Breakdown of Y-DNA distribution in Belgium by province

About the creation of the maps: what does the percentages represent? The current distribution of the DNA-donators or the distribution of the oldest known forefathers?

Jehan
 
Any ideas as to why Kempen would have such a relatively 'high' [3.6%] share of I2a?

I am curious about type of I2a found..

both Kempen in Belgium and Noord Brabant in the Netherlands have somewhat elevated I2a

I2* is in continental west Europe found on line from Italy to Denmark... and in continental east Europe from Black sea to Baltic along Dnieper...

I2a2 spread seems to follow Danube, Dniester and lower part of Dnieper...

my guess is that I2a in central and west Europe didnot spread only along Danube but also from Bodensee (Constance lake) along Rhine...

elevated J2 might be similar story... though J2 are in general sea people, while I2a are river people, so J2 might have even arrived with ships around coasts of Europe...


Europa_Ludwigskanal_Rhein_Main_Donau.png

dniester.jpg

dnieper.jpg

Haplogroup_I2a.gif


I2a = east/west linear pottery?

800px-European_Middle_Neolithic.gif
 
Last edited:
No, the Franks were almost certainly an admixture of R-U106, R1a, I1, and I2b.

R-U152 is either Gaulish Celt (descended from La Tène migrants who settled in the Meuse-Mosel region in the 5th century BCE) or "Roman" (Italic). Unfortunately there is no way to distinguish Gaulish from Roman at present, which is a shame as I always wanted to know in what proportion the Romans settled in different parts of their empire.

Yes so I think the Frankish elite was G2a3b1 according the Capetian bloodline and the 1/3 of G2a3b* found in a Merovingian high rank burial in Ergolding (Bavaria)
 
There a few more results for Wallonia the total number is now 69. I have added 5 more results from people I know who didn't take part to the project, so the total is 74.

Wallonia (n=74)

I1 : 11.2%

I2* : 1.4%

I2a : 0%

I2b : 6.8%

R1b : 59.5%
- R1b-U106/S21 : 21.6%
- R1b-P312/S116 (including L21) : 17.6%
- R1b-U152/S28 : 17.6%

R1a : 4%

G2a : 4%

E1b1b : 9.5% (including 6.8% of E-V13)

T : 2.7%

J1 : 0%

J2 : 1.4%

L: 0%

Q : 0%

Let's note the complete absence of haplogroup I and R1a from the provinces of Liège and Luxembourg at the moment. This is ironic as these are the two provinces that border Germany and that have the most Germanic place names (see Belgian toponymy). The province of Namur has a remarkable 28.5% of hg I (19% of I1 and 9.5% of I2b), 9.5% of R1a and 14% of R1b-U106, which makes it the most Germanic province of Wallonia (61.5% of lineages). This is also ironic since it is the province with the greatest proportion of Latin and Celtic place names.

There is only one Walloon belonging to haplogroup J (J2a) so far, which is also surprising considering the relatively high incidence of E1b1b (n=7) and T (n=2).

Also interesting is the absence of R1b-U152 from the province of Hainaut and Luxembourg. The latter is surely due to the tiny sample size for that province (only 5 participants). R1b-U152 is otherwise found at 17.4% of lineages in Walloon Brabant, 19% in the province of Namur and a staggering 40% in the province of Liège (where the total incidence of R1b reaches 80%).

R1b-P312 ranges from only 4.8% in Namur province to 18.8% in Hainaut, 20% in Luxembourg, 23.5% in Walloon Brabant, and 26.6% in Liège.

Notwithstanding the 40% in the province of Luxembourg that is undersampled, R1b-U106 peaks in Hainaut with 31.2%, followed by Walloon Brabant with 23.5%, 14% in Namur, 13.3% in Liège. This makes sense as the Hainaut is a geographic and historical continuity of Flanders, and Walloon Brabant as its name indicate is just a Francophone expansion of an otherwise Dutch-speaking region.

Wallonia appears therefore to be far more heterogeneous than Flanders. The gradient seems to be mostly from the hilly east to the flat west, with a hotspot of haplogroup I1, I2b and R1a in the Meuse valley in the middle. Only haplogroup E1b1b is found in similar proportions everywhere.
 
Last edited:
I am curious about type of I2a found..

both Kempen in Belgium and Noord Brabant in the Netherlands have somewhat elevated I2a

There is no I2a2 in Belgium. It's mostly the south-west European I2a.
 
There is no I2a2 in Belgium. It's mostly the south-west European I2a.
yes.
I think Bohemia/Bavaria is roughly where I2 splits.... I2a2 goes to east along Danube and north of Carpathians, I2a1 to west and south, I2b to north and west...
later some I2b also travelled to east with Germanic people...

interestingly, there is significant I2* in Wallonia....
do you perhaps know where else in west/central Europe we find significant I2*?
 
interestingly, there is significant I2* in Wallonia....
do you perhaps know where else in west/central Europe we find significant I2*?

