Wednesday, June 8, 2011

Northern Italians are biologically SOUTHERN European.

Above : Pictures of Venice Italy

It is often stated that Northern Italians are "Germanic" (don't know why they single out Northern Italians as Germanic 'barbarians' also invaded and controlled the south of Italy and Sicily in historic times), or more related to Germans and Northern Europeans than to the Southern Italians or other Southern Europeans; this notion however is completely incorrect. More recent genome wide association studies generally show Northern Italians clustering with Southern Europeans, pulling further south and distant from Northern Europeans than Spaniards. Here are several genetic studies demonstrating this fact.

European Population Substructure: Clustering of Northern and Southern Populations. Seldin et al (2006).
"Using a genome-wide single nucleotide polymorphism (SNP) panel, we observed population structure in a diverse group of Europeans and European Americans. Under a variety of conditions and tests, there is a consistent and reproducible distinction between “northern” and “southern” European population groups: most individual participants with southern European ancestry (Italian, Spanish, Portuguese, and Greek) have >85% membership in the “southern” population; and most northern, western, eastern, and central Europeans have >90% in the “northern” population group. Ashkenazi Jewish as well as Sephardic Jewish origin also showed >85% membership in the “southern” population, consistent with a later Mediterranean origin of these ethnic groups. Based on this work, we have developed a core set of informative SNP markers that can control for this partition in European population structure in a variety of clinical and genetic studies."

The recent development of methodologies for defining population structure has provided the ability to identify the major ethnic contributions in individual participants in diverse populations. These statistical approaches utilize non-hierarchical clustering algorithms in which Markov chain Monte Carlo methods are used to infer ancestry, based solely on genotyping information. Furthermore, related algorithms provide methods for controlling for population stratification in genetic studies. These methods are important in assessing the results of case-control and other non–family-based association tests. In addition, defining population structure is potentially useful both in clinical outcome studies and in the examination of pharmacologic response and toxicity. Previous studies of human population structure have primarily considered different continental populations or admixed populations between two or more different continental populations. However, some of these studies have also suggested that sub-continental differences in population structure can be discerned. The examination of population differences within Europe using mitochondrial or Y chromosome haplogroups has been particularly useful in tracing part of the routes of migration and populating of Europe, but these haplogroups do not provide strong inferences on population genetic structure. Autosomal studies using small numbers of classical genetic markers (nuclear protein polymorphisms) have suggested broad genetic gradients across Europe, leading to the proposal of demic diffusion models. These include a principal component analysis of small numbers of classic genetic markers that suggested three clines accounting for a proportion of the genetic variation in the continent.

A total of 1,094 participants were genotyped with more than 5,700 SNPs distributed over the entire genome. After excluding participants with > 10% estimated non-European ancestry (see Methods), 928 participants were selected for further analysis.

Populations studied: First sample set.

European Americans of different regional European origins (681 participants), East Asian Americans (13), African Americans (22), South Asian Americans (48), Amerindians (48), and Swedish (92), Finnish (13), Italian (91), Portuguese (3), southern France (1), and Spanish participants (82) were included in this study. None of the individuals were first-degree relatives of other participants in the study. These populations were based on self-identified ethnic affiliation. The European Americans, African Americans, and East Asian Americans were recruited from across the United States, and the majority of the participants, including all of the European Americans, were RA probands identified as part of the North American Rheumatoid Arthritis Consortium (NARAC) as previously described (61). The South Asian American participants were recruited from Houston, Texas, and Amerindian participants were self-identified as Mayan (Kachiquel language group) and were recruited in Chimaltenango, Guatemala, as previously described. The Italian participants were normal healthy volunteers recruited from throughout Italy: 38 from northern Italy, 23 from central Italy, and 30 from southern Italy. The Swedish and Finnish participants were healthy normal controls collected in these countries. The other participants recruited in southwestern Europe included 86 from Spain, three from Portugal, and one from southern France. Of the Spanish participants, there were 43 from northern Spain, 12 from central Spain, and 19 from southern Spain. Of these participants from Spain and Portugal, 61 were probands for a multiple sclerosis study. Blood cell samples were obtained from all individuals, according to protocols and informed-consent procedures approved by institutional review boards, and were labeled with an anonymous code number linked only to demographic information.

For the European Americans, grandparental information was available for the majority of the participants. These included the following self-identifier classifications of grandparents: western European (United Kingdom, northern France, Holland, Belgium, and Switzerland), eastern European (Russia, Poland, Romania, Ukraine, Lithuania, Latvia, Estonia, and Czech Republic), central European (Germany, Austria, and Hungary), southern European (Spain, Portugal, Italy, and southern France), Scandinavian (Denmark, Norway, Sweden, and Finland), and eastern Mediterranean (Greece, Turkey, Croatia, Bosnia, Yugoslavia, and Albania), Sephardic Jewish American, and White French Canadian. All participants with any reported mixed-continental origins (e.g., African) were excluded".

"Grouping of individuals with different north–south contributions from the k = 2 analysis further illustrates this division of individual participants from different European population sets and some of the variability observed (Figure 2). Italy (84 of 86 individuals), Spain (66 of 74), Portugal (3 of 3), and Sephardic Jewish Americans (3 of 3) had majority contributions from the “southern” population group as defined by this population structure analysis. In addition, a large fraction of southern European Americans (7 of 11) without other reported European heritage had majority “southern” contribution. Those Americans with self-identified mixed “southern” and “northern” heritage showed a substantial but less impressive “southern” population component (8 of 23 with majority “southern”). Those American participants with mixed eastern Mediterranean–reported heritage also had two of ten individuals with a majority “southern” population component. All other groups showed only a few isolated participants with more than a limited “southern” population component.
Trends in both the Italian and Spanish participants were also consistent with this north–south pattern: ten of 32 participants from northern Italy had greater than a 10% “northern” component compared with two of 28 from southern Italy; and 23 of 43 from northern Spain had greater than a 10% “northern” contribution compared to five of 19 from southern Spain."Analyses were performed with 2,657 SNPs under the condition of two population (Pop) groups (k = 2)."
FIGURE 1 Analysis of Population Structure in Participants of European Ancestry"Analysis was performed without any prior population assignment using STRUCTURE. The European ancestry groups, western European American (WEURA), central European American (CEURA), eastern European American (EEURA), Finland (FND), Italy (ITN), Spain (SPN), Sweden (SWD), and other European American (OEURA) are indicated by color code. The latter group consisted of individuals with mixed European ancestry from several regions and additional smaller groups (see Methods)....In (D), the proportion of each cluster (ordinate) is shown for each individual."

As we see above, each little line represents 1 Italian. As we can see, the Italians are scorching red here, even the Northern Italians; it looks like 1 (possibly although not certainly Northern) Italian is half blue.

FIGURE 2 - Distribution of “Southern” Population Components among Participants with Various Self-Identified Ethnic or Regional European Origins
"For each self-identified group, the fraction of individual participants in each group with the color-coded frequency “southern” contribution is shown. For southern European American I (SEA1), only southern European grandparents were identified. For southern European American II (SEA2), grandparents were self-reported as being of both southern European decent and western, central, or eastern European decent. For the eastern Mediterranean American (EMEDA) group, four of ten were of mixed-European decent with one or more grandparents of western, central, or eastern European decent.

CEURA, central European American; EEURA, eastern European American; FND, Finland; ITN, Italy; PTG, Portugal; SCNA, Scandinavian; SJA, Sephardic Jewish American; SPN, Spain; SWD, Sweden; WEURA, western European American; WFC, White French Canadian."
As we can see above, the Italians (38 of 87 of whom are Northern Italians) are scorching red (90%+ 'Southern'), only a minority are orange (60-90% 'southern'), while none are entirely or predominately in the blue (Northern) cluster. There are more Swedes (1 outlier) who are in predominately 'Southern' cluster than Italians who are predominantly in the 'Northern' cluster.

So yes, virtually every Northern Italian clustered predominantly in the 'Southern' cluster, while only 10 of 32 Northern Italians showed less than a 90% contribution to the 'Southern' cluster (compared to 23 of 43 from Northern Spain, and 5 of 19 from Southern Spain), and only 2 of 28 participants from Southern Italy had less than a 90% contribution to the 'Southern' cluster. The remaining Italians were between 60% and 90% in the Southern cluster. Northern Italians are overwelmingly 'Southern' in biological affinity when speaking in a European wide context; Southern Italians more-so, but Spaniards slightly less. 

SOURCE : Seldin MF, Shigeta R, Villoslada P, Selmi C, Tuomilehto J, et al. (2006) European Population Substructure: Clustering of Northern and Southern Populations. PLoS Genet 2(9): e143. doi:10.1371/journal.pgen.0020143

Refinement of ancestry informative markers in Europeans. Tian et al (2009).
"The definition of European population genetic substructure and its application to understanding complex phenotypes is becoming increasingly important. In the current study using over 4,000 subjects genotyped for 300,000 single-nucleotide polymorphisms (SNPs), we provide further insight into relationships among European population groups and identify sets of SNP ancestry informative markers (AIMs) for application in genetic studies. In general, the graphical description of these principal components analyses (PCA) of diverse European subjects showed a strong correspondence to the geographical relationships of specific countries or regions of origin. Clearer separation of different ethnic and regional populations was observed when northern and southern European groups were considered separately and the PCA results were influenced by the inclusion or exclusion of different self-identified population groups including Ashkenazi Jewish, Sardinian, and Orcadian ethnic groups. SNP AIM sets were identified that could distinguish the regional and ethnic population groups. Moreover, the studies demonstrated that most allele frequency differences between different European groups could be controlled effectively in analyses using these AIM sets. The European substructure AIMs should be widely applicable to ongoing studies to confirm and delineate specific disease susceptibility candidate regions without the necessity of performing additional genome-wide SNP studies in additional subject sets."
"Figure 1. Principal component analyses of substructure in a diverse set of subjects of European descent. Graphic representation of the first two PCs based on analysis with >250K SNPs are shown. Color code shows subgroup of subjects for each population group.
The subjects included Adygei (ADY, 12 subjects), Ashkenazi Jewish American (AJA, 40 subjects), Basque (BAS, 12 subjects), Bedouin (BDN, 23 subjects), CEPH European American (CEU, 48), Druze (20 subjects), Eastern European American (EEUR, 11 subjects), German American (GERM, 17 subjects), Greek American (GRK, 7), Hungarian American (HUN, 4 ), IRISH (84 subjects), Italian American (ITN, 20 subjects), northern Italian (ITN_N, 13 subjects), Dutch American (NETH, 3 subjects), Orcadian (ORC, 14 subjects), Palestinian (PAL, 22 subjects), Russian (RUS, 13 subjects), Sardinian (SARD, 28 subjects), Scandinavian American (SCAN, 6 subjects ), Spanish (SPAIN, 12 subjects), Swedish (SWED, 591 subjects), Tuscany (TUSC, 8 subjects), and United Kingdom American (UK, 5 subjects). Each of the specific country or ethnic color coded origins had consistent 4 grandparent origin information. The total number of individuals in this analysis was 4446. In panel A European Americans (EURA) without 4 grandparental information are shown (contains both NYCP and CHOP). Panels B and C illustrate the distribution of the EURA from NYCP (1873 subjects) and CHOP (1488 subjects), respectively."

As we can see by PCA analysis, Tuscans (TUSC) and Northern Italians (ITL N) group well within the Southern European cluster, far away from the Northern/Central/East European cluster. Italian Americans (ITN) with 4 Italian grandparents of Southern Italian/Sicilian origin overlap with the Greeks, while Northern Italians (ITL N) seem to be smack dab in the middle of the Southern European cluster, between Spaniards and Tuscans. Tuscans also pull even further south than the Northern Italians, between Southern Italians/Greeks and Northern Italians.

The lack of important European groups like the French and Swiss does seem to make the gap between Southern Europe and Northern/Central/Eastern Europe very large and not just gradual or clinal, but if those two populations were to be included the gap from Southern Europe to Central Europe would appear less significant with the alps showing a large (but not massive) barrier. Anyway, this study also irrefutably proves the Southern European biological affinities of Northern Italians and that is what is important.

