, Frisians + Danes) to be around 15% and Normandy (Danes+ Franks) + Picardy (Franks) to be around 10%.
Mertens even put 20% I1 in the flemish population
Journal of Genetic Genealogy, 3(2):19-25, 2007
Y-Haplogroup Frequencies in the Flemish Population
Gerhard Mertens