The Dago Divide

Eyetalian genetics.

It’ll be interesting to see how Italianthro tries to parse this.

In any case, the bottom line is that the differences between North and South Italy are mostly of ancient origin – and while these ancient differences have been reinforced by additional migrations, these additional migrations have not (significantly) added novel components.

Abstract (emphasis added):

The cline of human genetic diversity observable across Europe is recapitulated at a micro-geographic scale by variation within the Italian population. Besides resulting from extensive gene flow, this might be ascribable also to local adaptations to diverse ecological contexts evolved by people who anciently spread along the Italian Peninsula. Dissecting the evolutionary history of the ancestors of present-day Italians may thus improve the understanding of demographic and biological processes that contributed to shape the gene pool of European populations. However, previous SNP array-based studies failed to investigate the full spectrum of Italian variation, generally neglecting low-frequency genetic variants and examining a limited set of small effect size alleles, which may represent important determinants of population structure and complex adaptive traits. To overcome these issues, we analyzed 38 high-coverage whole-genome sequences representative of population clusters at the opposite ends of the cline of Italian variation, along with a large panel of modern and ancient Euro-Mediterranean genomes.

Only 38?!  I realize that high-coverage whole-genome sequencing is not trivial, but 38 is an absurdly low sample size for a population genetics study.

Results

We provided evidence for the early divergence of Italian groups dating back to the Late Glacial and for Neolithic and distinct Bronze Age migrations having further differentiated their gene pools. We inferred adaptive evolution at insulin-related loci in people from Italian regions with a temperate climate, while possible adaptations to pathogens and ultraviolet radiation were observed in Mediterranean Italians. Some of these adaptive events may also have secondarily modulated population disease or longevity predisposition.

Conclusions

We disentangled the contribution of multiple migratory and adaptive events in shaping the heterogeneous Italian genomic background, which exemplify population dynamics and gene-environment interactions that played significant roles also in the formation of the Continental and Southern European genomic landscapes.

Now, this is interesting (emphasis added):

Sharing of chromosome chunks among individuals belonging to the identified population clusters was investigated with CHROMOPAINTER. Accordingly, both Italian groups were found to share similar proportions of DNA segments with Sardinians (N_ITA, 48%; S_ITA, 43%) and Northern Caucasian populations (~ 10%), which have been previously supposed to be suggestive respectively of Early Neolithic and Bronze Age contributions to the ancestral pan-European genetic background [7, 8], while presenting considerably different painting profiles for the rest of their genomes (Fig. 1b). In particular, S_ITA showed substantial sharing (30%) with Near Eastern populations, while this signature is completely absent in N_ITA. Moreover, S_ITA presented 17% of chromosome chunks in common with Southern Caucasian groups in contrast to the 9% observed for N_ITA, although this pattern might be influenced by the fact that populations from Southern Caucasus are genetically close to those from the Near East [77]. N_ITA finally turned out to share DNA segments also with Eastern and Northern European groups (19%) and the Basques (12%), differently from what was observed for S_ITA (Fig. 1b).

CHROMOPAINTER yielded data showing northerners and southerners shared a majority of ancestry, but differed in certain minority affiliations, with a “Near Eastern” component only in the South, while in the North, some “Eastern and Northern European” and “Basque” similarities were found.

Now, here is the key point.  The CHROMOPAINTER data - the same data - were then analyzed with GLOBETROTTER and a significantly different outcome was observed.  Both northerners and southerners were now modeled as a mix of “North European” (and “Basque” for Northern Italians) and “MENA” (“Near Eastern and North African”) – albeit in different proportions.  Thus (emphasis added):

CHROMOPAINTER painting profiles were then used to infer admixture proportions of N_ITA and S_ITA with respect to the other Euro-Mediterranean clusters by calculating co-ancestry curves with the GLOBETROTTER method (see the “Methods” section). Admixture events involving a Northern European (and Basque in the case of N_ITA) gene pool and a Near Eastern/North African source of gene flow were found to have affected both Italian population clusters (Fig. 1c, Additional file 1: Figure S2). Nevertheless, N_ITA and S_ITA showed inverse proportions of these admixture sources, with respectively 59% and 32% of Northern European (and Basque in the case of N_ITA) contribution, coupled with 41% and 68% of Near Eastern and North African one.

While in general the two analyses are similar in that the northerners are “more northern” and the southerners “more southern (as expected), the pattern is different, in that the first analysis was more disjunctive and the second more clinal (and likely more realistic considering Central Italians represent a mid-way point in the clines).

Note also how this is wholly consistent with the many posts I have made being critical and skeptical of “admixture” analyses such as what is shown in this paper – as the outcome is completely dependent on the methodology used (and of course the reference samples used) – so findings vary from paper to paper and here actually vary from method to method within the same paper.  Note, amusingly, how the authors use the word "Nevertheless" to distract the reader from the point that the two methods of analyses yield completely different patterns (e.g. ,the first shows zero "Near Eastern" in northerners, but the second shows 41%!). 

Keep in mind, again, that how you model populations depends on the reference samples used.  And models are just that - models.  Just because you can model a population as a mix of "Northern European" and "MENA" doesn't mean they literally are such a mix, just that their gene frequencies, when measured against whatever reference samples used, and given the particular methodology used, resemble that more than not.  A different set of reference populations would allow the same sample to be modeled based on those particular references. And we have learned from 23andMe how dependent on context (reference samples, confidence levels, etc.) "chromosome painting" is.  

Indeed, if the authors of this paper wanted to get the results obtained, they certainly used the "appropriate" reference samples and methodology.  And when things go awry, use the word "Nevertheless."

But there are some general take-away points that likely have some validity.  See this. Thus, in summary, the northerners are more like “South-West Europe” (e.g., Spain) and the southerners more like “South-East Europe” (e.g.., Crete).  But, based on some genetic kinship data I’ve seen for southerners, some overlap must exist (e.g., high S.Italian kinship hits to Tuscany, “Italy” [Bergamo], Basques, French, Sardinia, Iceland, etc.).  

As regards Central Italy (emphasis added):

Gene flow from the Near East instead seems to have affected mainly Central Italy and for a longer period than other regions of the peninsula…people from Central Italy present variable degree of admixture between them, but no additional private ancestry fractions.

Why Central Italy is so much more “Near Eastern” than the other areas is uncertain, perhaps the Etruscan influence or perhaps Imperial Roman influxes

There are also functional data in the paper for those interested, although nothing there about “singing on balconies,” military ineptness, or Schettinoism.

If Der Movement is going to get all excited (*) over these data, I’d advise them to calm down before their priapisms get out of control. Three points to consider from these data (and previous studies) as the Type Is wipe the sweat off their foreheads:

1. That differences exist between Northern and Southern Italians has already been long known (albeit often exaggerated) by virtually everyone except perhaps Italiathro/Racial Reality.  These sorts of studies have been done (repeatedly) already.

2. Northern Italians are not “Celto-Germanic Nordics.”

3. Der Movement’s True Romance racial history of Southern Italy – an initially northern-like population becoming darkened through admixture with (or replacement from) Negroes/Moors - is not correct.

The fundamental differences were in place thousands of years ago, they are very ancient, and are part of the European ethnogenesis of the various types of Italian peoples.

*Of course, Amren breathlessly ran a summary of these data but I don’t recall that they ever focused on the recent study on Ancient Roman genetics. Interesting, no?  Also ignored at that site (but discussed here) are population genetic studies demonstrating the modern Middle Eastern origin of the Jews, the tropical origin of East Asians, and East Asian/Siberian admixture in Northern Europe.  Won't find that at Amren, but obsession with the wops, yes indeed. Interesting, no?