Croat K13 regional (and national) averages

[Jana]

it was based on Kajkavian speakers clustering away from the rest. But now Istria turned out to be quite north plotting too, maybe you should do a new weighed average with diferrent weights for each region.

PM all database (samples i sent you + what you and commonsense or mingle collected) pls. My PC is a mess. I need to know exact sample number to work with aevrages.

[vbnetkhio]

PM all database (samples i sent you + what you and commonsense or mingle collected) pls. My PC is a mess. I need to know exact sample number to work with aevrages.

i sent it.
i recommend you average out the results from the same town uploaded by the same email, and treat them as one result. like the kits from Vrdi for example.

[MoroLP]

What happened to all the regional averages? East is Slavonia? North is Zagorje + Međimurje? Littoral is confusing, is it well Littoral? What happened to Dalmatia then or it includes both Littoral (+Istria?) and the Dalmatian region? The northern and southern Adriatic coast shouldn't be put together. Names are a bit confusing, uncommon for Croatian regions. What is the sample size? New averages:

Croat_Bosnia&Herzegovina,25.53,32.41,15.86,8.50,13.91,1.38,0.59 ,0.25,0.56,0.46,0.35,0.04,0.11
Croat_East,27.42,34.37,13.89,8.65,11.79,1.56,0.03, 0.14,1.16,0.62,0.35,0.02,0.01
Croat_Littoral,26.81,32.09,15.78,8.11,13.54,1.40,0 .33,0.24,0.56,0.49,0.39,0.13,0.11
Croat_north,31.28,33.43,14.32,7.22,9.91,1.18,0.61, 0.31,0.54,0.59,0.35,0.13,0.13
Croat,28.38,33.18,14.95,7.74,11.78,1.42,0.50,0.26, 0.56,0.53,0.45,0.10,0.13

[Jana]

What happened to all the regional averages? East is Slavonia? North is Zagorje + Međimurje? Littoral is confusing, is it well Littoral? What happened to Dalmatia then or it includes both Littoral (+Istria?) and the Dalmatian region? The northern and southern Adriatic coast shouldn't be put together. Names are a bit confusing, uncommon for Croatian regions. What is the sample size? New averages:

Croat_Bosnia&Herzegovina,25.53,32.41,15.86,8.50,13.91,1.38,0.59 ,0.25,0.56,0.46,0.35,0.04,0.11
Croat_East,27.42,34.37,13.89,8.65,11.79,1.56,0.03, 0.14,1.16,0.62,0.35,0.02,0.01
Croat_Littoral,26.81,32.09,15.78,8.11,13.54,1.40,0 .33,0.24,0.56,0.49,0.39,0.13,0.11
Croat_north,31.28,33.43,14.32,7.22,9.91,1.18,0.61, 0.31,0.54,0.59,0.35,0.13,0.13
Croat,28.38,33.18,14.95,7.74,11.78,1.42,0.50,0.26, 0.56,0.53,0.45,0.10,0.13

Hey. We decided to decrease number of averages since lot of them were very similar to each other, and that way create more reliable averages with much bigger sample size.
Divison was proposed by Vbnethkio and I agreed, although I had slightly different idea (my idea was Panonnian, Dinaric and Adriatic)

Litorral (Istria+Kvarner+Lika+Dalmatia) - sample size 48
North (Zagorje+Central+ Gorski kotar) - sample size 41
Bosnia-Herzegovina (Bosnia_proper+West_Bosnia+Herzegovina) - sample size 52
East (Slavonia) - sample size 7

General average is weighted according to percentage of regional populations in total population historically.

Any luck getting 70 Slavonian samples?

[Lucas]

I posted it in Vahaduo thread but should be rather here.