Also, here is an interesting Fst genetic distance table from the study, although it unfortunately doesn't include Northern Italians it does include Tuscans (ie where the Renaissance first sprung up).


The results as seen in Table 1.

Tuscany to Southern Italy - 0.0004
Tuscany to Greece - 0.0005
Tuscany to Germany - 0.0032
Tuscany to Ireland - 0.0055
Tuscany to Sweden - 0.0061
Tuscany to Russia - 0.0108

SOURCE : Tian et al. (2009). European Population Genetic Substructure: Further Definition of Ancestry Informative Markers for Distinguishing Among Diverse European Ethnic Groups. Mol Med; doi: 10.2119

Link PDF

Genetic Structure of Europeans: A View from the North–East. Nelis et al (2009).
"Using principal component (PC) analysis, we studied the genetic constitution of 3,112 individuals from Europe as portrayed by more than 270,000 single nucleotide polymorphisms (SNPs) genotyped with the Illumina Infinium platform. In cohorts where the sample size was >100, one hundred randomly chosen samples were used for analysis to minimize the sample size effect, resulting in a total of 1,564 samples. This analysis revealed that the genetic structure of the European population correlates closely with geography. The first two PCs highlight the genetic diversity corresponding to the northwest to southeast gradient and position the populations according to their approximate geographic origin. The resulting genetic map forms a triangular structure with a) Finland, b) the Baltic region, Poland and Western Russia, and c) Italy as its vertexes, and with d) Central- and Western Europe in its centre. Inter- and intra- population genetic differences were quantified by the inflation factor lambda (λ) (ranging from 1.00 to 4.21), fixation index (Fst) (ranging from 0.000 to 0.023), and by the number of markers exhibiting significant allele frequency differences in pair-wise population comparisons. The estimated lambda was used to assess the real diminishing impact to association statistics when two distinct populations are merged directly in an analysis. When the PC analysis was confined to the 1,019 Estonian individuals (0.1% of the Estonian population), a fine structure emerged that correlated with the geography of individual counties. With at least two cohorts available from several countries, genetic substructures were investigated in Czech, Finnish, German, Estonian and Italian populations. Together with previously published data, our results allow the creation of a comprehensive European genetic map that will greatly facilitate inter-population genetic studies including genome wide association studies (GWAS)."

Yet again, as we can see by PCA analysis, Northern Italians blend perfectly in the Southern European area, and are the second most "southerly" population sampled here after the Southern Italians. Northern Italians pull further south than Spaniards from Madrid, and while they seem to be a little closer to the Spanish and some Swiss (who are partly Italian) than to Southern Italians, they are equidistant between Southern Italians and the French, and they're much closer to the Southern Italians than to Southern Germans. The Northern Italians also surprisingly pull quite a bit further to the south end of Europe than Bulgarians do, which might mean Bulgarians have substantial medieval Slavic admixture (Bulgarians are southern Slavs after all) which would have brought more "northerly" gene-flow into Bulgaria; there's no telling really regarding them. 

One more thing of note, unlike the other previous two studies, in this study the samples that represented Northern and Southern Italy only came from 2 different locations; here are the sample set locations as representative of "Northern Italy" and "Southern Italy".
Northern Italy:
"The Northern Italian samples have been randomly collected from the Borbera Valley. The Borbera Valley is located in Northern Italy, namely the northern part of the Apennines mountains, between Liguria and Piedmont, about 80 km south of Milan. For the current analysis, 96 individuals were selected from the database where all the samples have complete phenotypic data."

Southern Italy:
"The Italian samples were randomly chosen from those enrolled in a population study named Carlantino Project, which is focused on inhabitants arising from an isolated village at the border between Central and Southern Italy (Province of Foggia, Region of Apulia with 1200 inhabitants)."

Also, it is important to note that while genetic distance is often heavily correlated to geographical distance, genetic barriers can be created by substantial geographical barriers which limit gene flow and cause breeding isolation. In Europe the Alps represent such a genetic barrier and it is not surprising that this study confirms that.
"Barrier analysis revealed genetic barriers between Finland, Italy and other countries, as has been described before [12]. Interestingly, barriers could be demonstrated within Finland (between Helsinki and Kuusamo) and Italy (between northern and southern part). Another barrier emerged between the Eastern Baltic region and Sweden, but not between the Eastern Baltic region and Poland (Figure S4). The barrier between Bulgaria and Western Russia, Poland and Lithuania may have arisen due to the fact that several populations are missing in between those countries"

As stated above there is a noticeable difference between Northern and Southern Italians, according to PC analysis the medians of North and South Germans seem to be roughly 0.005 apart, while those of North and South Italians seem to be 0.005 apart. So no big difference among the ethnic stratification between the two countries. The biological difference between Northern and Southern Italians though rather noticeable, does seem to have been somewhat exaggerated by laymen 'racialists' of past and present.
"Interestingly, PC analysis was also capable of highlighting intra-population differences, such as between the two Finnish and the two Italian samples, respectively. A low level of intra-population differentiation in Germany has been reported previously, and was confirmed here. In addition, we detected intra-population differences within the Czech and Estonian samples."

SOURCE : Nelis M, Esko T, Mägi R, Zimprich F, Zimprich A, et al. (2009) Genetic Structure of Europeans: A View from the North–East. PLoS ONE 4(5): e5472. doi:10.1371/journal.pone.0005472

Also, other than the Greeks in Southern Italy, historical admixture seems to be quite minimal in Italians.
"An analysis of the genetic makeup of Italy's modern population argues that the various distinctive genetic combinations currently found in different regions within the peninsula by and large track the linguistic distribution that resulted from the migrations of the Iron Age. No data indicate the subsequent large-scale infusion of new genetic material into the populations of these regions except in the case of southern Italy and eastern Sicily, which is explained by the well-documented Greek migrations there."
SOURCE : Rosenstein, Nathan. Rome at War: Farms, Families, and Death in the Middle Republic. Chapel Hill: University of North Carolina Press, 2004.


The genetic article which that book used as a source for that statement is Piazza's "A Genetic History of Italy", and although unfortunately I don't have access to the entire study, the abstract is available online. It's conclusions are clear that the biological stratification seen in today's Italy is reflective of pre Roman times.

"Statistical techniques for displaying the geographical distribution of many genes in few synthetic images have been used to represent the various patterns of gene frequencies in Europe and in the world (Menozzi et al. 1978; Piazza et al. 1981a). It has also been shown that such synthetic displays are particularly useful in detecting clines of genetic differentiation associated with movements of populations like those accompanying the Neolithic expansion of farmers from the Near East or, in more recent times, the putative diffusion of Indo-European-speaking populations (Ammerman & Cavalli-Sforza, 1984; Gimbutas, 1973).

In this paper we use the same combination of statistical and graphical techniques to study the genetic structure of Italy, a European country whose unity of people and cultures was quite a recent event. The possibility of studying genetic differentiation in a small geographical area is tested and trends of genetic differences are tentatively interpreted in terms of historic and linguistic knowledge. The few demographic pieces of information taken from historical sources and compared with linguistic records support the hypothesis that the genetic structure of Italy still reflects the ethnic stratification of pre-Roman times."
SOURCE : Piazza, A., N. Cappello, E. Olivetti, and S. Rendine. 1988. A genetic history of Italy. Ann. Hum. Genet. 52:203-213.


(EDIT 2015) Admixture ratios in modern populations based on prehistoric genomes

All populations in Europe can be modeled as being a mixture of 3 major and distinct ancestral strains from prehistoric migrations. These are 

1. Paleolithic West European Hunter Gatherers

2. Early European Farmers of the Neolithic revolution who had most their genetic origin in Mesolithic and Neolithic Anatolia and brought the earliest agricultural practices to Europe.

3. Early Bronze age Yamnaya steppe pastoralists from the Pontic-Caspian steppe, who's ancestors were the first to domesticate the horse. They were also one of the first groups on Earth who knew of and used wheeled vehicles. The Yamnaya steppe pastoralists are believed by most linguists to have been the speakers of the late stage proto Indo-European language (see Kurgan hypothesis). 

As we can see in Figure 3 below, all Europeans can basically be modeled as a composite of the same 3 ancestral mixtures, it's just that the ratios are different. We can also see that that the biggest differences in Europe are between the North of Europe, and the South of Europe. As we can see in their admixture ratios, the Northern Italians (here represented by a sample set from Bergamo Lombardy) are unsurprisingly most similar to other Southern Europeans (ie Northern Spanish, Greeks, Albanians, Bulgarians, etc) in their admixture ratios, and that is being of heavy Early Neolithic Farmer ancestry. Tuscans as anyone can see also come out very heavily Neolithic.

SOURCE : Haak, et al. (2015). "Massive migration from the steppe was a source for Indo-European languages in Europe". Nature. doi:10.1038/nature14317.edit


And lastly, for the most part irrelevant, I found some old hair and eye pigmentation data from Livi on Italian conscripts. Interesting stuff but for the most it's not that important.

SOURCE : Antropometria Militare. Ridolfo Livi. Two volumes, 4°, with atlas. Rome: Preso il Giornale Medico del Regio E Sercito, 1896, 1905..

Summary and conclusions
1. Northern and Central Italians are biologically SOUTHERN European...broadly speaking. In so far that they share their overall closest genetic relationships with other Southern Europeans, and that they are more distinct from Northern European groups than the Spanish.

2. Northern-Central Italians may be a bit closer to North/Central/East Europeans than Southern Italians are (just as Southern Germans are a bit closer to Southern Europeans than Northern Germans are), but Northern-Central Italians still biologically pull to Southern Europe quite strongly as seen that they ALWAYS pull further south than Spaniards, and also one study showed them a bit below Balkan Southern Slavs like Bulgarians. They also always structure with Southern Europeans in European exclusive STRUCTURE analysis studies. And their admixture ratios based on actual ancient genomes using formal admixture modeling is much more similar to the ratios present in other Southern Europeans than in Northern Europeans.

3. Slight Germanic admixture is possible for all Italian populations, not just the Northern Italians. I would venture that actual Germanic ancestry is larger in Northern Italy and that the Lombards in Northern Italy had a bigger impact than the Goths or Normans in Southern Italy. However one thing is for certain, and that is that large scale Germanic admixture in any Italian groups is certainly not the case and easily refuted by the data. Anything beyond that is total conjecture at this point.

4. There is a noticeable difference between the Northern Italians and Southern Italians, and I don't want people to think that Northern Italians cannot be argued to be a genetically distinct. Also, the genetic distance between Northern and Southern Italians, although pretty large from a single European 'nationality' point of view, is only roughly equal [marginally greater] to that of the genetic distance between Northern Germans and Southern Germans.


  1. I think the differences between Northern and Southern Italians is clear; it doesn't come from Germanic admixture in the North, or Middle Eastern admixture in the South, but from the fact Northern Italy was already *well* populated by Gauls prior to being incorporated into Roman Italy (which one could argue was only done to fortify the Italia homeland, with the Alps).

    1. Yes indeed, not to mention Greek admixture in the South.

      I think as far as Germanic admixture is concerned, it's minimal in both Southern and Northern Italy, perhaps slightly greater in the North. Either way it's pretty clear since the time of the Roman empire Italians have not seen much gene flow from the outside.

  2. It's not minimal in some areas of northern Italy. I'm northeastern Italian and on genetic testing with three companies, I grouped with Austrians and not Italians. I think it depends on where you get genetic samples from. There are small inbred groups here and there in Europe. Italy has some inbreeding in the northeastern area or at least one study I read recently stated that. The Basque are between Spain and France, yet are an inbred group. Italy has smaller inbred groups in the north and that is a problem with these tests. Italian geneticists know this and are studying the bones of buried Lombards. They want to know who they mixed with and if modern day Italians are related. My family has been in the same area of Italy for eons, yet I'm not genetically close to Italians, but grouped right in the middle of Austria.