1.6% isn't really significant. Flemish Brabant has 3% and East Flanders 2.5% of I2*. I don't know any other region that has more than 1% of I2*, but that may just be because older Y-DNA studies didn't differentiate I2 subclades clearly, and also because there are no good detailed regional studies of France and Switzerland at the moment. I expect that both countries could potentially hide pockets of I2* just like Belgium, especially in isolated areas where Paleolithic lineages could have survived in greater number.
 
I have updated the maps of U152 and U106 based on the new data from Wallonia.


Benelux-map-R1b-S21.jpg


Benelux-map-R1b-S28.jpg
 
There a few more results for Wallonia the total number is now 69. I have added 5 more results from people I know who didn't take part to the project, so the total is 74.

Wallonia (n=74)

I1 : 11.2%

I2* : 1.4%

I2a : 0%

I2b : 6.8%

R1b : 59.5%
- R1b-U106/S21 : 21.6%
- R1b-P312/S116 (including L21) : 17.6%
- R1b-U152/S28 : 17.6%

R1a : 4%

G2a : 4%

E1b1b : 9.5% (including 6.8% of E-V13)

T : 2.7%

J1 : 0%

J2 : 1.4%

L: 0%

Q : 0%

Let's note the complete absence of haplogroup I and R1a from the provinces of Liège and Luxembourg at the moment. This is ironic as these are the two provinces that border Germany and that have the most Germanic place names (see Belgian toponymy). The province of Namur has a remarkable 28.5% of hg I (19% of I1 and 9.5% of I2b), 9.5% of R1a and 14% of R1b-U106, which makes it the most Germanic province of Wallonia (61.5% of lineages). This is also ironic since it is the province with the greatest proportion of Latin and Celtic place names.

There is only one Walloon belonging to haplogroup J (J2a) so far, which is also surprising considering the relatively high incidence of E1b1b (n=7) and T (n=2).

Also interesting is the absence of R1b-U152 from the province of Hainaut and Luxembourg. The latter is surely due to the tiny sample size for that province (only 5 participants). R1b-U152 is otherwise found at 17.4% of lineages in Walloon Brabant, 19% in the province of Namur and a staggering 40% in the province of Liège (where the total incidence of R1b reaches 80%).

R1b-P312 ranges from only 4.8% in Namur province to 18.8% in Hainaut, 20% in Luxembourg, 23.5% in Walloon Brabant, and 26.6% in Liège.

Notwithstanding the 40% in the province of Luxembourg that is undersampled, R1b-U106 peaks in Hainaut with 31.2%, followed by Walloon Brabant with 23.5%, 14% in Namur, 13.3% in Liège. This makes sense as the Hainaut is a geographic and historical continuity of Flanders, and Walloon Brabant as its name indicate is just a Francophone expansion of an otherwise Dutch-speaking region.

Wallonia appears therefore to be far more heterogeneous than Flanders. The gradient seems to be mostly from the hilly east to the flat west, with a hotspot of haplogroup I1, I2b and R1a in the Meuse valley in the middle. Only haplogroup E1b1b is found in similar proportions everywhere.

Maciamo, thanks again for posting these numbers. Do you have the U152 breakdown of the 17.6% in Wallonia?
 
Maciamo, thanks again for posting these numbers. Do you have the U152 breakdown of the 17.6% in Wallonia?

In the text you quoted.
 
If I understood you good , you said there is not I2a2 -Din in Belgium , but because Belgium was on border there is some amount of J1 and J2 in it - mainly from Italy and Greece . Could that mean that there is no I2a2-Din in Balkans ( Greece ) in that time , and my theory of Sarmatian origin of I2a2-Din could be true ? Soory for imposing my own interest:ashamed2:
 
thank you for the percentages on Wallonia
i'm very hungry to see mor data because these result are evidently too contradictory, almost impossible to take in account in this stade
I 'd think Walloons are more on the U152 side than on the U106 side (more 'Gauls' than the North Picards Frenchmen) it's my bet
waiting to read some new data
 
Maciamo, thanks again for posting these numbers. Do you have the U152 breakdown of the 17.6% in Wallonia?

the topic got cold enough!!! but I find the sample is very very too scarce, even more when we consider the provinces of Wallonia (typical is the opposition between Ardennes and the Liege region: unreliable - wait and see... even the Hainaut, close to France Nord (pseudo Flander) has to high level of RU106
 
It's been over 6 years since the Brabant Y-DNA Project was completed. I have tried to gather additional Y-DNA samples from Wallonia as the sample size was tiny (n=55) compared to Flanders (about 1000). I have recalculated the frequencies based on the new sample size.


