New map of haplogroup J2 (Y-DNA)

[Maciamo]

I was fed up to constantly see badly mistaken distribution maps of haplogroup J2 (the latest in date being from the Thangaraj et al. study), so I decided to make my own, perusing all the data for each region of Europe and the Middle East with scrupulous attention.

There were often very conflicting studies, so I had to compromise and use common sense. For example, one study found only 3% of J2 in Corsica while another claimed it to be as high as 21%.

 

[how yes no 2]

thanks Maciamo! this is very interesting map that tells lot of stories about history...

btw. can you stress for which areas is data solid and for which is interpolation...

e.g. Liburnia - is there data that shows lack of it there? I was expecting exactly the opposite - a hotspot... was there any sampling there?

 

[Wilhelm]

Catalonia has only about 3% of J2, not in the range of 5-10 %

 

[Maciamo]

thanks Maciamo! this is very interesting map that tells lot of stories about history...

btw. can you stress for which areas is data solid and for which is interpolation...

e.g. Liburnia - is there data that shows lack of it there? I was expecting exactly the opposite - a hotspot... was there any sampling there?

Don't worry, I checked it already. Percici et al., who made the most detailed study of Croatian Y-DNA and mtDNA to date, found between 2.3 and 3.5% of J2 in the Liburnian islands of Brac, Hvar and Korcula. Only Krk (Veglia in Italian), closer to Venice, had 11%, but not the Croatian mainland opposite (1.9%).

 

[Maciamo]

Catalonia has only about 3% of J2, not in the range of 5-10 %

Semino et al. (2004) found 6%, Adams et al. (2008) also 6%. Flores et al. (2004) found 0%, but their sample for Catalonia was a useless 16 individuals.

 

[how yes no 2]

Don't worry, I checked it already. Percici et al., who made the most detailed study of Croatian Y-DNA and mtDNA to date, found between 2.3 and 3.5% of J2 in the Liburnian islands of Brac, Hvar and Korcula. Only Krk, closer to Venice, had 11%, but not the Croatian mainland opposite (1.9%).

be careful there: Croatian mainland is a mix that include area as far as Zagorje (north from Zagreb), Osijek (east Slavonia close to border with Serbia), and Dubrovnik (very far south close to Montenegro)..those areas are genetically extremely diverse... R1a being strong in north and scarce in south, R1b strong in west and scarce in east, I2a2 scarce in northwest of Croatia and extreme in south Dalmatia....

J2 did spread along sea coasts mostly... my guess that in area of Liburnia it is around 11% as Island of Krk is close to mainland area of Liburnia...

Brac, Hvar and Korcula are on the map of Liburia south most islands, very far from Liburnia mainland.... I do not think they can be used to extrapolate values for whole Liburnia because those 3 islands are quite separated from the rest of Liburnia and for instance have I2a2 around 55% while island of Krk on north of Liburnia has 9.5% I2a2 (and 17.8% of other I haplogroups, likely I1)

besides, area that looks as a hole is exactly the settlement of Croats according to http://www.euratlas.net/history/europe/700/entity_1523.html
which further complicates interpretation of the supposed hole in J2 spread... because that area was not sampled at all... all mainland samples are taken further away from it...

I have even heard discussion that J2 might have been main marker of Liburnians before Croats settled the area

also if you look at variance for J2, it is pretty high along sea coast in area of Liburnia, and in general has increased values along sea coasts which indicates that J2 initially spread along sea coasts...

 

[Maciamo]

be careful there: Croatian mainland is a mix that include area as far as Zagorje (north from Zagreb), Osijek (east Slavonia close to border with Serbia), and Dubrovnik (very far south close to Montenegro)
J2 did spread along sea coasts mostly... my guess that in area of Liburnia it is around 11% as Island of Krk is close to mainland area of Liburnia...
Brac, Hvar and Korcula are on the map of Liburia south most islands, very far from Liburnia mainland.... I do not think they can be used to extrapolate values for whole Liburnia because those 3 islands are quite separated from the rest of Liburnia and for instance has I2a2 around 55% while island of Krk on north of Liburnia has 9.5% I2a2
besides, area that looks as a hole is exactly the settlement of Croats according to http://www.euratlas.net/history/europe/700/entity_1523.html
which further complicates interpretation of the supposed hole in J2 spread...
I have even heard discussion that J2 might have been main marker of Liburnians before Croats settled the area
also if you look at variance for J2, it is pretty high along sea coast in area of Liburnia

You are just extrapolating. I need proof, and all the evidence I have show that the average for Slovenia and Croatia is around 2.5% of J2, and Bosnia-Herzegovina around 6%. You can't just guess that some haplogroup must have spread because this or that area was colonised by this or that people.