  3. I doubt any significant Celtic ancestry in Northern Italy. This idea of a Celtic Northern Italy contradicts both history and logic. These are the main arguments:

    1. There were just a few Celtic tribes settled in the Po valley. At the time of their invasion, the area was fully populated, it was not the Celts who populated it, they simply settled in the midst of native Italian populations: Etruscans (and also their relatives in the north, the Rhaetians), Ligurians, Venetians, Umbrians. The area was by no means a Celtic land.

    2. Even this ethnic composition, in which the Celts were a minority among natives, did not last for long. The Romans always resented this Celtic presence south of the Alps and wanted to expel them from Italy. They had already begun the colonization of the Po valley right before the war with Hannibal, when the colonies of Placentia (Piacenza) and Cremona were established. The war prevented future developments for a while, because the Romans had to defend against Hannibal and his Celtic allies. After the war, the Celts paid a heavy price for this alliance. The Romans continued to wage war on them, and most of the tribes were destroyed, enslaved or expelled. Their land was distributed to Roman and Italian colonists and many cities were founded. This influx of settlers made the region a densely populated one, prosperous and similar ethnically to the rest of Italy in a short time. Although still a province for a while and misleadingly called "Gallia" Cisalpina, it was finally incorporated into Italy by Augustus, as a natural recognition of its Roman/Italian nature.

    To conclude, Celtic ancestry of Northern Italians is just another fantasy of the Padanian Nordicists. They should replace their comic Celtic cross with a Roman aquila, which would represent them much better, but it's their business if they want to be ridiculous. The Celts contributed even less then the Germans to the Italian gene pool.

    In my opinion, there was much more contribution of the Italians (Romans) to the population of the neighbouring countries then the other way around. The Romans settled in large numbers in some of these regions. For example, Southern France (Gallia Narbonensis) was considered in Roman times almost like Italy itself.

    As stated above, the only significant difference between Northern and Southern Italy is the Greek contribution to the latter.

  4. First of all i do not consider it a stigmata to be (Genetic) South European; Especially not in the light of the reason studies (Lazaridis et al 2013 / Haak et al 2015 / Gamba et al 2014) that show that the so-called South_European component (peaking with overwhelming majority in Sardinians) is identical with Neolithic-farmers i.e. a Neolithic inheritance (of Ötzi and the likes); North Italians are no exception to this; South Europeans with an overwhelming Neolithic ancestry;

    Haak et al 2015 {supp. p.120}
    "Sardinians are the population that is closest to early European farmers with an estimated ~90% descent from them"

    North Italians have also been tested (Tian et al 2009 / Sikora et al 2014) to be the closest Italian group to the Sardinians; Regardless of what Geography might allude; And it is indeed the North Italians and Tuscans that cluster the closest to Greeks and Albanians, where as South Italians are between Greeks and Cypriots;

    As for the Spaniard comparisons; Yes it is true that Spaniards have a much higher North European affinity but at the same time also a high and significant African affinity which is absent in North Italy and Tuscany; Just one (recent) example - Kushniarevich et al 09/2015:

    Also your remark at '(EDIT 2015) "..all Europeans basically are composed of the same 3 mixtures, it's just that the ratios are different..." is incorrect; As already illustrated by Lazaridis et al 2013 and his 3way mixture, 4 populations are not fit for such an analysis due to and extra-external admixture; These were due to an extra Near East admixture Maltese, Sicilians and Ash. Jews and due to an extra African admixture Spaniards {supp. p.88/p.146} and indeed Haak et al 2015 created a 5way mixture {supp. p.125} adding Nganasan and BedouinB and the result was this:

    Spain: 18.5% BedouinB / 54.4% EN / 2.4% Nganasan / 0% WHG / 24.7 Yamnaya
    North Italy (Bergamo): 1.7% BedouinB / 74.4% EN / 3% Nganasan / 0% WHG / 20.9% Yamnaya

    Clearly showing that North Italians are proper South Europeans and have a higher ancient European ancestry than Spaniards; Keeping in mind that BedouinB is an ambiguous component that can be considered MENA; If Haak would have used a proper African component Spain would have scored as in Lazaridis {supp. p.88} 12.6% Mozabite and 1.5% Yoruba; Components which are absent in North Italy (Bergamo), Tuscany and other South Europeans acc. to the K=20 analysis in both Haak and Lazaridis;

    On that note, yes North Italians are proper South Europeans;

    1. "North Italians have also been tested (Tian et al 2009 / Sikora et al 2014) to be the closest Italian group to the Sardinians; Regardless of what Geography might allude; "

      North Italians and Sardinians cluster worlds apart. They may have similar amount of WHG ancestry, but so do many Balkanians and Iberians.

      Bergamo is about 40% Yamnaya+WHG, whereas Sardinians are only 10% (See Haak et al.)

      Moreover Bergamo samples are not really representative of all North Italians. Swiss Italian samples in Novembre et al 2008 are halfway between Germans and Bergamo, while Furlans are closer to Austrians than to Bergamo (See Tõnu Esko et al. 2012)

      "And it is indeed the North Italians and Tuscans that cluster the closest to Greeks and Albanians, where as South Italians are between Greeks and Cypriots;"

      Lazaridis et al had only Greek samples from the far North, and even few of them clustered near Cypriots and Turks, whereas no Sicilian did it. Greeks as whole are more Southern than S.Italians (see Paschou et al). Moreover Greeks from Peloponnese are among the most Northern Shifted ones (see Kushniarevich et al.) since they did not mix with Pontics and Anatolians.

      "If Haak would have used a proper African component Spain would have scored as in Lazaridis {supp. p.88} 12.6% Mozabite and 1.5% Yoruba; Components which are absent in North Italy (Bergamo),"

      Supp info p.85 has Spaniards scoring 14.8% African (12.6% Mozabite and 2.2% Mbuti/Yoruba). Other Southern Euros had zero.

    2. @Roma Sabinus (=Nobody) & Unknown:

      African DNA is absent in Northern Italy and in Tuscany only in your dreams & fantasies and your dishonest manipulations. First of all, you mysteriously "forgot" to mention that the Lazaridis et al. 2013 study questioned its own results regarding their Spanish sample and very plainly said this:

      "It is possibly that this is due to the presence of low levels of Sub-Saharan ancestry in the Mediterranean or of North African admixture as has been reported previously. Such ancestry has also been suggested to occur at low levels in other European populations, and perhaps the Spanish stand out in our analysis because of their large sample size."

      So the authors are recognizing that the larger size of the Spanish sample size might be exaggerating these supposed differences. We can see something similar happening in Busby et al. 2015, but this time it was not the Spanish who had the larger sample size, but your Tuscan heroes, and alas, not surprisingly it was them who came out with the largest amount of West African DNA (orange-brown) in the ADMIXTURE analysis as a result:

      As for North African DNA, Lopez-Herraez et al. 2009, Atzmon et al. 2010, Hodgson et al. 2014 and the already mentioned Busby et al. 2015 all found it with no problem at all among Sardinians, Tuscans and/or Northern Italians.

      Shriner et al. 2014, on the other hand, found no significant sub-Saharan or North African anywhere in their sampled Europeans populations except in Sardinia (6.1% Berber), plus they also found quite more Middle Eastern DNA among Northern Italians and Tuscans (6.9 to 9% Arabian, 17.2 to 21.1% Levantine-Caucasian) than among Spaniards (0 to 5% Arabian, 6.8 to 9.6% Levantine-Caucasian.)

      Hodgson et al. 2014 in the maps that accompany the results of their analysis at K=10 to K=12 also show way more Middle Eastern (Eurasian and Arabian) DNA in Italy than in Spain, while the only sub-Saharan DNA they put in Spain is some East African (Ethio-Somali) around Andalusia. As for North African DNA, this same study shows it quite widespread throughout most of Europe, including most of Italy.

      Regarding Haak et al. 2015: that "Maciamo" guy from Eupedia estimated the "Bedouin" component in the sample labelled "Spanish" as actually about 10%, not 18.5% as you claim. Plus you mysteriously "forget" to mention that what the Bergamo sample had quite more of was the "Caucaso-Gedrosian" component (greyish green), which peaks among people like Pakistanis and Indians:

      So on that note, no, Northern Italians are not more "proper South Europeans" than Spaniards.

    3. 1.) you misunderstood Lazaridisis' statement. Theat the Spaniards stand out because of a larger sample size says something about the sample size of other populations, but nothing at all about the result obtained from that large Spanish sample size. On the contrary the larger the sampe size the more representative and credible the result - and as Lazaridis pointed out, the amount of Spanish samples was very large. The result being 12.6% Mozabite (NW African) and 2.2% sub-Saharan (Black African)

      2.) Talking about "manipulations", from which ass did you pull that Busby et al admixture run image out of? It is not the same [K=14) run that is displayed in (p.3) of Busby et al 2015! At K=14 the larger Tuscan samples (23/70) all have 0% west-African and only the minorty of (3) and (4) samples have sub-Saharn admixture and also the other samples tsi(2)/tsi2a(2)/tsi2b(2) have no sub-saharn admixture - meaning oinly a minorty of Tuscans have sub-saharan admixture (and that at minimal amounts). North and South Italains both having none.
      Not to mention Figure4 (GLOBETROTTER’s admixture) to compare Spaniards with North Italains and Tuscans!

      However there are also other admixture runs such as Kushnirevich et al [K=6] or Behar et al [K=10] or Malick et al [K=7] in which almost ALL Spanish samples display sub-Saharan admixture only a minorty doesnt, and the amounts of sub-Saharan admixture in the Spanish individual is far higher than in any other individual of any other population in Europe (incl. the along tested North Italians, Tuscans, Abruzzo and Sardinians).

      3.) as for the rest of your Mambo-Jambo, in Lopez-Herraez et al. 2009, Atzmon et al. 2010, Hodgson et al. 2014 Spaniards were not even tested, and all the Europeans tested in Atzmon had North African admixture (incl. the Russians). There is also a fine difference between Tunisian_Berber and Mozabite or Saharawi; In Botigue et al 2013 in Lazaridis et al 2013 Spaniards score the highest in Mozabite/Saharawi where as Sardinians have Tunisian_Berber admxiture (Hodgson et al / Shriner et al). the difference is that Tunisian_Berber is unadmixed where as Saharawi/Mozabite is heavily sub-Saharan admxied. And both Hodgson et al [K=20] and Lopez-Herraez et al [K=6] show no sub-saharan in Tuscans, North Italians or Sardinians whereas Haak et al [K=20] shows the large Spanish samples riddled with Dinka and Yorzuba admixture (also here Bergamo and Tuscany have 0%) !

      4. That "Maciamo-guy" is neither an Academic nor a Scientist, thus i wikll stick to the academic data from Haak et al and this is ( p.125) 18.5% BedouinB in Spaniards. Lazaridis already hinted that Spaniards posses a substantial additional strain.

      PS: Nobody needs to guess who these samples are in EUR (Pardo-Seco et al) with the high amount of Near east and only ones with sub-Saharan admixture. i doubt it is the Finns (FIN), the Brits (GBR) or the Utah-whites (CEU) leaves only the Spaniards (IBS/SPA) and they are the only ones with sub-Saharan admxiture in Europe and a sizable Near Easta dmixture on top; Tuscans 0.0% sub-Saharan! My heroes

    4. 1)No, you are the one who misunderstands what they plainly say. They are pointing out that the Spanish sample due to its larger size might be giving the impression of a larger amount of the "African" components that they plainly say other studies have also found in other European populations but at lower levels:

      "It is possibly that this is due to the presence of low levels of Sub-Saharan ancestry in the Mediterranean or of North African admixture as has been reported previously. Such ancestry has also been suggested to OCCUR AT LOW LEVELS IN OTHER EUROPEAN POPULATIONS, and perhaps THE SPANISH STAND OUT IN OUR ANALYSIS BECAUSE OF THEIR LARGE SAMPLE SIZE."