Wallonia (n=137)

I1 : 10.9%

I2a1/I2c : 1.4%

I2a2a : 3.6%

R1b : 60.5%
- R1b-U106/S21 : 17.5%
- R1b-P312 (xU152) : 13.9%
-- R1b-U152/S28 : 13.1%
- other R1b (V88, M269, L23, L11, L51): 3.6%
- undetermined R1b subclade : 12.4%

R1a : 7.3%

G2a : 5.1%

E1b1b : 5.8%

T : 3.6%

J1 : 0%

J2 : 1.4%

L: 0%

Q : 0%


No change for I1, R1b or T, but E1b1b, I2a2a and J2 decreased, while G2a and R1a increased. The percentage of R1a has doubled and now vastly exceeds that of Flanders.

If the undetermined R1b subclades are ignored, the percentage of R1b-U106 is Wallonia is actually 22%, while R1b-U152 is 16.5%. With the previous data those percentages were respectively 18% and 16.5%, so U106 has increased. I was able to get a more accurate provincial breakdown for U106 and U152 in Walloon provinces.

Benelux-map-R1b-S21.jpg


Benelux-map-R1b-S28.jpg



The western province of Hainaut and the centre-north Walloon Brabant have an unusually high percentage of I1: 16% and 21% - like the Netherlands and Schleswig-Holstein respectively. But as the sample sizes are still small, it may not be correct.

If I add up the typically Germanic haplogroups (I1, I2a2a-L801, R1a and R1b-U106), I get the following percentages:

Namur : 50%
Hainaut : 45%
Walloon Brabant : 35%
Liège : 32%
Luxembourg : 25%

Ironically, Liège scores pretty low despite being the only part of Belgium where German is spoke. It was also the seat of the Carolingian dynasty and it remained part of Germany until the French revolution. The historical Duchy of Luxembourg was split in two in 1839 when the country of Luxembourg was created. The country was mainly German speaking, while what became the Belgian province of Luxembourg was the predominantly French speaking part.

It's odd that Namur should have the highest percentage of Germanic haplogroups as it was the most Romanised part of Belgium and still has the highest percentage of Roman or Celtic toponyms in Belgium.
 
Last edited:
When they did these tests in the Brabant or previously in Normandy how deep was the y testing? I notice results are always listed as I1 or R1b-U106. Is that the extent of the y testing or just the way the results are being presented. I'm curious about the deeper subclades for these areas.
 
When they did these tests in the Brabant or previously in Normandy how deep was the y testing? I notice results are always listed as I1 or R1b-U106. Is that the extent of the y testing or just the way the results are being presented. I'm curious about the deeper subclades for these areas.

They only tested top levels haplogroups, except for E1b1b (V13, M34 and M81), J2 and and R1b (M269, P312, U152, U106). Keep in mind that the project was mostly done in 2010.

In 2013, the re-tested R1b participants for deeper subclades.
 
Last edited:
Here are the frequencies for the Grand Duchy of Luxembourg (n=38)

I1 : 2.5%

I2a1/I2c : 2.5%

I2a2a : 5%

R1b : 60.5%
- R1b-U106/S21 : 23.5%
- R1b-P312 (xU152) : 8%
-- R1b-U152/S28 : 10.5%
- other R1b (V88, M269, L23, L11, L51): 2.5%
- undetermined R1b subclade : 15.5%

R1a : 2.5%

G2a : 10.5%

E1b1b : 5%

T : 0%

J1 : 2.5%

J2 : 8%

L: 0%

Q : 0%


Although the percentage of R1b is identical to Wallonia and Flanders, the other haplogroups are completely different. Luxembourg is more shifted toward the Eastern Mediterranean, which is expected considering the high level of Romanisation (in the 3rd century, nearby Trier was the capital of the Gallic Empire then briefly also of the Western Roman Empire) and the local phenotypes (few light hair, shorter statures, bigger and more aquiline noses). The percentages of G2a and J2a are particularly elevated, but there is also a J1 sample, while so far there had only been one for all Belgium with 30x the sample size of Luxembourg. The three J2 samples all belong to J2a1-M67.

What's more, the only R1a sample from Luxembourg is a Middle Eastern Z93 (Z2123), not a Germanic L664 or Z284 like in Wallonia.

What is also surprising is the low frequency of I1 compared to Wallonia or Germany. Mind you, it's even lower than in northeast Italy, Catalonia or Galicia! Rather odd for a German-speaking country. That apparent contradiction may be explained by the unusually high level of U106 in Luxembourg.

If we exclude the R1b samples not tested for deep clades, 32% of the population belongs to R1b-U106, a higher proportion than anywhere in Belgium, Germany or Scandinavia, and close to the high levels observed in the Netherlands and England.

