New map of the Atlantic admixture (Eurogenes K15)

[Joey D]

Population	Percent
East_Med	30.31
West_Med	17.77
West_Asian	17.34
Atlantic	12.77
North_Sea	9.65
Baltic	4.01
Eastern_Euro	3.49
Red_Sea	1.73
Northeast_African	1.23
Oceanian	1.09
Sub-Saharan	0.6
Population (source)	Distance
South_Italian	5.19
Central_Greek	5.69
East_Sicilian	6.66
Italian_Abruzzo	7.68
West_Sicilian	8.79
Italian_Jewish	8.8
Ashkenazi	8.99
Sephardic_Jewish	10.03
Algerian_Jewish	10.37
Greek	10.62

Using 2 populations approximation:
1 50% Central_Greek +50% South_Italian @ 5.471951
Using 3 populations approximation:
1 50% Central_Greek +25% Cyprian +25% Italian_Abruzzo @ 5.225592
Using 4 populations approximation:
1 Algerian_Jewish + Armenian + North_Italian + South_Italian @ 4.703165
2 Armenian + Italian_Jewish + North_Italian + South_Italian @ 4.792818
3 Algerian_Jewish + Armenian + Central_Greek + North_Italian @ 4.795398
4 Algerian_Jewish + Armenian + Greek + Tuscan @ 4.940431
5 Armenian + Greek + Italian_Jewish + Tuscan @ 4.969768

[New Englander]

# Population (source) Distance
1 Italian_Abruzzo 4.84
2 West_Sicilian 5.71
3 Greek 6.85
4 Tuscan 7.53
5 Central_Greek 7.66
6 East_Sicilian 8.05
7 Ashkenazi 8.75
8 Greek_Thessaly 9.29
9 South_Italian 9.68
10 Bulgarian 12.26
11 North_Italian 12.79
12 Italian_Jewish 13.09
13 Sephardic_Jewish 13.16
14 Romanian 13.93
15 Algerian_Jewish 14.33
16 Serbian 16.34
17 Tunisian_Jewish 17.68
18 Spanish_Extremadura 19.07
19 Spanish_Andalucia 19.23
20 Spanish_Murcia 19.36


Mixed Mode Population Sharing:


# Primary Population (source) Secondary Population (source) Distance
1 54.8% Southwest_French + 45.2% Iranian_Jewish @ 3.53
2 94.1% Italian_Abruzzo + 5.9% MA-1 @ 3.54
3 55.1% Southwest_French + 44.9% Kurdish_Jewish @ 3.6
4 92% Italian_Abruzzo + 8% Kargopol_Russian @ 3.73
5 93.2% Italian_Abruzzo + 6.8% Chuvash @ 3.76
6 94% Italian_Abruzzo + 6% Mari @ 3.79
7 92.6% Italian_Abruzzo + 7.4% Erzya @ 3.82
8 89.3% West_Sicilian + 10.7% Kargopol_Russian @ 3.83
9 88.4% West_Sicilian + 11.6% Tatar @ 3.83
10 76% West_Sicilian + 24% Romanian @ 3.84
11 89.8% West_Sicilian + 10.2% Erzya @ 3.86
12 88.8% West_Sicilian + 11.2% Ukrainian_Belgorod @ 3.87
13 79.1% West_Sicilian + 20.9% Serbian @ 3.88
14 92.5% Italian_Abruzzo + 7.5% Estonian_Polish @ 3.9
15 88.6% Italian_Abruzzo + 11.4% Austrian @ 3.91
16 89.3% West_Sicilian + 10.7% Tabassaran @ 3.92
17 92.6% Italian_Abruzzo + 7.4% East_Finnish @ 3.93
18 89.1% West_Sicilian + 10.9% Southwest_Russian @ 3.94
19 92.2% Italian_Abruzzo + 7.8% Russian_Smolensk @ 3.95
20 92.4% Italian_Abruzzo + 7.6% Southwest_Russian @ 3.96

Using 3 populations approximation:
1 50% Cyprian +25% French_Basque +25% Serbian @ 3.616444