      2)Are you blind? That is in fact just a close up view of K=14 ADMIXTURE analysis of Busby et al. 2015 of your very own link, and it very clearly shows that the Tuscan samples collectively had the largest amount of West African component (orange-brown) of any of the sampled European populations. It is likely not any "coincidence" that in this study it was in fact the Tuscans who were the most sampled Europeans. The "GLOBETROTTER" analysis is experimental and it is quite in contradiction to the standard ADMIXTURE analysis of their very own study, which is the method used by most genetic studies. Notice, for example, that the ADMIXTURE analysis shows no North African (yellow) component in their Sardinian samples, yet strangely enough the experimental GLOBETROTTER analysis shows Sardinians as the only Europeans with North African component! Talking about your real manipulations and lies: Behar et al. 2010 does not show any sub-Saharan African (dark & light brown on the chart) in Spaniards at K=10, or in fact at any other K above K=1:

    5. 3)The fact that Spaniards were not sampled in Lopez-Herraez et al. 2009 and Atzmon et al. 2010 has nothing to do with it, but Italians were indeed sampled, and they do show that Northern Italians and Tuscans have North African component in their DNA as well. Yes, and so do many other Europeans. That's the point. And the North Africans used for those studies were not Tunisian Berbers either, they were Mozabites:

      And Shriner et al. 2014, a study which sampled A LOT of African populations from all over that continent, did not make any distinctions among all their Berber samples, which also included Mozabites:

      See Supplementary Table 1. Mozabites are clearly there among all the North Africans sampled. Sardinians were the only sampled Europeans in that study who had a significant amount (6.1%) of the Berber component. So the Berber component found in the Sardinian sample is also at least partly derived from Mozabites.

      Hodgson et al. 2014 does show a small amount of sub-Saharan African (dark blue = West African, dark green = East African) among North Italians and Tuscans:

      Even though Spaniards (except Basques) were apparently not included in the ADMIXTURE analysis, the authors apparently relied on other data gathered somewhere else to make general assessments about the occurrence of these different components all over Europe, as seen in the maps that accompany the ADMIXTURE analysis. Notice that the only sub-Saharan component they placed in Spain was East African (dark green), not West African (dark blue), and only around Andalusia. In contrast, the Eurasian (orange) and Arabian (brown) component are practically all over Italy, while in Spain they are found only in a few parts (at K=12 there is in fact no Arabian in Spain, except in a corner of Andalusia close to Portugal.) Haak et al. 2015 only shows a very small amount (probably not even 1%) of East African in the Spanish sample.

      4)He is not a geneticist, but neither are you (or me, for that matter.) On page 124 (not 125) of the supplementary information of Haak et al. there seems to be shown two different figures based on the "BedouinB" sample, and not just for the Spanish, but for several other populations as well. The Tuscans themselves also show differences in value.

      PS: yes, you do need to assume it, just look at you expressing "doubt" about it as proof of it. If the samples had been clearly identified in the graph you would not need to "doubt" it, you could easily show it. But you can't, because the fact is that they did not clearly label any of the other Europeans samples on the graph, while the Tuscans are clearly labelled. We don't need to assume anything about the larger amount of Middle Eastern in them. The study is very clear about this, unlike the other thing.

    6. 1.) Obviously Lazaridis was shocked by the high amounts of African DNA in the (large) Spanish sample, and so was i. But the result remains and Lazaridis also concludes that "a fourth wave affected spain more than others".

      2.) No i am not blind. In fact i can zoom in on my "very own link" (the actual paper itself) 1000% on PDF. And there is no orange-brown on tsi(70) and tsi(23) even when you analyze each pixel. The abnormal Tuscans of tsi(3) and tsi(4) do show it. whereas North Italians and South Italains dont. And for your information i was referring to Behar et al 2013 not 10! there the K=10 is quite different than the K=14 of Busby concerning Spaniards and Sicilians.

      3.) Lopez-Herraez et al. 2009 and Atzmon et al. 2010 did not use a Mozabite component. Lopez-Herraez used the ambigiiuous [MENA] BedouinB and Atzomon used the Tunisian_Berber as in Quinto-Sanchez et al. But your in luck a recent study (Günther et al) did test NorthItalians, Tuscans and Spaniards with a Mozabte component and the K=15 run concluded that neither Tuscans nor NorthItalians had any Mozabite (not even a single sample had some) were as plenty of Spanish samples had Mozabite admixture and that a very subsatntial level ( p.71). In Hodgson et al Tunisian_Berber was the proxy for NorthAfrica and Georgians (a Caucasus population) were the proxy for Caucasus-Levantine in Shriner et al.

      And before you get confused with colors again you just might want to look at Table_S6 of Hodgson et al, Niger-Congo: 0.00% Tuscans(HapMap) / 0.00% NorthItalian / 0.01% Tuscans(HGDP). The 0.01 in TuscanHGDP is also found in BasquesHGDP and Orcadians (not to mention Russians).

      And what you describe as East African in Haak et al is not Ethio-Somalian (aka Hamitic) but Nilo-Saharan aka Dinka. Not to forget the Yoruba (i believe its coloring is bright Blue).

      4.) well now, if he is not a scientist than why is he distorting academic data? and why do you than highlight such nonsense?
      No, i am not a scientist either, and therefor i dont incline that BedouinB in Spaniards is 88.5%.

      PPS: Well, i guess we both know who those samples belong to and tehrefor none has to guess. In fact anyone who reads the entire paper will know.

    7. 1)Lazaridis et al. offered their own possible explanation for the unusual result, plus they in fact end up dismissing the result as "not significant".

      2)Yes, apparently you are, because anyone can plainly see the West African orange-brown component in the TSI70 sample by zooming in at 1000%. In fact, I spotted the orange-brown component in ALL Tuscan samples! The TSI70 sample also has a small amount of East African (reddish-brown) component as well, and so does TSI4. And the TSI3 and TSI4 samples have the largest West African of any European sample. Put all the Tuscan samples together and beyond any shadow of a doubt they top the list of West African found among Europeans in this ADMIXTURE analysis.

      Behar et al. 2013 shows a small "African" (brown) component in the Spanish sample up to about K=9/10. After that, it is gone.

      3)Lopez-Herraez et al. and Atzmon et al. both plainly used Mozabites, there is no argument here. Tunisians are not mentioned anywhere in either study. Just an example, from Lopez-Herraez et al., to even more plainly show that they used Mozabites:

      "The three Italian groups are alone among European groups in having low amounts of the Mozabite ancestry component, possibly indicating gene flow across the Mediterranean."

      I find it very strange how you desperately want to deny that these studies plainly used Mozabites for their North African samples. As if it wasn't already plainly enough seen from the graphs, clearly labelled as "MOZABITE", even the text of the studies themselves plainly point out that Mozabites were used as the North African samples.

      I took a look at Table S6 of Hodgson et al., and yes, I think I mistook the purple colors for the dark blue this time. What the Northern Italians show in the ADMIXTURE graph must be Nilo-Saharan, Ethiopic and Khoesan.

      The East African in Haak et al. is the dull pink component, peaks in Dinkas, Datogs and Masai. The Yoruba is light blue.

      4)Since one of the values given for the BedouinB component in the Spanish sample is like 12% I suppose he rounded it up to about 10%.

      PPS: No, we don't know for sure since the graph does not clearly label the other European samples.

    8. " So the Berber component found in the Sardinian sample is also at least partly derived from Mozabites "

      Eh, no.
      The proxy was not Mozabite but Tunisian_Berber. And Tunisian_Berber is the (purest) uadmixed Berber component (Henn et al / Botigue et al / Quinto-Sanchez et al) in contrast to the Mozabite and Saharawi compenents which are heavily sub-Saharan admixed. Of course the North African groups display a large Berber ancestry (plus others) but only if the Tunisian_Berber is the sole component but if one factors in also the Saharawi/Mozabite component - North Africans tend to be overwhelmingly Saharwi/Mozabite and not pure-Berber. (Henn et al K=8 / Botigue et al K=6).

      Where as Sardinians dont correspond to the Mozabite/Saharawi component (Günther et al) but only to the pure Tunisian_Berber component (Shriner et al/Paschou et al). Which is an old south mediterranean component. Spaniards on the otehr hand do not correspond to the pure-Berber component either but to the Mozabitre/Saharawi admixture instead (Botigue et al).

    9. Shriner et al. refers to "the Berber component" (dark golden), which is highest of all in Tunisians and Mozabites out of all the North African samples. Mozabites are predominantly Berber too. Shriner et al. did not support Botigue et al.'s claims of any supposed significant connection between Spaniards and North Africans, and if, as you say, they used people who are even more Berber than Mozabites as proxies, then no wonder that they have even more reason to disagree with Botigue et al.

    10. Also, according to Botigue et al. Tunisian Berbers are also one of the North African groups that has "the lowest proportion of the European cluster in ADMIXTURE", and they too tried to use Tunisians to supposedly make a connection with Iberians:

      "For example, the Andalucians share many IBD segments with the Tunisians (Fig. 3), who present extremely minimal levels of European ancestry."

      Again, no wonder that Shriner et al. did not support their claims. Their own samples of Tunisians and Mozabites did not show any such supposed connection. Only Sardinians had any significant (6.1%) North African component.

    11. This comment has been removed by the author.

    12. 2) Where do you see Orange-Brown in tsi(70)? There is a bright stripe that differs completely from the Orange-Brown stripe in Ceu(2). Yes, there is a Dark-Brown (EastAfrican) stripe in tsi(70) and in tsi(23) and the 7 samples of tsi(3) and tsi(4) will certainly top the WestAfrican of any other group in that admxture run - but not the 93 samples from tsi(23) and tsi(70) because they have none.

      3.) i have read the full Lopez-Herraez et al paper, and looked at the analysis chart of K=5 and Bedouin samples score 100% of that Mozabite component. also in Günther et al K=5 has a MENA component dominant in both BedouinB and Mozabite that later diverge into Mozabite and BedouinB proper. Thats what i meant with K=5 in Lopez-Herraez et al MENA. But here, point taken in Lopez-Herraez et al K=5 North Italians and Tuscans have "low amounts of the Mozabite ancestry component"

      And in Hodgson et al it can only be the Ethio-Somali, Nilo-Saharan and Khoisan that North Italians (and Tuscans) show in the K=12 run. But that is overall very minimal and at the exact same rate as all the other Europeans tested (Basques, CEU, French, Russians, Orcadians). Table_S6 in full detail.

      But Spaniards were not even tested in any of those studies.
      East African in Haak et al is Nilotic (Dinka) as is Nilotic (Anuak) the 'Nilo-Saharan' in Shriner et al.

      4.) i guess some call it "rounding up" others distorting. why he feels he has the credentials to "round up" academic data in the first place is beyond me.

      PPS: Oh, i am very certain that the both of us know for sure.

    13. 2)The orange-brown is very clearly visible in all the Tuscan samples. In TSI70 and TSI4 is right below the East African reddish-brown. Right next to TSI70 is the Turkish sample, TURKI4, showing the exact same pattern, except that the Turkish sample has more East African (reddish-brown) than the Tuscan sample, and about the same amount of West African (orange-brown.) Even leaving TS4 and TS3 aside, if you put all the Tuscan samples together they still would have more West African than any other European sample.

      CEU3 (not CEU2, which has no African at all) does not have orange-brown (West African) but it does have East African (reddish-brown.)

      3)I never said that Lopez-Herraez et al. found huge amounts of North African in North Italians and Tuscans, but that it DID find it. Atzmon et al. looks like it found a more substantial amount.

      Spaniards were not part of Lopez-Herraez and Atzmon et al., but Hodgson et al. must have used data on Spain and Portugal, otherwise it is difficult to envision why they have component distributions for these countries on the maps that accompany their ADMIXTURE analysis.

      One of the Tuscan samples in Hodgson et al. does have a small amount of West African as well.

      PPS: I think that neither one of us knows for sure. It is not clearly labelled on the graph.

    14. Spaniards were not part of Lopez-Herraez and Atzmon et al., but Hodgson et al. must have used data on Spain and Portugal, otherwise it is difficult to envision why they have component distributions for these countries on the maps that accompany their ADMIXTURE analysis.

      In the absence of samples from a population of an area visible on a map, that area is colored automatically based on the algorithms of the map painting software, not based on some prior knowledge on their genetics. So colorings of unsampled/unrepresented areas should not be taken seriously.