She has all 4 grandparents kajkavian Croats
So, her father is from Zlatar, Krapinsko/zagorska županija
Her mother is half from Cestica, on border with Slovenia and half from mursko središče on border with Slovenia and Hungary


Population
North_Atlantic 32.54 Pct
Baltic 33.63 Pct
West_Med 14.34 Pct
West_Asian 6.01 Pct
East_Med 9.51 Pct
Red_Sea -
South_Asian -
East_Asian 0.55 Pct
Siberian -
Amerindian 1.54 Pct
Oceanian 1.3 Pct
Northeast_African 0.58 Pct
Sub-Saharan -


# Population (source) Distance
1 Hungarian 3.55
2 East_German 5.78
3 Croatian 5.83
4 Austrian 6.17
5 Moldavian 8.24
6 South_Polish 9.03
7 Ukrainian_Lviv 9.96
8 Serbian 10.09
9 Ukrainian 10.74
10 Polish 12.47
11 West_German 13.7
12 Romanian 14.28
13 North_Swedish 14.88
14 Southwest_Russian 14.97
15 South_Dutch 15
16 North_German 15.23
17 Southwest_Finnish 15.39
18 Ukrainian_Belgorod 15.42
19 Russian_Smolensk 15.64
20 Estonian_Polish 16.09

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 77.8% Croatian + 22.2% Southeast_English @ 2.48
2 60.6% Ukrainian_Belgorod + 39.4% Spanish_Cataluna @ 2.69
3 62.5% Ukrainian_Belgorod + 37.5% Spanish_Valencia @ 2.7
4 61.3% Southwest_Russian + 38.7% Spanish_Cataluna @ 2.75
5 78.9% Croatian + 21.1% Orcadian @ 2.76
6 80.1% Croatian + 19.9% Southwest_English @ 2.81
7 77.8% Croatian + 22.2% North_Dutch @ 2.85
8 80.7% Croatian + 19.3% West_Scottish @ 2.89
9 74.9% South_Polish + 25.1% Spanish_Murcia @ 2.9
10 77.6% Croatian + 22.4% Danish @ 2.91
11 73.1% South_Polish + 26.9% Portuguese @ 2.91
12 74.8% South_Polish + 25.2% Spanish_Valencia @ 2.92
13 72.7% Croatian + 27.3% West_German @ 2.93
14 80.1% Croatian + 19.9% Irish @ 2.96
15 63.3% Southwest_Russian + 36.7% Spanish_Valencia @ 2.97
16 62.8% Ukrainian_Belgorod + 37.2% Spanish_Murcia @ 2.97
17 50.3% East_German + 49.7% Croatian @ 2.98
18 51.9% Croatian + 48.1% Austrian @ 2.99
19 76.1% South_Polish + 23.9% Spanish_Andalucia @ 3.03

[Jana]

Reason why Istria&Kvarner was grouped with Lika and Dalmatia was the fact it was not different enough from them, at least based on sample size we have so far.
And it's mostly Istria due to history of southern migration + Italian input which pull it south (although it has very northern/Slavic shifted samples too).
If we had more samples from Kvarner, I think that average would be more distinct from Dalmatia & Lika because Kvarner is more native and received almost no southern migration compared to Istria (except Novi Vinodolski hinterland - Bunjevci). I posted sample from Rijeka recently who is Slovenian like genetically (1/8 German but I include 1/8 foreign admix, that's my upper limit-even without such input he should be Slovenian like)

So, until we get more samples from western Croatia I think Littoral average makes sense, all three averages which made it were similar. (and BiH is pretty similar as well).

Gorski kotar on the other hand is distinct and very central Euro as expected so it was merged into north Croatia togheder with Zarorje, Prigorje, Međimurje, Podravina, wider Zagreb region etc.

[Jana]

My idea was the following: Dinaric (BiH and Lika, because Lika population has lot of ties with BiH Croats, almost same genetics too)
Adriatic (Istria,Kvarner,Dalmatia)
Panonnian(Gorski kotar, Central, Zagorje, Slavonia)

I wanted to camouflage Slavonian average by mergining it into Panonnian because it's weakest average we have - only 7 samples. So put togheder with other northerners it would be much stronger.
But maybe Vbnethkio idea was better, because Slavonians are not the same as Kajkavians - less western Euro input and more eastern, they are more like NE Bosnians).

I'm happy Vbn made these averages for us, I could never do it alone.