Using 4 populations approximation:
1 Austrian + Cyprian + French_Basque + Kurdish_Jewish @ 2.742715
2 French_Basque + Kurdish_Jewish + Sephardic_Jewish + Serbian @ 2.942163
3 Algerian_Jewish + French_Basque + Kurdish_Jewish + Serbian @ 2.979804
4 Algerian_Jewish + French_Basque + Iranian_Jewish + Serbian @ 2.980447
5 French_Basque + Iranian_Jewish + Sephardic_Jewish + Serbian @ 3.016441
6 Algerian_Jewish + Assyrian + French_Basque + Serbian @ 3.021501
7 Cyprian + French_Basque + Lebanese_Muslim + Serbian @ 3.062569
8 French_Basque + Iranian_Jewish + Italian_Jewish + Serbian @ 3.071012
9 Austrian + Cyprian + French_Basque + Iranian_Jewish @ 3.077204
10 French_Basque + Kurdish_Jewish + Moldavian + Sephardic_Jewish @ 3.101743
11 French_Basque + Italian_Jewish + Kurdish_Jewish + Serbian @ 3.136735
12 Algerian_Jewish + French_Basque + Kurdish_Jewish + Moldavian @ 3.164120
13 Cyprian + East_German + French_Basque + Kurdish_Jewish @ 3.170987
14 Kurdish_Jewish + Serbian + Spanish_Aragon + West_Sicilian @ 3.171661
15 Assyrian + French_Basque + Sephardic_Jewish + Serbian @ 3.172798
16 Kurdish_Jewish + Serbian + Spanish_Valencia + West_Sicilian @ 3.175521
17 Iranian_Jewish + Serbian + Spanish_Aragon + West_Sicilian @ 3.203618
18 Azeri + Lebanese_Druze + Southwest_French + Southwest_French @ 3.234926
19 Iranian_Jewish + Serbian + Spanish_Valencia + West_Sicilian @ 3.241805
20 Croatian + Cyprian + French_Basque + Kurdish_Jewish @ 3.243646

[Hauteville]

Population	Percent
East_Med	30.31
West_Med	17.77
West_Asian	17.34
Atlantic	12.77
North_Sea	9.65
Baltic	4.01
Eastern_Euro	3.49
Red_Sea	1.73
Northeast_African	1.23
Oceanian	1.09
Sub-Saharan	0.6
Population (source)	Distance
South_Italian	5.19
Central_Greek	5.69
East_Sicilian	6.66
Italian_Abruzzo	7.68
West_Sicilian	8.79
Italian_Jewish	8.8
Ashkenazi	8.99
Sephardic_Jewish	10.03
Algerian_Jewish	10.37
Greek	10.62

Using 2 populations approximation:
1 50% Central_Greek +50% South_Italian @ 5.471951
Using 3 populations approximation:
1 50% Central_Greek +25% Cyprian +25% Italian_Abruzzo @ 5.225592
Using 4 populations approximation:
1 Algerian_Jewish + Armenian + North_Italian + South_Italian @ 4.703165
2 Armenian + Italian_Jewish + North_Italian + South_Italian @ 4.792818
3 Algerian_Jewish + Armenian + Central_Greek + North_Italian @ 4.795398
4 Algerian_Jewish + Armenian + Greek + Tuscan @ 4.940431
5 Armenian + Greek + Italian_Jewish + Tuscan @ 4.969768

It is impressive that you score a very low Red Sea, kinda like a French or an German. For example I score 3% instead.

[New Englander]

^ Looks like his North Sea and Baltic are taking away from his Red Sea. Everything else seems about right. Might mean Norman Ancestry.

[Angela]

There are always going to be slight variations, even between siblings and other family members. I don't think there is enough appreciation for how much chance comes into all of this when you're looking not at averages but at an individual's results.

All this over-interpretation of results and trying to correlate them with specific historical events is highly problematic.

I know that probably because of gedmatch Admixture calculators are very popular, but they're not easy to interpret and, in fact, very easy to misinterpret.

You also need to look at formal stats to get a more complete picture.

[Hauteville]

^ Looks like his North Sea and Baltic are taking away from his Red Sea. Everything else seems about right. Might mean Norman Ancestry.

More likely North Western Italian judging by his countries, since there are town near of him that are Lombard or part so, like Randazzo, Maniace etc. At the same time, i also myself score high East Euro and Baltic admix (on Eurogenes K13 the last component is almost 12%) and my Y-DNA is common in the Balkans but I've find contacts from Germany and even England. Strange, really strange.

[Hauteville]

There are always going to be slight variations, even between siblings and other family members. I don't think there is enough appreciation for how much chance comes into all of this when you're looking not at averages but at an individual's results.

All this over-interpretation of results and trying to correlate them with specific historical events is highly problematic.

I know that probably because of gedmatch Admixture calculators are very popular, but they're not easy to interpret and, in fact, very easy to misinterpret.

You also need to look at formal stats to get a more complete picture.