    15. Are you sure that it is the case in this study as well? From the general distribution of some of these components shown on the maps one would be very tempted to believe that they were done on purpose, not the product of some random decision by a map painting program. For example, just like the maps show quite more of the Middle Eastern components in Italy than in Iberia, which is the general consensus, it also shows more of the North African component in Iberia than in Italy, again reflecting the general consensus on the subject.

    16. Unsampled locations are as a rule automatically painted as on a cline between the closest sampled locations, not based on any prior knowledge. This is how map painting algorithms work. You can consult any map painting professional or the authors of that paper to get more info.

    17. 2.) thats exactly what i am talking about but i am looking at this.
      tsi70 - turki4 - ceu2

      It is the ceu2(2) that has a clear orange-brown stripe. And that stroipe is of adifferent coloring than the bright stripe in tsi70 (or turki40 and many others). And if tsi23 and tsi70 have none (orange-brown) than they can hardly have the most westafrican.

      3.) It does depend which North Afrcian component was used (Mozabite or Tunisian_Berber). And studies which have a clear Mozabite or Saharawi component like Behar et al, Henn et al or Günther et al show no admixture in Tuscans or North Italians, but in the tested Spaniards admixture was substantially present.

      In Hodgson et al the minimal amount of WestAfrican is the aformentioned (Table_S6) 0.01% in TuscanHGDP (0.00% in TuscanHapMap). That 0.01% is also found in Basques(HGDP), Orcadians and CEU.

      And i agree with Onur, spatial maps only mean something when a tested population is directly involved.

    18. 2)I simply followed your original advice of taking a look directly at the graph in the study itself, found here:

      Zooming in at least 1000%, and looking at K=14. And this is exactly what can be clearly seen by doing that:

      The orange-brown is clearly seen in all the Tuscan samples, including TSI70. Whatever is found under the reddish-brown East African component is the West African orange-brown component, the very first component in the graph. It is the one that shows up at the very bottom of any sample that has it. If you see the reddish-brown East African component in any given sample and there is something underneath it, it is the West African component. There is no other component that occupies a lower position on the graph than the West African. West African = 1, East African = 2, North African (yellow) = 3.

      3) Berber component is Berber component. It does not matter in what population it's found. Lopez-Herraez et al. and Atzmon et al. used Mozabites for their samples and had no problem finding the North African component among Northern Italians, Tuscans and Sardinians either. It is not necessary at all to use Tunisians for this to happen.

    19. Note: For those who want to do even deeper zooms into the graphic and see things even clearer, I have uploaded the PDF page of Busby et al. 2015 that contains the ADMIXTURE analysis, to make it easier to handle than the whole document:

      I recommend to zoom in at 3200% for fuller details. All the components, no matter how small, are very clearly discernible at such a close-up look.

    20. 2) The image you show is not a zoom-in on the admixture run from the actual paper. It even differs from the first image you posted which was also not from the actual paper. I posted a zoom-in of the run from the actual paper and there is no ornage-brown in tsi70. But apart from that, Busby et al is 1 admixture run that can be compared (double tested) to many otehrs. To name a few: Behar et al K=10, Malick et al K=7 or the more recnet Haak et al K=20 or Kushniarevich et al K=6, the results concerning Tuscans and Spaniards are completely different. In that only a few Tuscan samples have sub-Saharan and that at most minimal level whereas almost all Spanish samples have sub-Saharan (only a few dont) and that at levels higher than others. In Pardo-Seco et al 98 Tuscans were tested and all were o.oo% sub-Saharan (K=4) - same result as in Henn et al or Quinto-Sanchez et al. And you might also be interrested in the results of another recent study (Hofmanova et al nov.2015) Figure3, i think a zoom-in of 650% will do.

      3) Exactly this: "Berber component is Berber component".
      And the only unadmixed Berber component is Tunisian_Berber, Saharawi and Mozabite are admixed components. And papers such as Haak et al and Behar et al (2013) (who didnt use a Mozabite component) revealed that Mozabites are heavily sub-Saharan admixed and those results in Mozabites are similar to that of Atzmon et al. If Atzmon would have used a Mozabite component than that component would be near fixation in Mozabites as in this recent paper (Omrak et al 2016) K=12, and all otehrs that use a Mozabizte component.

      So whoever descended into sardinia, these must have been pure(unadmixed)Berbers for the modern Sardinians to still have pure Berber (6.1%) DNA and not Mozabites and Saharawi.

    21. 2)The image I posted is a screen-capture of a close up at 1000% of the ADMIXTURE graph straight from the paper itself. The picture I gave the link for the first time also came from the paper itself, but it was not zoomed in enough, it was made by one of the guys from Anthroscape. I also posted the PDF page containing the ADMIXTURE analysis straight from the study. Anyone can download it and see for themselves by zooming in. Or if they prefer they can download the whole study from the link above. The brown-orange West African component is very obviously there in TSI70 sample, and anyone who bothers to go 3200% zoom will more than very plainly see it. And they will also see that all the Tuscan samples had the orange-brown West African component. The image you posted was not close enough, but the fact that you can tell there is something below the East African component already proves that TSI70 does have West African as well. The West African component when found in any given population always occupies the lowest position in the graph, right below the East African (if this component was also found in the same population.) Since TSI70 also had the East African component and another component right underneath it, it goes without saying it also has the West African one as well. It is obvious that this ADMIXTURE analysis using 106 Tuscans found more West African among them than among any other Europeans.

      Behar et al. 2013 stops showing any sub-Saharan for Spaniards after about k=9/10.

      If you are referring to the Yoruba (blue) of the pie charts in Hofmanova et al. 2015, it is found quite widely spread over most of Europe, even as far north as Scotland. Notice also the pie chart for Poles, if quantities is what you are trying to imply. The majority of the Spanish groups had less than the Poles. In fact, only Galicians had about as much as the Poles. Also notice the Central and Northern Italian pie charts and the Northern Spanish charts. Only Galicians had more of the blue component. The other Northern Spanish actually had less than the Italians. In fact, the non-Galician Northern Spanish have a bit less than the French. The Italians have about the same level as the Andalusians and Extremadurans, but lower than the Valencians.

      3)Tunisians also have admixture from other groups. Less than Mozabites, according to some studies, but they still have it.

      In Shriner et al. and Hodgson et al. Mozabites in fact scored the highest Berber component after Tunisians.

      The graph you posted from Omrak et al. in fact shows that Mozabites have less of other components than Tunisians. They are almost wholly "yellow", while Tunisians have about 3/4 of the same component.

      A study that used both Mozabites and Tunisians, but as different components, was Kopelman et al. 2009, and some amount of both of them were found in Italians:

      Intermediate purple = Tunisians
      Dark purple = Mozabites

      Basques and Orcadians had smaller amounts of both than Italians and French.

    22. Clarification: the Tunisian component in Kopelman et al. 2009 was Jewish, so that likely explains why it is quite different than the Mozabite. Had it been Berber Tunisian it would be more similar to Mozabites. In any event, this is yet another study that found some North African (here rather mislabelled as "Middle Eastern") component in Italians as well by using Mozabites.

  5. "North Italians and Sardinians cluster worlds apart. They may have similar amount of WHG ancestry, but so do many Balkanians and Iberians."

    I agree, they aren't even close to being the same population. Sardininians represent an isolated Neolothic population which has only seen minimal geneflow since the first Neolithic settlers settled Sardinia. Sardinians are 90% EEF genetically, Northern Italians on the other hand are typical Southen Europeans after the Indo-Europeanization of Europe took place..and are more on the middle to northern genetic periphery of modern Southern Europe, and are only about 64% EEF.

    "Moreover Bergamo samples are not really representative of all North Italians. Swiss Italian samples in Novembre et al 2008 are halfway between Germans and Bergamo,"

    Swiss Italians don't reside in Northern Italy, so that's why I didn't mention them. Makes a lot of sense to me they would be noticeably more Northern shifted then Northern Italians residing in Lombardy

    "Greeks as whole are more Southern than S.Italians (see Paschou et al)."

    No they aren't, only Cypriots are are more "southern" than Southern Italians and Sicilians. Mainland Greeks genetically look like a population that is 3/4ths South Italian 1/4th Slovakian or Czech. Southern Italians from Naples southward are about equal to Cretans in their genetic affinity to Northern Europeans. Northern Greeks with no Pontian/Anatolian Greek admixture blow out Southern Italians in Northern affinity and are genetically closer to Southern Slavs or Northern Italians from as far north as Lombardy than to Southern Italians.

    "Lazaridis et al had only Greek samples from the far North, and even few of them clustered near Cypriots and Turks, whereas no Sicilian did it."

    If they show a pull to Anatolia they are obviously Pontians. Do you know where Haak's Greek samples come from? Because obviously the Greeks here are considerably more northern influenced than his Sicilian samples. Not that Sicilians are the only Southern Italians of course.

    1. "Southern Italians from Naples southward are about equal to Cretans in their genetic affinity to Northern Europeans."

      On Paschou et al. Cretans and Dodecanese are slightly more southern-shifted than Sicilians and surely othern Southern Italians.

    2. That's not true because mainland Greeks from Athens in Novembre et al. 2008 are about equal to to South Italians in their genetic affinity to Northern Europeans (PC1).

      In Paschou et al. 2014 Sicily, Pelopponese and Laconia plot together, while Dodecanese, Cretans and Cappadocia are further south.

      In Alena Kushniarevich et al. 2015, all mainland Greeks are less Northern Euro (blue admixture) than either Bergamo or Tuscany.

    3. No Northern/Central/East Europeans featured in that Paschou et al plot, so it's pretty irrelevant since plots are dataset dependent..

      In Alena Kushniarevich et al. 2015, all mainland Greeks are less Northern Euro (blue admixture) than either Bergamo or Tuscany

      I see them roughly equal with Tuscany, Peloponese Greeks have about as much dark blue as Tuscany.

      Interesting that the Greek Macedonians show a lot of diversity. Have a feeling a lot of the ones scoring low in dark blue [by Greek standards] are Pontians. Looks like several of those Greek Macedonians score as high as North Italians. The Northern genetic character of Northern Italians is so over rated, lol. They score slightly less than FYROMians and Bulgarians, lol

    4. "No Northern/Central/East Europeans featured in that Paschou et al plot, so it's pretty irrelevant since plots are dataset dependent.."

      There are still the Basques who have plenty of Mesolitich Euro ancestry, actually more than most Northern Europeans. so it's correct.

      "I see them roughly equal with Tuscany, Peloponese Greeks have about as much dark blue as Tuscany."

      You need to buy a pair of glasses because all Greeks have clearly less blue admixture and more Dark Green and light blue. They are clearly less Northern.

      That's also confirmed in Haak et al, who used the same Northernmost Greeks from Lazaridis, and still get less WHG+Yamnaya than Tuscany and Bergamo.

    5. Basques definitely don't have as much Mesolithic ancestry as Northern Europeans. They score higher in EEF than the French in Haak, much less Northern Europeans. So putting them as any kind of surrogate for North Europeaness is some serious reaching.

      What i said was correct, I don't need glasses

      "That's also confirmed in Haak et al, who used the same Northernmost Greeks from Lazaridis, and still get less WHG+Yamnaya than Tuscany and Bergamo."

      Wrong again, Tuscany came more EN (Early Neolithic) than Greeks. Greeks scored 67%, Tuscans scored 72%. I have the data posted right here on this blog in this very entry man. Maybe try and look at it?

      Not to mention that while Greeks scored 20% Yamnaya compared to 27% for Tuscans. 'Yamnaya' itself is only half hunter gatherer, half the ancestry of the Yamnaya came from some type of Near Eastern source in the Caucasus. Which means that's only roughly a 3.5-4% increase in Mesolithic/Hunter Gatherer ancestry for Tuscans from scoring more in the Yamnaya category.

      On the other hand, Greeks scored 12% WHG while Tuscans scored 0%. And WHG in Haak was based on the complete Mesolithic genome of Loschbour man who didn't have a drop of near eastern admixture unlike the Yamnaya mutts.