The last map you posted (for J2b2) show level under 5% for Liburnia. Variance isn't the same as frequency. What's your point ? Do you even know what you are posting ?

 

[how yes no 2]

You are just extrapolating. I need proof, and all the evidence I have show that the average for Slovenia and Croatia is around 2.5% of J2, and Bosnia-Herzegovina around 6%. You can't just guess that some haplogroup must have spread because this or that area was colonised by this or that people.

The last map you posted (for J2b2) show level under 5% for Liburnia. Variance isn't the same as frequency. What's your point ? Do you even know what you are posting ?

I am only saying that there is not enough data for exact area that you have drawn as a hole in J2 spread, as there was no sampling in that area... look at Figure 2 of work Pericic to see exact locations of sampling... none of them falls in the area that you see as a hole in J2... and whole Croatia is genetically extremely diverse area (compare Figure 4 with Figure 2), which makes any guessing relatively hard... in fact, I would put question mark on complete mainland testing of Croatia presented in that work as it is in total 108 people on wide area that shows extremely diverse genetic landscape...

there is absolutely no reason why this area would follow the spread of south most islands Krk, Hvar and Brac, and not the spread of north most one - Krk...

high variance suggest that early inhabitants were strong in J2 marker... and old inhabitants of the area are Liburnians... so I would expect higher frequencies of J2 along the coastline and on islands.. and in mainland part of what you draw as hole were also Iapodes.. again because no samples were taken from area we have no clue what were genetic markers of Iapodes

closest sampling point to the area is Delnice...
but again Delnice are not representative of old settlement...
according to http://hr.wikipedia.org/wiki/Delnice
due to lot of forest and no roads area was settled only in late 13th century, first by cakavian speaking people..than area of Delnice was depopulated due to Turkish conquests, and in 17th century Delnice was populated by Kaikavian speaking people from areas that are now in Slovenia... this makes any data from Delnice likely outlier in wider surroundings south of it

btw. I just noticed, chakavian dialect of Croatian (in fact difference between chakavian, kaikavian and stokavian dialects of Croatia are more like the ones you see in different languages) correlates fairly well with Liburnians and Histri and not with Dalmatae

 

[how yes no 2]

Maciamo, it is clear to me that when trying to make map one often needs to interpolate values, as not all areas are really sampled... however, clues that one takes into account when interpolating might be sometimes misleading...

so, I was wondering if it is possible that areas that are not really sampled are in your maps covered with some pattern e.g. stripes , or xxx

in attempts to interpret the maps, such markings would help in avoiding to take assumptions for granted

btw.

You are just extrapolating. I need proof, and all the evidence I have show that the average for Slovenia and Croatia is around 2.5% of J2, and Bosnia-Herzegovina around 6%. You can't just guess that some haplogroup must have spread because this or that area was colonised by this or that people.

if average for Slovenia is around 2.5% why is on your map the area of Slovenia mostly in zone 10-15%?
also for Bosnia it is not flat 6% as in Bosnian Croats (west part of Herzegovina) it is around 1%, while in Serbs and Muslims it is larger than 6% (around 13% for Muslims, around 9% for Serbs, note that Serb samples come mostly from north areas of Bosnia, so Serb populated east Herzegovina might still be low in J2 )
http://onlinelibrary.wiley.com/doi/10.1111/j.1529-8817.2005.00190.x/pdf

also
_____________Serbia___Montenegro

J1______________0.6% ___0.5%
J2a ____________3.3%____4.7%
J2b ____________1.7% ___4.5%

http://onlinelibrary.wiley.com/doi/10.1002/ajpa.21235/suppinfo

thus,

J2a + J2b________5% _____9.2%

while on your map large part of Serbia is in 10-15% area, and large part of Montenegro is in 20-30% area

 

[Maciamo]

also for Bosnia it is not flat 6% as in Bosnian Croats (west part of Herzegovina) it is around 1%, while in Serbs and Muslims it is larger than 6% (around 13% for Muslims, around 9% for Serbs, note that Serb samples come mostly from north areas of Bosnia, so Serb populated east Herzegovina might still be low in J2 )

Regarding Bosnia, I am not going to draw dots and enclaves by ethnic and religious groups. That's impossible. That's why I have made an average. But it's true that the Bosnian Croats have less J2, so I have extended a bit the 1-5% zone towards western Bosnia.

I am aware that the map isn't perfect, but it cannot be without having detailed studies of every nook and cranny of every country.

 

[how yes no 2]

I am aware that the map isn't perfect, but it cannot be without having detailed studies of every nook and cranny of every country.

ok, I think it is significantly improved with changes for Serbia, Montenegro, and extending the hole somewhat towards Slovenia...

I am doubtful about west Bosnia as that is part inhabited with Serbs and Muslims while Croats are more south of it..

But anyway, in general I am satisfied with corrections...