Indeed, I agree and the results can change from calculator to calculator. My results have variation, for example on my DNA Land I score 31% Balkan while on 23andme I've 5%, at the same time on this last tool I'm mostly Italian, close to 90+%.

[Angela]

Not all calculators are equal. I can guarantee you that most people get higher northeastern and eastern Europe on the eurogenes calculators. They're designed that way.

Then you have to factor in that not only in calculators, but in commercial testing results, clusters that bear the same name may not represent the same exact genetic group.

As for DNA Land versus 23andme, as we've discussed before, 23andme has a cluster centered on the Italian peninsula. It also has a specific Balkan cluster. DNA Land looked at the alleles spread across north/north Central Italy and the Balkans and called the cluster Balkan. That gave a lot of Italians a lot of Balkan. Had they called it North/Central Italian a lot of people in the Balkans would have gotten a lot of "Italian". It's all the same genes.

You just can't take these things all that literally.

In Joey's case, given the location of his ancestral area, northwestern Italians are probably a better bet, but it might just be chance.

At 23andme my only non-northwest Italian close "cousins" are from Denmark. Does it have something to do with the "Lombard" castles dotting every hill or is it just another one of those "false flags", IBD sharing that is just the result of recombination of stray bits of dna? Who knows?

[Joey D]

I agree with these posts, these various tests throw up all sorts of results, and clearly some place a greater weighting on one or another region.

So for me, it's just a bit of fun, and to be frank, it doesn't keep me up at night.

Having said that, there are obvious pointers which keep cropping up which I think do carry some significance: Greece (obviously), the Near East and the Caucasus - I'm especially interested in the latter, and also this Ashkenazi result which keeps popping up everywhere (granted, I'm starting to understand why the Ashkenazi and Sicilians plot closely together).

[Hauteville]

In Joey's case, given the location of his ancestral area, northwestern Italians are probably a better bet, but it might just be chance.

It is interesting that in the Fiorito's IBDs the first Italian region is not Basilicata and Calabria who makes more sense for geography and history but Liguria. A signal of North Italian settlements for sure judging also by the position of Val d'Aosta.

https://docs.google.com/spreadsheets...iCw/edit#gid=0

[Angela]

It is interesting that in the Fiorito's IBDs the first Italian region is not Basilicata and Calabria who makes more sense for geography and history but Liguria. A signal of North Italian settlements for sure judging also by the position of Val d'Aosta.

https://docs.google.com/spreadsheets...iCw/edit#gid=0

Thanks a lot for that link. One quibble: in the notes I think "Northwest" is meant, not "Northeast".

Extraordinary that Liguria comes in first for Sicilia, and Aosta second for Calabria. I think I've mentioned before that one of my husband's best clients was a Sicilian American whose family came from Catania. He was an avid genealogist, and had traced his family tree in the male line all the way back to a Genovese who had relocated to Sicilia in the Middle Ages. Apparently they continued to get wives from home for quite a long time. I was more than fond of him for many reasons. He became, in fact, a surrogate father for me when my own father died. He didn't look like him at all, but he looked unnervingly like my Spezzino maternal grandfather: lean and spare, small boned, elegant, light on his feet, and very bright and quick witted. The surname appears both in Liguria and Sicilia, and I don't think modern immigration is the only reason.

Interesting in the link also is that sometimes geography doesn't really explain the genetic relationships. As you say, Toscana is equidistant to Liguria and Emilia, but the link with Emilia is much stronger. Toscana is also equidistant from Emilia and Lazio, and in fact there is the huge barrier of the Appennini between Emilia and Toscana, and yet Toscana is closer to Emilia.

[Pax Augusta]

Thanks a lot for that link. One quibble: in the notes I think "Northwest" is meant, not "Northeast".

The doc "IBD states, Fiorito 2015" was made by Fire Haired14 based on the Fiorito 2015 paper.

http://www.eupedia.com/forum/threads...l=1#post470195

Extraordinary that Liguria comes in first for Sicilia, and Aosta second for Calabria. I think I've mentioned before that one of my husband's best clients was a Sicilian American whose family came from Catania. He was an avid genealogist, and had traced his family tree in the male line all the way back to a Genovese who had relocated to Sicilia in the Middle Ages. Apparently they continued to get wives from home for quite a long time. I was more than fond of him for many reasons. He became, in fact, a surrogate father for me when my own father died. He didn't look like him at all, but he looked unnervingly like my Spezzino maternal grandfather: lean and spare, small boned, elegant, light on his feet, and very bright and quick witted. The surname appears both in Liguria and Sicilia, and I don't think modern immigration is the only reason.