      So I've really seen no evidence, especially in more recent studies that mainland Greeks are more 'southern' or 'Neolithic' than Tuscans or Central Italians as a whole. I'm confident the opposite is true when considering Northern Greeks. Though differences are probably trivial. It actually appears that Northern Italians only slightly beat Northern Greek mainlanders out. The early genome studies from 2008-2009 (like Tian et al) which showed high Greek-Southern Italian overlap appear to be wrong. Either that or there were a lot of Islander Greeks or Cretans sampled, which is probably the most logical conclusion to draw up based on the more recent data I've seen. Virtually all the plots I see from guys like Eurogenes too show mainlander Greeks always considerably more Northern shifted than Southern Italians.

      Lastly, what page in Haak is the information given where the Greek samples come from?

    6. "Basques definitely don't have as much Mesolithic ancestry as Northern Europeans. They score higher in EEF than the French in Haak, much less Northern Europeans. So putting them as any kind of surrogate for North Europeaness is some serious reaching."

      EEF is a mixture of Neolitich middle eastern and WHG, so it's actually hiding a good deal of mesolitich ancestry in it. If the WHG in EEF is separated from the ENF, the Basque would get more mesolitich ancestry than most North Europeans.

      In any case we can agree that the Basque are much more mesolitich than other South Europeans, while the Cappadocian Greeks are completely middle eastern. so the plot by Paschou et al, where the Pelopponese and Lakonia Greeks plot with Sicilians halfway between Basques and Cappadocians, is quite indicative.

      "Wrong again, Tuscany came more EN (Early Neolithic) than Greeks. Greeks scored 67%, Tuscans scored 72%. I have the data posted right here on this blog in this very entry man. Maybe try and look at it?

      Not to mention that while Greeks scored 20% Yamnaya compared to 27% for Tuscans. 'Yamnaya' itself is only half hunter gatherer, half the ancestry of the Yamnaya came from some type of Near Eastern source in the Caucasus. Which means that's only roughly a 3.5-4% increase in Mesolithic/Hunter Gatherer ancestry for Tuscans from scoring more in the Yamnaya category.

      On the other hand, Greeks scored 12% WHG while Tuscans scored 0%. And WHG in Haak was based on the complete Mesolithic genome of Loschbour man who didn't have a drop of near eastern admixture unlike the Yamnaya mutts."

      Where are you getting those numbers from? In Lazaridis et al, Tuscans get 0.151 of WHG, while Northernmost Greeks from Macedonia get only 0.58. Bergamo is at 0.177. Quite a big difference to me.

      "So I've really seen no evidence, especially in more recent studies that mainland Greeks are more 'southern' or 'Neolithic' than Tuscans or Central Italians as a whole. I'm confident the opposite is true when considering Northern Greeks. Though differences are probably trivial. It actually appears that Northern Italians only slightly beat Northern Greek mainlanders out. The early genome studies from 2008-2009 (like Tian et al) which showed high Greek-Southern Italian overlap appear to be wrong. Either that or there were a lot of Islander Greeks or Cretans sampled, which is probably the most logical conclusion to draw up based on the more recent data I've seen. Virtually all the plots I see from guys like Eurogenes too show mainlander Greeks always considerably more Northern shifted than Southern Italians."

      Tian et al used Greek Americans and Italian Americans. Their Greeks samples plot exactly where mainland Greeks from Athens do in Novembre et al. 2008 compared to Italians.

      "Lastly, what page in Haak is the information given where the Greek samples come from?"

      Wolfgang Haak and Iosif Lazaridis work in the same laboratory of David Reich. They obviouly use the same samples.

    7. "EEF is a mixture of Neolitich middle eastern and WHG, so it's actually hiding a good deal of mesolitich ancestry in it. If the WHG in EEF is separated from the ENF, the Basque would get more mesolitich ancestry than most North Europeans."

      We now know that EEF is 90% Anatolian farmer, 10% WHG. So if all the populations in Haak were modeled based on Anatolian farmers and not EEFs, their 'Neolithic' ancestry would be slightly less. But still EEF is by far the most Neolithic/non European Mesolithic component in Haak's 3 way admixture run. Yamnaya is 50% EHG, 50% Near Eastern Caucasus, And of course WHG is 100% WHG.

      "Where are you getting those numbers from? In Lazaridis et al, Tuscans get 0.151 of WHG, while Northernmost Greeks from Macedonia get only 0.58. Bergamo is at 0.177. Quite a big difference to me."

      I'm getting them from Haak, in his 3 way admixture run (WHG, EEF & Yamnaya) as i explained in my blogpost.

      In the PDF, it's on page 153 of 172..figure C. In fact even going by figure A which replaces Yamnaya with Eastern Hunter Gatherer, the Greeks still come out less Neolithic than Tuscans.

      As for Lazaridis, his admixture run was a 3 way run on modern populations using WHG, EEF & ANE. His article came out before we had Yamnaya genomes or Eastern Hunter Gatherers sequenced.

      Ancient North Eurasian is based on a genome from 20,000 BC Siberian populations. Any admixture run based on genomes so old, and who's ancestry is spread so far throughout the world (including Native Americans and South Asians) is bound to be less informative than one where it replaces ANE with Yamnaya, who are only from 3,500-2,500 BC, lived on the Eurasian steppe, and are the direct ancestors of all Europeans both genetically and likely linguistically. Or even Eastern Hunter Gatherers who made up 50% of the ancestry of the Yamnaya, and are from 5,500 BC or so. Also your initial comment on my blog was based on WHG + Yamnaya ratios, not WHG + ANE ratios.

      _"Tian et al used Greek Americans and Italian Americans. Their Greeks samples plot exactly where mainland Greeks from Athens do in Novembre et al. 2008 compared to Italians."

      Thanks for that Novembre plot. You know the Italian samples in that plot range from everything from Swiss Italians to Sardinians, as well as the entire peninsula. I think we can easily guess which of those Italians are Sardinians. But anyway I notice only some Greeks overlapping with the southern end of the Italian cluster, others are more Northern and Balkan shifted, closer to Bulgarians. Difficult to tell about the rest though [which are Tuscans, Northern Italians, etc]

    8. I have to say that i am quite disgusted by your filthy attitude;
      What is wrong with Macedonians ("FYR") and Bulgarians? and calling Yamnaya folks as mutts?really?

      Degenerated types such as Jennifer Aniston (aka Anastasakis) and her fake plastic-Nordic wanna-be look or madame "i am too dumb to speak proper english" Huffington; Another Anglo/Nordic Greek wanna-be joke; You are prob. just like them;
      Modern Greece LOL

    9. This wont be my last word on the filthy Greeks (modern not ancient / big difference) now that Sardinians (shepherd type), Albanians and north Italians (also Bulgarians) are the real archaic Europeans (mediterranid+alpinoid variants) who needs the dirty modern Greeks, a joke to begin with.
      Maybe i'll also start a blog, seems like fun;

    10. Yes, no one short of 55 Beers will mistake any common Greek as North European no matter how butthurt they are or how hard the likes of Jenn Aniston try (and they try very hard). Truly sad.

    11. 1. I don't have a problem with Southern Slavs, and absolutely nothing with Bulgarians. FYROMian nationalists that identify with Ancient Macedon are a bunch of goofball clowns but that's another topic.

      2. The Yamnaya / Proto Indo-European folks were indeed mixed mutts, the product of Eastern Hunter Gatherers mixing with a Neolithic Caucasus population from the Near East that migrated north into the steppe. Although I admit 'mixed mutts' was just a snide comment I made mocking outdated Nordicists worldviews. I am interested in the Proto Indo-Europeans and consider them my ancestors. The old Nordicist worlview, mostly from the 19th and early 20th century basically postulated the Proto Indo-Europeans were some sort of pure Hyperboreans, blonde haired and blue eyed Ubermen from the far North who swept down and civilized the South. We now know this is a quite false and horribly exaggerated view of history. We now know that it was the South that migrated into the North first to help form Proto Indo-European genetic identity and culture itself. Only then later, this newly created Indo-European language and culture expanded in all directions merging with and influencing pre Indo-European cultures. This is not taking away of course the 50% Mesolithic Eastern Hunter Gatherer ancestral component of the PIEs which I'm sure played a very vital role as well.

      3. Haha, yeah. As far as Greeks are concerned, Greeks both modern and I'm sure ancient are happy to be Greeks and look Greek and not much like Northern Europeans. I know I am. I think Jennifer Aniston is pretty, but I prefer my Greek women looking more like Helena Paparizou. Helena fits more the common and proper Hellenic example/ideal of beauty. Though I doubt Jennifer would raise any eyebrows if she was living in Greece speaking proper Greek.

  6. In the real world people don't care about WHG, ANE or PCA plots, you'll be judged for your phenotype...

    1. Yeah, and while Northern Italians don't look like Sicilians, they sure as hell don't look like Norwegians either. No one short of about 40 beers would confuse Berlusconi for a Norseman.

    2. Of course Norwegians are too far off, they clearly stick out in Italy.
      However you seem to be very excited at the idea that the spanish cluster closer to northern europeans than northern italians and this is the comparision I had in mind: as someone who have experience with both groups I don't think there is any doubt on who have more fair people( altought they still aren't a majority) and who have much more dark and ambiguos people; and like it or not this is the criteria used by real life people to determine who is "more european"( which is a retarded concept I know)
      Actually even the average of both places is quite different with the spanish clearly closer to the mediterranean stereotype; denying it would be disillusional.

    3. I'm not "excited" because I'm not Northern Italian or Spanish, i don't have much ideological bout in their little "battle for Whiteness". I'm an American and ethnically a mix of 50% Greek, 25% Southern Italian + French, and 25% Northwest European. And genetically, I come out more 'Northern' than any 'pure' Southern Europeans from Europe on 23andme including Northern Italians.

      Example, my European PCA plot.

      Not that I give much a crap. I don't think it's terrible to plot far south on a European genome wide PCA plot, and I don't think it's a terrible thing to look more like Al Pacino or John Stamos than Max Von Sydow. *_^

      Anyway, Northern Spanish are definitely just as fair as Northern Italians IMO. There's no way they belong more in the 'Mediterranean stereotype' than Northern Italians other than length of skull [since the Spanish are noticeably more dolichocephalic on average than Northern Italians]. But I'm not going to pretend I've visited either Northern Italy nor Spain/Northern Spain. Just what I see from photos of individuals and crowds on race fora and stuff. I'd certainly agree with you for the Southern regions of Spain in comparison to Northern Italy though. But that makes sense since the climate is warmer and more Mediterranean-like there.

    4. Northern Iberians make less than 10% of Iberian population. Most Iberians live either near the southern coast or in the big internal cities. Half of population in Catalonia and Pais Vasco come from the South.

      From the paper “DNA-based eye colour prediction across Europe with the IrisPlex system.” South Western Venetians from Verona have 41% of light and mixed eyes, compared to just 15% of Northern Greeks from Thessaloniki and 22% of South Eastern Spaniards from Alicante/Valencia.

      Those Spaniards are lightest of the bunch. Much more representative ones from Madrid are as dark as Greeks, which just 15% of light and mixed eyes. From Gender is a major factor explaining discrepancies in eye colour prediction based on HERC2/OCA2 genotype and the IrisPlex model (2013).

      Of course a comparison with Northern and Eastern Venetians or even Furlans who are very mixed with Germans and Slovenes, and have more than 50% of light and mixed eyes would blow out any Greek in Northern affinity.

    5. "South Western Venetians from Verona have 41% of light and mixed eyes, compared to 22% of South Eastern Spaniards from Alicante/Valencia."

      That's cherry picking. Compare Northern Italians from say Emilia-Romagna to Northeastern Spaniards from Catalonia. Id bet the Spaniards would come out fairer.

    6. Cherrypicking my ass. Verona is almost in Emilia Romagna and Spaniards from Alicante are actually lighter than the ones living North of them. See the second study I've quoted. Anyway Barcellona is a melting pot of Anadalusians, Canarians... and many of swarthiest famous Spaniards like Sergio Busquets are from there.

      Needlees to say North East Italians from Friul and Bolzano would have at least 50% or even more of light/mixed eyes almost on par with Slovenes.