[MoroLP]

It is fine to decrease number of averages as discussed before, but this has been done with some strange reasoning:

1) The naming, Littoral (Primorje) is a separate region, especially in English language, which is rarely or never used in the meaning of Istria, Lika and Dalmatia. This is some strange mixture of medieval and modern appellations. People from diaspora are certainly going to be confused. A more reasonable name would be Adriatic. However, it is confusing to have averages named in a different manner; by cardinal directions (North, East), regional name (Littoral), and country (BiH). It needs only a single and easily understandable style. The averag "_north" obviously needs letter "n" in all caps "N".

2) Only because a sub-regional average is more closer to another it doesn't mean Gk should be grouped with North average, neither is in the continental or Pannonian part of Croatia. Exactly otherwise, considering the geographical proximity Gk should be grouped with Kvarner and Istria, perhaps even with Lika, making a "West" average.

3) Averages should been according to this Mrsic et al. map. According to it we should have "West" or "Littoral" (Istria, Kvarner, Gk, Lika), "North" or "Pannonia" (Central, Zagorje, Medimurje), "East" or "Slavonia" (Slavonia), and "South" or "Dalmatia" (Dalmatia). Of course, fifth average would be "Bosnia-Herzegovina", and we would have a national average for Croats from Croatia and another cumulative for all Croats (I doubt other national averages are done considering samples from foreign countries). To me that's most simple, understandable and of interest to everyone. Can you or Vbn do this instead or at least try to see current results?

No, I still did not hear anything about those Slavonian samples.

[vbnetkhio]

3) Averages should been according to this Mrsic et al. map. According to it we should have "West" or "Littoral" (Istria, Kvarner, Gk, Lika), "North" or "Pannonia" (Central, Zagorje, Medimurje), "East" or "Slavonia" (Slavonia), and "South" or "Dalmatia" (Dalmatia). Of course, fifth average would be "Bosnia-Herzegovina", and we would have a national average for Croats from Croatia and another cumulative for all Croats. That's most simple, understandable and of interest to everyone. Can you or Vbn do this instead or at least try to see current results?

I had this map in mind, except I merged Southern and Western into "Littoral", and Northern and Central (and Gorski Kotar) into "North", because they look genetically identical, with the current samples at least.

There is no need to include multiple averages which represent the same genetic cluster, because then they just overflood the oracle.

maybe a map should be included along with the oracle so the users can see the geographic range of each region.

[Dunai]

It is fine to decrease number of averages as discussed before, but this has been done with some strange reasoning:

1) The naming, Littoral (Primorje) is a separate region, especially in English language, which is rarely or never used in the meaning of Istria, Lika and Dalmatia. This is some strange mixture of medieval and modern appellations. People from diaspora are certainly going to be confused. A more reasonable name would be Adriatic. However, it is confusing to have averages named in a different manner; by cardinal directions (North, East), regional name (Littoral), and country (BiH). It needs only a single and easily understandable style. The averag "_north" obviously needs letter "n" in all caps "N".

2) Only because a sub-regional average is more closer to another it doesn't mean Gk should be grouped with North average, neither is in the continental or Pannonian part of Croatia. Exactly otherwise, considering the geographical proximity Gk should be grouped with Kvarner and Istria, perhaps even with Lika, making a "West" average.

3) Averages should been according to this Mrsic et al. map. According to it we should have "West" or "Littoral" (Istria, Kvarner, Gk, Lika), "North" or "Pannonia" (Central, Zagorje, Medimurje), "East" or "Slavonia" (Slavonia), and "South" or "Dalmatia" (Dalmatia). Of course, fifth average would be "Bosnia-Herzegovina", and we would have a national average for Croats from Croatia and another cumulative for all Croats (I doubt other national averages are done considering samples from foreign countries). To me that's most simple, understandable and of interest to everyone. Can you or Vbn do this instead or at least try to see current results?

No, I still did not hear anything about those Slavonian samples.

For example Hungarian average takes into consideration also those from neighboring countries, but in a weighted manner based on regional population, so it would be representative of the real numbers of Hungarians and not the number of samples collected. As we know from certain regions you have more or less samples, even if the population of that region can be either much higher or much lower. It surely takes more math and usage of equations, but ultimately it's not super difficult to create weighted national averages.