Yes, it's not so uncommon in Sicily and other parts of South Italy, especially for Sicilians, medieval ancestry from North-west Italy.

Interesting in the link also is that sometimes geography doesn't really explain the genetic relationships. As you say, Toscana is equidistant to Liguria and Emilia, but the link with Emilia is much stronger. Toscana is also equidistant from Emilia and Lazio, and in fact there is the huge barrier of the Appennini between Emilia and Toscana, and yet Toscana is closer to Emilia.

Interestingly enough the Tuscan sample is from Southern Tuscany (Siena and Arezzo) that borders Lazio (Viterbo), but despite this proximity the Tuscan sample is closer to the Emilian sample from Ferrara, one of the most distant Emilian province from Tuscany (Ferrara borders Rovigo in Veneto). The Ligurian sample is from Savona, so unsurprisingly closer to Piedmont, that part of Liguria has a lot of shared history with Piedmont.

[Sile]

Yes, it's not so uncommon in Sicily and other parts of South Italy, especially for Sicilians, medieval ancestry from North-west Italy.

the "link" did decrease in numbers after the 1430 - fifth Genoese-Venetian war.............where Genoa and Milan where fighting against Venice and Aragon.

Clearly Aragon must have owned most of southern italy as well as sardinia

Their annoyance with Genoa and its Corsican ownership was very much against any Genoese association.

[Pax Augusta]

the "link" did decrease in numbers after the 1430 - fifth Genoese-Venetian war.............where Genoa and Milan where fighting against Venice and Aragon.

Clearly Aragon must have owned most of southern italy as well as sardinia

Their annoyance with Genoa and its Corsican ownership was very much against any Genoese association.

I was talking more of these

http://www.treccani.it/enciclopedia/...#39;Italiano)/

https://it.wikipedia.org/wiki/Lombardi_di_Sicilia

https://it.wikipedia.org/wiki/Gallo-italico_di_Sicilia

[Angela]

One country can rule another and the impact genetically can be negligible if discernible at all. It's different when you're talking about a virtual state sponsored colonization, as is the case we're discussing with the so called "Lombard" cities and other immigration.

Perhaps someone should inform Fire-Haired that Genova, Aosta, etc. are in northwestern Italy, so he can correct the typos.

[Joey D]

Interesting spreadsheet, I would not have picked that in a million years.

[Pax Augusta]

Perhaps someone should inform Fire-Haired that Genova, Aosta, etc. are in northwestern Italy, so he can correct the typos.

Also that Emilia-Romagna is in north Italy and not in central Italy.

[Angela]

Given medieval Sicilian history it's not really a surprise to me that a lot of IBD sharing with northwestern Italians shows up.

I'm more surprised by the results for Calabria, Basilicata, etc. as I'm not aware of any actual "colonization" type migration from the northwest during the Middle Ages.

Some of it has got to be older then. Could it be from Roman veteran settlements? Is it older yet? Italics?

@Pax Augusta,
Isn't the Fiorito paper the one where they say Emilia-Romagna is in central Italy and then find that their Emilian samples plot with Northern Italians and not Central Italians or even Tuscans? :) In that case it would be their error, not Fire-Haired's.

Even some Italian researchers can be clueless sometimes.

[Hauteville]

Basilicata had also colonization from North Italy in some zones.

[MOESAN]

There are always going to be slight variations, even between siblings and other family members. I don't think there is enough appreciation for how much chance comes into all of this when you're looking not at averages but at an individual's results.

All this over-interpretation of results and trying to correlate them with specific historical events is highly problematic.

I know that probably because of gedmatch Admixture calculators are very popular, but they're not easy to interpret and, in fact, very easy to misinterpret.

You also need to look at formal stats to get a more complete picture.

totally agree

[Pax Augusta]

Given medieval Sicilian history it's not really a surprise to me that a lot of IBD sharing with northwestern Italians shows up.

I'm more surprised by the results for Calabria, Basilicata, etc. as I'm not aware of any actual "colonization" type migration from the northwest during the Middle Ages.

Some of it has got to be older then. Could it be from Roman veteran settlements? Is it older yet? Italics?

Basilicata, southern Campania and at some extent northern Calabria, all places around the gulf of Policastro, have had the same migrations from northwest Italy during the Middle Ages.

https://it.wikipedia.org/wiki/Dialet..._di_Basilicata

Savona, especially the Val Bormida, was one of the places of origin of these colonisers.