    7. You can't mix data from different studies unless they're using the exact same method and unit of measurement. Especially when talking of pigmentation, since what qualifies as "light mixed" or even blue are subjective cut off points. Welcome to color terminology and common sense 101. This is why pigmentation maps you see spammed all over race fora always come out a bit different in various studies showing different frequencies. It's one thing to suggest Northern Italians are lighter than Northern Spanish, but It's laughable to suggest Southern Spaniards are lighter than actual Northern Spanish Catalonians. That right there tells me you trying to mix and match data from two different studies is severely flawed.

      Does your first study, which is peer reviewed, and gives figures 22% figure for Valencia and 41% for Venetia have figures for Catalonians or Italians from Emilia-Romagna? If not then you have no proof Emilians are fairer than Catalonians in eye pigmentation. This does not even address hair or skin either.

      "Verona is almost in Emilia Romagna."

      The eye color pigmentation differences between Venetians and Emilians is shown right here in this blog entry. The data from Livi you see in for instance (which is based on thousands of samples from each region), while North Italians Venetians were 41% light eyed, North Italians from Emilia were only 32%. Emilians are without question darker than Venetians in eye color frequencies and according to Livi by nearly a margin of 10%. So yeah, It's disingenuous to consider them the exact same population, or that Veneto represents all of Northern Italy.

      Anyway, relooking at a geographical map of Europe, it appears Northern Spain is in the same general latitude as only Central Italy, not Northern Italy. Considering pigmentation is highly correlative with adaption to geographical climatic clines and not just from long distant genetic ancestry [which Northern Spanish undoubtedly show more affinity with Northern Europeans than the vast majority of Northern Italians] then it's definitely possible now rethinking it that Northern Italians are lighter in the broadest sense than Northern Spanish. But it's likely to be pretty small and trivial. But yeah, comparing Southeastern Spaniards to Venetian Northern Italians is definitely cherry picking, especially since my remark to the 'Nobody' guy I already said that Southern Spanish were undoubtedly darker than Northern Italians. So you were just repeating info which all 3 of us already agreed with [and then tried to mix and match this data with that of a completely different study!]. Thumbs up buddy.

    8. You are either trolling or just can't understand simple things.

      1. Verona is not the whole Venetia and is almost in Emilia Romagna. Get a fucking map and see it. Lightest areas of Venetia are Belluno, Trento, and parts of Alto Adige. Furlans are even lighter than the lightest Venetians, especially Eastern ones.

      2. In this actual peer reviewed study

      "Gender is a major factor explaining discrepancies in eye colour prediction based on HERC2/OCA2 genotype and the IrisPlex model (2013). "

      Spaniards from Alicante/Valencia are much lighter than the ones living North of them. THAT'S A SCIENTIFICT FACT! Either you like it or not. If you have some problems with the actual study then contact the authors.

      3. Barcellona and most of Catalonia have been invaded by millions of Southern Spaniards and even Canarians. So your idea that modern "Catalans" are lighter than Emilia Romagna is simply ludicrous. It would be using 50% North African Marseille to represent Southern France.

    9. @ ArchHades
      You do know that all these results (studies quoted by Silver) are based on Genetic SNPs; These SNPs can only be INHERITED and thus cant be corrupted by external factors (sunlight, warmer climate etc.); Its like testing Bloodgroups or Haplogroups;

      Lets take a look at data from the 1000-Genome-project:
      IBS (IberianSpain) / TSI (Tuscany) / GBR (Britain) / CEU (Utah-whites)

      rs16891982 G/G
      IBS = 66.4% / TSI = 94.4% / GBR = 95.6% / CEU = 96.0%
      rs12913832 G/G
      IBS = 9.3% / TSI = 15.9% / GBR = 68.1% / CEU = 57.6%

      Seems that Spaniards are much darker than even Central Italians; despite being more northern! Yet they are also very African compared to Central Italians (Pardo-Seco et al 2014 = 0.0% Tuscany/north Italy) maybe thats the trick;

    10. Iberians are hopeless. They can't get over the fact they have far less of those alleles associated with light pigmentation than 98% of Europeas. At this point they usually either claim to have 850 zillions of mesolitch admixture from La Brana HGs or worse quote ludicrous papers like Candille et al or Jablonski et al who claim that Indians are lighter than Japanese or Portuguese paler than Poles.

    11. "Spaniards from Alicante/Valencia are much lighter than the ones living North of them. THAT'S A SCIENTIFICT FACT! Either you like it or not. If you have some problems with the actual study then contact the authors."

      Why are the figures for Alicante/Valencia (blue + light mixed?) suddenly up to 31% now? You give me a figure of 22% from an earlier study, now your new study shows the combined frequency of 31%. So what's going on man? Could it be you're making the noob mistake I warned you not to make? But yes, the study does show the Spanish 'north of them' are just barely north and are at 26%. I would not consider that 'much lighter', considering a 5% difference is 1 extra light eyed person in a room of 20. Lastly this does not even include Catalonians. I don't know what population reference B is, but it doesn't seem to be Catalonia.

      "Barcellona and most of Catalonia have been invaded by millions of Southern Spaniards"

      And many Northern Italian major cities has been invaded by millions of Southern Italians, what your point? I'm not talking about the recent arrivals.

      "Verona is not the whole Venetia and is almost in Emilia Romagna. Get a fucking map and see it."

      So why not just get some samples from Emilia Romagna then? lol. My challenge of Emilia [N Italy] vs Catalonia [N Spain] is still completely unanswered by you.

    12. You could simply take a look at Hoyos Sainz who studied over 3000 Spaniards from all the regions and found that Catalonians are among the darkest Spaniards. Indeed by using the same scale, Spaniards end up below Southern Italians regarding the frequency of light eyes.

      Perhaps only Pyrennaic regions (less than 10% of Spanish population) can be compared with central and northern Italy.

      Anyway less than one million of South Italians live in the north.

    13. @Silver: Actually the Spanish population concentrates around the North, the East coast and the South, the center being the areas of less population density (except the whole Madrid Community):

      And Spaniards from Alicante/Valencia are hardly the lightest eyed people in Spain. I am not talking about those silly "predictions" that you like so much (which if taken at face value would even suggest that Italians are almost as light-eyed as Orcadians!) but the results of actual pigmentation surveys. The lightest eyed regions of Spain are in the Northeast.

      So no, it is Italians who are hopeless. You want to hilariously disparage studies like Jablonski & Chaplin 1999 and Candille et al. 2012 because of your rather strange and very selective interpretation of the actual measured data they present, yet you want to desperately believe in "predictions" based on some alleles that would suggest such absurdities as Orcadians and Andalusians being darker than Moroccans and Druze, or Germans from Lower Saxony being darker than Tuscans.

    14. @Roma Sabinus (a.k.a. Nobody): No, I am pretty sure that that rather disingenuous "explanation" doesn't hold water. By taking the results of other studies using frequencies of the exact same rs16891982 marker, I can also come to the very dubious suggestion that Germans from Lower Saxony (who ironically are one of the lightest people in that country) are darker than a bunch of other Europeans, including Spaniards, Greeks and Italians:

      Norwegians: 98%

      Poles: 98%

      Finns: 96%

      North Italians from Verona: 94%

      Spaniards from Madrid: 92%

      Greeks from Thessaloniki: 86%

      Spaniards from all over Spain (including islanders): 82%

      Germans from Lower Saxony: 66%

      Source: ALFRED alleles and 1000genomes.

      What you are carefully trying to conceal is that many people can carry those markers, and yet at the same time still not manifest the lighter pigmentation that they allegedly convey. That's why such "predictions" can be very misleading, unless they are accompanied by an actual observation and measurement of the pigmentation. Very few studies have tried to do this, and you already know very well the results of one of them (Candille et al. 2012 The "predictions" based on alleles often did not match the actual observed values.

      And the reason why anthropologists usually measure skin tones in areas unexposed to sunlight (armpits, pelvic area) is precisely to avoid the issue of the effects of UV rays on the skin.

      Plus Pardo-Seco et al. 2014 did not clearly label the Spanish sample in the analysis chart, so you seem to be just assuming the "African" component has to be all for it. What it does clearly show, though, is that Tuscans have more Middle Eastern DNA than any of the other Europeans sampled, and since they did clearly separate and label the Tuscan sample from the rest of the European samples we do not have make any assumptions here.

    15. The reason the Candille et al UV reflection data does not correspond to the Candille et al Genetic SNP data as well as the Genetic SNP data from Lucotte et al or Norton et al (and the list goes on) is that UV reflection method is corruptable by UV radiation aka sun exposure/tanning. And none of the groups in Candille et al were study groups - making the results none academic to begin with and dubious if the Polish turn out darker than the Portuguese. You mention Chaplin and Jablonski? well not even Jablonski believed in her UV reflection dat hence she created a formula in which the UV reflection data is factore in - i present the result:

      And where did you obtain the Low Saxony data? In Lucotte et al Westphalians and Bavarians are at 96% derived rs16891982 G/G where as Spaniards and Portuguese linger around North African dimensions.

      And you are correct North Italians are Brachycephalic and belong to the Dinaroid and Alpinoid sub-races; But there is heavy Mediterranid strain. These are two Italains one of the minorty Noric type and the other of the dominant Medit-Alpinoid type.

    16. Nope, that excuse doesn't work, and I already explained to you why: Candille et al followed the very standard procedure of taking the skin tone measurements from parts of the body not exposed to UV rays. So did all the data in Jablonski and Chaplin. In fact, so do the vast majority of anthropologists, precisely to avoid the possible problems that exposure to UV rays can present.

      Candille et al. gathered and used its own samples, nothing wrong with it. A bunch of studies use their own samples. It is not necessary to borrow them from some database or something of that nature.

      Jablonski and Chaplin did NOT make that map. Some anonymous person did pretending to be following their work. You can see what their "predictions" based on environmental factors are really like here, page 76:

      Quite different! Plus look at the actual measured values from many populations (including two Northern Spanish ones) in pages 74-75. Once again, a pretty different story! "Predictions" based on environmental factors not infrequently do not go hand in hand with measured values.

      The Lower Saxony results are gathered in the ALFRED alleles database (made by Yale University and the US National Science Foundation), which has compiled a lot of data about frequencies of some of these alleles from many genetic studies. You will see how many different results for a whole bunch of populations there are recorded there. Sometimes they score low, sometimes high, and so on. It is simply impossible to derive very solid conclusions from such frequencies. They often vary too much within different samples of the same populations, plus they also often contradict measured pigmentation values.

      The Mediterranean strain in North Italy is a minority and more concentrated in Northwest Italy (around Liguria.) The darker types of Northern Italians more often have Alpine and Dinaric traits, and no one should be surprised at all about it, since these are the two predominant craniofacial types there.

    17. This comment has been removed by the author.

    18. Candille chose the inner upper arm to "avoid as much as possible" the exposure of sunlight. But corruption of UV radiation (especially when its not a study group) can never be ruled out. It can though by Genetics, and those SNPs are uncorruptable. And predictions are not based on thin air but on those inherited SNP derived mutations, in other words the biological reality (and that is quite substantial).

      Blue and Brown eyes can be predicted with a 95% accuracy rate (8plex/IrisPlex) and the basal skin tone with a 99% accurracy rate (8plex). Based on numerous testing groups. And so far the Anthroplological data seems to correspond almost to the point with the Genetic data concerning eye-color. Though skin pigmentation was never a significant point in any Anthropological work (some have data based on the von Luschen scale). Who knows what low Saxon smaples are in the ALFRED database (im not familiar with it). So far none of the groups tested in published peer reviewed papers had an odd result.

      The North west of Italy might have pure Mediterranid types (Ripley) but the Medit strain is all over North Italy (Banse) and that quite extensivly. I think it was Kraitschek who differed the Alpinoids north of the Alps and south of the Alps with former being more Nordid and latter more Medit.

    19. Most people do not get sun exposure in the armpits or in the pelvic area, therefore measurements from these areas are the most accurate to assess skin tones. Frequencies of SNPs can be very misleading because a person can carry them and yet still not manifest the light pigmentation attributed to them. It is not any guarantee regarding the visible pigmentation. The only way to assess this for sure is to actually measure it.