The Alpine Celtic and Roman R1b-U152 is at 14%, similar to levels observed in Belgium.
 
Last edited:
Here are the frequencies for the Grand Duchy of Luxembourg (n=38)

I1 : 2.5%

I2a1/I2c : 2.5%

I2a2a : 5%

R1b : 60.5%
- R1b-U106/S21 : 23.5%
- R1b-P312 (xU152) : 8%
-- R1b-U152/S28 : 10.5%
- other R1b (V88, M269, L23, L11, L51): 2.5%
- undetermined R1b subclade : 15.5%

R1a : 2.5%

G2a : 10.5%

E1b1b : 5%

T : 0%

J1 : 2.5%

J2 : 8%

L: 0%

Q : 0%


Although the percentage of R1b is identical to Wallonia and Flanders, the other haplogroups are completely different. Luxembourg is more shifted toward the Eastern Mediterranean, which is expected considering the high level of Romanisation (in the 3rd century, nearby Trier was the capital of the Gallic Empire then briefly also of the Western Roman Empire) and the local phenotypes (few light hair, shorter statures, bigger and more aquiline noses). The percentages of G2a and J2a are particularly elevated, but there is also a J1 sample, while so far there had only been one for all Belgium with 30x the sample size of Luxembourg. The three J2 samples all belong to J2a1-M67.

What's more, the only R1a sample from Luxembourg is a Middle Eastern Z93 (Z2123), not a Germanic L664 or Z284 like in Wallonia.

What is also surprising is the low frequency of I1 compared to Wallonia or Germany. Mind you, it's even lower than in northeast Italy, Catalonia or Galicia! Rather odd for a German-speaking country. That apparent contradiction may be explained by the unusually high level of U106 in Luxembourg.

If we exclude the R1b samples not tested for deep clades, 32% of the population belongs to R1b-U106, a higher proportion than anywhere in Belgium, Germany or Scandinavia, and close to the high levels observed in the Netherlands and England.

The Alpine Celtic and Roman R1b-U152 is at 14%, similar to levels observed in Belgium.

Hello Maciamo,

How can R-U106 be that high in Luxembourg? (higher than Belgium) Did the Franks settled in Luxembourg in great number ?
Do you think that maybe we need to wait for more testers before drawing conclusion ? (like the undersampled 40% of R-U106 province of Luxembourg in Wallonia).

Thank you for your work.
 
Here are the frequencies for the Grand Duchy of Luxembourg (n=38)

I1 : 2.5%

I2a1/I2c : 2.5%

I2a2a : 5%

R1b : 60.5%
- R1b-U106/S21 : 23.5%
- R1b-P312 (xU152) : 8%
-- R1b-U152/S28 : 10.5%
- other R1b (V88, M269, L23, L11, L51): 2.5%
- undetermined R1b subclade : 15.5%

R1a : 2.5%

G2a : 10.5%

E1b1b : 5%

T : 0%

J1 : 2.5%

J2 : 8%

L: 0%

Q : 0%


Although the percentage of R1b is identical to Wallonia and Flanders, the other haplogroups are completely different. Luxembourg is more shifted toward the Eastern Mediterranean, which is expected considering the high level of Romanisation (in the 3rd century, nearby Trier was the capital of the Gallic Empire then briefly also of the Western Roman Empire) and the local phenotypes (few light hair, shorter statures, bigger and more aquiline noses). The percentages of G2a and J2a are particularly elevated, but there is also a J1 sample, while so far there had only been one for all Belgium with 30x the sample size of Luxembourg. The three J2 samples all belong to J2a1-M67.

What's more, the only R1a sample from Luxembourg is a Middle Eastern Z93 (Z2123), not a Germanic L664 or Z284 like in Wallonia.

What is also surprising is the low frequency of I1 compared to Wallonia or Germany. Mind you, it's even lower than in northeast Italy, Catalonia or Galicia! Rather odd for a German-speaking country. That apparent contradiction may be explained by the unusually high level of U106 in Luxembourg.

If we exclude the R1b samples not tested for deep clades, 32% of the population belongs to R1b-U106, a higher proportion than anywhere in Belgium, Germany or Scandinavia, and close to the high levels observed in the Netherlands and England.

The Alpine Celtic and Roman R1b-U152 is at 14%, similar to levels observed in Belgium.

Hello Maciamo,

How can U106 be that high in Luxembourg (near the level of Netherlands)? Did the Franks settled in great number in Luxembourg and replaced the celtic tribe ? or just ruled the already present population ?

Maybe we need to wait for more samples before drawing conclusion (like the undersampled 40% of U106 in the province of Luxembourg in Wallonia).

By the way, thank you for your work.
 

This thread has been viewed 51392 times.

Back
Top