I do expect that future sampling will show more J2 in narrow area along north part of Croatian Adriatic coast and on north-central islands...

my general impression is that J2 spread along sea coasts, and that people with strong (not necessarily dominant) J2 were thus very sea related people (Phoenicians, Illyrians, Crete, some of Greek tribes, some Italic tribes, Etruscans, Veneti, and I think Liburnians as well..).... and variance of J2 seems to indicate that as well...

 

[Wilhelm]

Semino et al. (2004) found 6%, Adams et al. (2008) also 6%. Flores et al. (2004) found 0%, but their sample for Catalonia was a useless 16 individuals.

The semino and Flores samples are also ridiculous. How about 193 catalans :

http://iberianroots.com/statistics/iberian_peninsula.html

 

[R.Rocca]

Maciamo, excellent map. May I point out that the higher J2 percentages in Italy seem to mirror the areas of the Italic tribes, including the Romans, Brutti, Siculi, and the north-eastern Veneti. Also, the Romans claimed to be descendants of Troy - another high % area on your map.

 

[Semitic Duwa]

When is the J1 map coming?:grin:

 

[Maciamo]

When is the J1 map coming?

A European map of J1 won't be possible due to its low frequency in every country besides Italy and Greece. What's more, typical country-wide studies show less than 1% of J1, but detailed studies at the provincial level often show minor "hotspots" in some well defined areas. For example, in the Brabant DNA Project for Belgium, most provinces had 0% of J1, but the Flemish Brabant province (Brussels and surrounding) had 3%.

 

[Semitic Duwa]

A European map of J1 won't be possible due to its low frequency in every country besides Italy and Greece. What's more, typical country-wide studies show less than 1% of J1, but detailed studies at the provincial level often show minor "hotspots" in some well defined areas. For example, in the Brabant DNA Project for Belgium, most provinces had 0% of J1, but the Flemish Brabant province (Brussels and surrounding) had 3%.

That is to say, M35 isn't that populous in Europe either (except in the Balkans with the V13 clade, but nvm).

 

[Maciamo]

That is to say, M35 isn't that populous in Europe either (except in the Balkans with the V13 clade, but nvm).

E-M35 reaches level of 5 to 10% in most of continental Europe, and between 10 and 30% in West Iberia, Italy, Greece and the Balkans. There is no comparison with J1. The frequency of J1 in Europe is more comparable to E1b1b1c (E-M123) or T. In fact it is entirely possible that these three haplogroups are related (all probably migrated from the southern Levant to Europe, in all likelihood in a larger group dominated by E-M78 and J2 individuals).

 

[MarTyro]

I would like to know if the last J2-map update was in Dec. 2010?

 

[Goga]

I was fed up to constantly see badly mistaken distribution maps of haplogroup J2 (the latest in date being from the Thangaraj et al. study), so I decided to make my own, perusing all the data for each region of Europe and the Middle East with scrupulous attention.

Hi Maciamo,

According to you and your legend is J2a Greco-Anatolian, Mesopotamian and Caucasian.

But you forgot the most important ethnic group that also belongs to this haplogroup! Iranic folks! You can find J2a in all Iranic peoples, from Kurdistan to Tajikistan. So according to me J2a is also an Iranic marker!

J2 is actually the most common haplogroup among all Iranic folks all over the world! Doesn't matter where they live, in the Mesopotamia or Central Asia..

 

[LeBrok]

I'd say, J2 was highly agriculturalist. Their epicenter was and is in Fertile Crescent, spreading nicely and continuously in southern and western Europe, with diminishing in Carpathian Mountains and other mountainous areas.

Their crops, early wheats where not as advanced as today's varieties, so it is logical that J2 spread was limited in northern Europe where climate was vastly different than the area of their inception. Plus North was already populated by successful herders of R1a haplogroup, the steep competition.
J2 is the richest in Greece and Italy. This might mean that well off agricultural societies were behind the engine of first strong city-states and consequently the first European empires of Greece/Macedonia, and Romans. Eventually they also became sea going people, but it was more around Bronze Age and later.
I have a feeling that J2 was most numerous of all haplogroups in southern Europe in mid Neolithic till the arrival of IE.
J2 also matches the borders of Roman Empire. Obviously RE helped J2 flow within it's borders, but I think it didn't do much difference. The truth might be that RE conqured these areas with rich populous farming communities (already with a lot of J2) and wasn't much interesting in more spares, poorer and wiled north.

Summarizing:
1. First farmers in Europe
2. No other HG matches spread of early farming that well.
3. Rich agricultural engine behind Greece and Rome
4. Most populous among HG in mid Neolithic till IE R1b arrival, and declining since Bronze Age (percentile wise)
5. J2 in Northern Europe might be more Jewish than early farmers.