      I am not aware of any supposed "99% accuracy rate" in predictions of skin tones by using any alleles. The very few studies that have bothered to contrast measured values with "predictions", like Candille et al., do not seem very encouraging.

      The data compiled in the ALFRED database come from genetic studies. And there's a wide variety of different results to be seen.

      Trying to blame Mediterraneans for the darker types of Northern Italians is not really convincing, since we can also easily find "swarthy" types in other areas of Europe where the Alpine and Dinaric also predominate. Even in places like Austria, Switzerland, Belgium, etc. one can see anthropologists easily noticing "swarthy" types.

    20. It can only be misleading if the Alleles are heterozygous. Than mostly given as inconclusive or undefined. 8plex can only predict the basal skin tone by homozygosity and that with an accuracy rate of over 99% (n=600) and an accuracy rate of a 100% (n=131). The test groups also revealed that homozygosity is the norm (74% and 61% respectively).

      It is an Anthropological fact that Mediterranid and Armenoid(>Dinarid) are darker pigmented than Nordid. And C.S. Coon derived the Alpinoid strain in Southern Italy as the lightening factor. And im not blaming anything at all and never said that the Mdeiterranid is the swarthiest or anything close to that.

    21. The reason for an apparent increase in accuracy is very likely due to using larger numbers of SNPs. Single ones are very misleading. If we had to believe in such single frequencies, like this one, for example, we would have to come to the counterintuitive conclusion that Orcadians are darker than Moroccans, Palestinians and Druze:

      Darker than Nordics yes, but than Alpines I find it hard to believe. The Italian anthropologist Biasutti characterized Mediterraneans as "skin color “matte”-white or brunet-white, chestnut or dark chestnut eyes and hair", and Alpines as "opaque-white skin color, chestnut hair and eyes".

  7. I didn't intend to be provocative.
    I 'll start with the premise that I'm not trying to portray italians as nordics and spaniards as arabs, but I have been in both place and I know for sure there are differences : while northern spanish are indeed a bit lighter than their countrymen they are still closer to them than to n.italians: coloring wise at best they could range from roman/ abruzzese like to tuscan like, coherently with their geographical position; but I remember seeing a study from Beddoe where only basques, navarrans and aragonese were clearly lighter than the spanish average, catalans and galicians weren't.
    The differences to me is that there people of the same typology of Villa, Fabregas or Andrea Duro are typical everywhere in Iberia while in Northern Italy,among the natives( they really wouldn't be at home(they could pass as southerners tough) .
    You can also find some swarthy/ very med n. Italians but definitely not in the same proportion and to a certain limit : even in Catalunya you have some border line phenos like Guardiola or Sergio Dalma and such people are outside the range of most of Italy.
    At the same time a good proportion of N. italians would look foreign in Spain, not because they are more european or something, but because they are different and would rather pass better in other places.
    These are my observations directly on the place.

    1. Take a look at Adams et al 2008 (Figure4);
      North Atlantic Spaniards are like Atlantic Morrocans;

    2. That's actually amusing, considering how easy it also is to find people like Simone Vergassola, Gianluca Zambrotta or Chiara Caselli in Northern Italy. And you seriously expect us to believe that Andrea Duro, David Villa (who looks a lot like Frenchman Jean-Pascal Lacoste, BTW) and even Fabregas would not fit in? Somehow I think that if we put them in the middle of these Lega Nord people no one would notice a thing:

      In fact, I would rather say that quite a number of them are actually more "exotic" than Villa or Duro.

      And by the way, none of the Spaniards you mentioned is exactly "typical" either. Common, yes, but not the typical Spaniards by any means. Think more along the lines of an Iker Casillas as a more average Spaniard.

      Apparently you also "misremember" seeing the Jablonski & Chaplin study, where the samples from Leon (Northwestern Spain) were almost as light as Basques and actually somewhat lighter than Belgians and South English. Anyone who has been to the area will tell you that no one can tell average Leonese apart from average Asturians, Galicians or Cantabrians just by looking at them. Same with Catalans and Aragonese. So I very much doubt that any of these Northern Spaniards are really much darker than others.

      The very swarthy types among Northern Italians are not predominantly "Med" either, but in fact predominantly Dinaric and Alpine, as befits the most common anthropometric racial types in the region.

    3. If that supposedly shows that, then Italy in general looks more like the Near East:

    4. This comment has been removed by the author.

    5. Oh now i get it, im supposed to be 'Nobody'? LOL
      archhades can surly help you out with the IPs, bet we dont even live in the same country.

      and i was talking about this

      something different than an outdated Y-Chromosome chart which looks "near east" if you didnt know by now that J2a is Neolithic European (Szécsényi-Nagy 2015).

      And this Medit.-Alpinoid

      is surly not swarthier than this

    6. Yes, I can easily tell you are "Nobody". Arch Hades might even confirm it.

      The study that you are using is in fact itself quite an outdated haplogroup paper that basically labelled anything that wasn't R1b and R1a as "North African" and "Jewish". If we follow their quite unacceptable classification method, it goes without saying that Italy in general is waaaaaaaaaaaaaaaaaaaaaaaaaay more "North African" and "Jewish" than Spain. But waaaaaaaaaaaaaaaaaaaaaaaaaaay more.

      And this Atlanto-Mediterranean:

      Is surely not swarthier than this:

    7. Oh can you.
      Than let Arch Hades clear up the mystery. Liitle excitement for me, because i already know that im not 'Nobody'.

      I will give your assessment of Adams et al a DOUBTFUL.
      Seems that you are confusing Figure2 with Figure4 - an honest mistake.
      In terms of Haplogroups, Italy is waaaaaaaay less North African than Spain - beacuse Italy is E-V13 (Neolithic) and Spain is E-M81 (Maghreb). And what exactly is a Jewish Haplogroup? Mitochondrial haplogroups overwhelmingly stem from European donors (Costa et al 2013). Botigue et al actually revealed that Italains are the least related to Jews (ItalainsJ/Ashkenazi) and that all Europeans range between 0-1.9% Jewish admixture (incl. Italains/Tuscans). And most of all Behar et al 2013 revealerd that Jews are like any other people, with several admixture components and that Sicilians and Cypriots are closest due to the similar amounts of the same components.

      I see your Atlanto-Mediterranean and i raise you a Nordo-Alpinoid from South Italy

      and he is surly not swarthier than this

    8. I forgot to add, that i do agree with you that the SephardicJewish in Adams et al is nonsensical and i didnt even bother with that.

      My point with Adams et al is that the Atlantic side of Iberia has in fact a higher North African ancestry than the Mediterranean side of Iberia. Best example Andalusia (east/west).

    9. I am pretty sure you are both the same user with two different accounts.

      You need to get a bit updated on E-V13 and other sub-clades of E1b1b. Spain also has E-V13 and Italy, France and the Balkans also have E-M81. Plus E-V13 itself in fact probably arose either in the Near East or North Africa, and from there made its way to Europe via Italy or the Balkans:

      Since when is Pique "swarthy"???

      Agreed on Adams et al., it made several quite incorrect assumptions. No other haplogroup study has supported its strange assumptions.

    10. And i am absolutely sure that we are not the same person.

      I know that Spain also has E-V13. well know that Ave07 (7ky ago) had E-V13 (Lacan et al) but now the update: Szécsényi-Nagy (2015) sequenced a nother two from the Lengyal and Sopot culture (also 5-7ky ag) who were E-M78 the ancestral clad of E-V13 maybe they were already dreived but not further sequenced.

      Pique is not at all swarthy, that was the point.

  8. North Italians are closest to Sardinians (Neolithic farmers / Ötzi) where as Greeks are somewhat Slavic (not really ancient Greek) !!!

    1. North Italians are closest to Iberians, whereas Greeks are closest to South Italians and other Balkanites. Sardinians are separate from everyone but closest to Basques.

    2. Basques are way more Northern Influenced than Sardinians. Basques are just an isolated population in Iberia that because of heavy cultural isolation and later inbreeding, kept the old pre Indo-European Iberian language, despite a large doze of steppe/Proto Indo-European admixture [27%]. But Basques are certainly not even close being a near pure Neolithic population like Sardinians are.

  9. I'm northern Italian with parents from Venice and Piacenza. I did a lot of genetic testing and I came out 25% northern European, 30% Italian and the rest was everything else in Europe. It was very spread out, because northern Italy was made up of many different people. We share markers with a lot of European groups. I had a Scandinavian component too. It's typical on 23andMe for northern Italians to only be 1/3 Italian. The difference between me and my southern Italian friends, is that I was 100% European and they were 85%. They were 15% North African and Middle Eastern. This shows up on 23andMe and on Geno 2.0, as well as Gedmatch and FTDNA. I had less in common with southern Italians than someone from Austria. Some other northern Italians came in with a lot of French and German. More than me. Same with most central Italians. Everyone south of Rome seemed to be closer to Greeks and Assyrians on Gedmatch as well as Middle Easterners. FTDNA has southern Italians also coming in with a lot of Middle Eastern. I grouped with many from Austria, Switzerland, Germany and the Alps in general. I did not group with southern Italians at all. Also remember that Sardinians are trapped hunter gatherers. They are one of the original European groups along with the Basque. The Basque are more Northern, but the Sardinians predate the Neolithic period or they wouldn't be I2 Y-DNA so heavily. So to sum up, I came out totally European and only half southern European. Germans come out 36% southern European on Geno 2.0. So, it's all relative and is a gradient with southern Italians coming in at 85% southern European with the rest being Middle Eastern and North African. I've seen some Germans come in at 40% southern European on Geno 2.0, so that's not much away from northern Italians who can be up to 50%. Northern Italians come in with up to 20% French and German on 23andMe and other sites like Geno 2.0. It's apples and oranges, but I think it's fair to say that northern Italians are between the Germanics and southern Italians. We are in the middle of these groups. I think that too many scientists have a political agenda and when you finally get tested and compare yourself to others in Italy, you then see a noticeable difference below Rome and above the Po River. I don't trust studies when the genetic testing companies are coming up with another answer. It's been very consistent. All genetic sites have me pegged well and are very consistent. I've tested with three companies and ran my data through UCLA and Gedmatch. I've also been privately tested. Northern Italians are not close to Iberians at all. I was closest to Austrians, because I'm part northeastern Italian. Northwestern Italians came in closest to the French. Unless you are Italian and share data with other Italians, you have no clue what is going on. Too many studies are wrong about northern Italians. We aren't German. We are predominately Y-DNA R1b, as was my father. Only my one X chromosome was part Iberain. The other was British and Irish... meaning Celtic. I'm 100% Italian stock. I wasn't different than many northern Italians who can prove their ancestors have been in northern Italy for hundreds of years.

    1. Those are commercial DNA testing companies, not peer reviewed science. Most of which i'm unfamilar with aside from 23andme. Often their data sets are odd and terminalogy questionable. Why the hell for instance would Germans be labled as 40% Southern European? Sounds retarded. Secondly my article does not try to convey that North and South Italians are the same population. Someone from Venetia or Lombardy might be equadistant between Austria and Sicily, or a bit closer to the French than Sicily. That's not the point or this article. Sicily is to Southern Europe what Finland is to Northern Europe, the outlier. But Northern Italians do not show higher ancestral strains that peak in Northern Europeans than most Spaniards or Southern Slavs...who are Southern Europeans. Northern Italians do not show a special genetic relationship to Northern or Central Europeans in comparison to many other Southern Europeans like the mentioned groups above. FYI, I've seen where dozens of Northern Italians cluster on PCA in 23andme over on The Apricity and Forum Biodiversity, they cluster with or most often marginally South of Spaniards, who then cluster marginally below Southern French. So again there's nothing spectacularly 'Northern' about Northern Italians at all. Listen i'm pro Padanian nationalism and respect the different culture Northern Italians have created in comparison to The Mezzogiorno, but it's best to have an objective and truthful understanding of your racial/genetic origins. Take care dear. And i'd be interested in seeing your 23andMeResults.