Calabria has had also "Occitan" migrations from val Pellice in Piedmont but the Calabrian sample used in the Fiorito paper is from Reggio Calabria.

http://www.chambradoc.it/occitaniaGr...iemontese.page

@Pax Augusta,
Isn't the Fiorito paper the one where they say Emilia-Romagna is in central Italy and then find that their Emilian samples plot with Northern Italians and not Central Italians or even Tuscans? :) In that case it would be their error, not Fire-Haired's.

Not really, the Fiorito paper is correct, Emilia-Romagna is grouped with the Northern Italians and is never considered in the paper as Central Italy. Why should it be? There is no reason to consider Emilia-Romagna as a Central Italian region. Linguistically, culturally and even geographically it's northern Italy. Obviously Emilia-Romagna is the northern Italian region that shares more with the central Italy and less with the far northern Italy, the Italian Alps. At some extent, and in a different way, this is also true for Liguria.

For this analysis, the Italian regions were grouped according to the five previously identified clusters (Northern, Central, Southern Italy, Aosta Valley, Sardinia).

Even some Italian researchers can be clueless sometimes.

I totally agree with you.

[Joey D]

Given medieval Sicilian history it's not really a surprise to me that a lot of IBD sharing with northwestern Italians shows up.

I'm more surprised by the results for Calabria, Basilicata, etc. as I'm not aware of any actual "colonization" type migration from the northwest during the Middle Ages.

Some of it has got to be older then. Could it be from Roman veteran settlements? Is it older yet? Italics?

@Pax Augusta,
Isn't the Fiorito paper the one where they say Emilia-Romagna is in central Italy and then find that their Emilian samples plot with Northern Italians and not Central Italians or even Tuscans? :) In that case it would be their error, not Fire-Haired's.

Even some Italian researchers can be clueless sometimes.

There are a couple of separate (but related) events happening over the space of some 75 years. I know you will know much of this already, but it is worth laying it out, if nothing else, for completeness.

So the Normans first started venturing into the far South around 1030, as part of a pilgrimage to view some holy relics somewhere (I can't even remember where now).

Around this time, the Byzantine grip on the far South was weakening, in part because Lombards had already set up come city-states in places like Salerno (I'm going from memory here, so bear with me). So in this period, from 1030 to 1050 there is already a sizeable Lombard presence in Southern Italy.

At this point, one word of caution - the Byzantines had a habit of calling anyone who didn't speak Greek on the Italian peninsula as being a "Lombard". So it's unclear whether we are talking about actual Lombards, and/or whether it's a mixture of Italic speakers from other parts of the peninsular which may have included some Lombards.

So, even before the Norman conquest of Sicily, we already have various city-states, principalities and settlements in Southern Italy, occupied by people called generically "Lombard" in an era where Byzantine power is weakening in Southern Italy.

The Normans use this opportunity to conquer all of Southern Italy, and on and off, these Lombards were their allies.

Then, we get the Norman conquest of Sicily, undertaken by the youngest of the Hauteville brothers, Roger. It takes some 30 years. Just prior to the completion of the conquest, 1089,, Roger marries his third wife, Adelaide del Vasto, with notable family links to Montferrat. She granted her brother Paterno' (Pr. Catania), and from that point there is large-scale immigration into central and Eastern Sicily to help re-populate the interior which had yet to recover from the recent wars.

Geoffrey Hull writes: "Most of the settlements in the east of the island were founded by pioneers from the Monferrat region of the western Po valley...Padanian immigration to Sicily had been promoted by Roger I's marriage to Adelaide, daughter of the Marquis of Monferrat.... her brother Henry married Roger I's daughter Blandina, ruled the County of Paterno', the nucleus of "Lombard" settlement in eastern Sicily.

AS a footnote, he writes: A third and smaller Padanian colony was established in the Trecchina district of north-west Calabria, above Maratea on the Gulf of Policastro.

[Joey D]

Ah, I see Pax has already mentioned Policastro, if you read my huge story above, you'll see that I eventually get to Policastro as well.

[Pax Augusta]

Having said that, there are obvious pointers which keep cropping up which I think do carry some significance: Greece (obviously), the Near East and the Caucasus - I'm especially interested in the latter, and also this Ashkenazi result which keeps popping up everywhere (granted, I'm starting to understand why the Ashkenazi and Sicilians plot closely together).

It depends by the calculator and the sample chosen. Generally speaking Ashkenazis and Sicilians plot closely together because they have similar amount of ancestral components, but it's not due to a direct source, and Ashkenazis are usually more eastern-shifted than Sicilians as a whole.

These are based on MDLP K23b, where Ashkenazis and many Greek samples plot closely together as well.

[New Englander]

If those regions have admixture from the past 1000 from the North, than what does that say about the population before than?