Ancient Roman DNA

November 28, 2019

We've already seen that modern Italians are genetically close to ancient Mycenaean Greeks, and now we finally have Ancient Roman DNA, which shows the same genetic closeness, refuting Nordicist fantasies about an original "Northern European" stock becoming "orientalized". The Romans (Latin and Etruscan) were all Southern European, similar to modern Italians from North to South. There were foreigners during the empire, but they weren't very exotic and didn't have much of an impact.

Most of the samples are from cemeteries in and around the city of Rome, so on one hand they're not representative of the whole country, but on the other, Rome was the biggest, most traveled-to city in all of Italy, so if there wasn't much diversity or mixing going on there, that means there would have been even less in other cities and (especially) in the small towns and rural areas that made up most of the country.

As in all of Europe, there were major changes between the Mesolithic and Iron Age, but from then on, the samples all have the same 4 main components: Western Hunter-Gatherer, Anatolian Neolithic (farmers), Iran Neolithic and Steppe Eneolithic (the last two arriving mostly with Indo-Europeans), plus very low levels of Morocco Hunter-Gatherer. Even the "diverse" imperial samples have those same components, just in slightly different proportions, so the foreigners were not that foreign.

Iron Age & Republic (dark blue)

Most of these early Latin and Etruscan samples cluster between Northern Italians and Spaniards, and a few others cluster with Northern Italians, Central Italians, Southern Italians and off toward Sardinians. So the original, pre-empire Romans were heterogeneous but fully Southern European, just like modern Italians.

By 900 BCE at the latest [before the founding of Rome], the inhabitants of central Italy had begun to approximate the genetics of modern Mediterranean populations. [...] The Iron Age individuals exhibit highly variable ancestries, hinting at multiple sources of migration into the region during this period (Figs. 2A and 3B). [...] Together these results suggest substantial genetic heterogeneity within the Etruscan (n = 3 individuals) and Latin (n = 6) groups. [...] In contrast to prehistoric individuals, the Iron Age individuals genetically resemble modern European and Mediterranean individuals, and display diverse ancestries....

Imperial Rome (teal)

At this time there appears a "tail" toward the Near East, reflecting migrants to Rome from the Eastern Mediterranean. However, most of them cluster no farther than Cyprus, which seems to confirm that "Middle Eastern" migrants to Rome were of predominantly Greek ancestry.

During the Imperial period (n = 48 individuals), the most prominent trend is an ancestry shift toward the eastern Mediterranean and with very few individuals of primarily western European ancestry (Fig. 3C). The distribution of Imperial Romans in PCA largely overlaps with modern Mediterranean and Near Eastern populations, such as Greek, Maltese, Cypriot, and Syrian (Figs. 2A and 3C).

Late Antiquity (green)

Here the "tail" to the Middle East disappears and the samples shift back westward. After the capital of the empire moved east and Rome was sacked, the population declined to almost nothing, and the city would have been repopulated by Italians from other parts of the country, and possibly other Europeans too.

The average ancestry of the Late Antique individuals (n = 24) shifts away from the Near East and toward modern central European populations in PCA (Fig. 3D). [...] This ancestry shift is also reflected in ChromoPainter results by the drastic shrinkage of the Near Eastern cluster (C4), maintenance of the two Mediterranean clusters (C5 and C6), and marked expansion of the European cluster (C7) (Fig. 4C).

This shift may have arisen from reduced contacts with the eastern Mediterranean, increased gene flow from Europe, or both, facilitated by a drastic reduction in Rome's population in this period to less than 100,000 individuals, due to conflicts and epidemics (1, 3).

Medieval & Early Modern (yellow)

Here the samples are back to being similar to the Iron Age samples, and they're identical to modern Italians from North to South, except for a few outliers tending toward Spain and France. The latter may be related to the Holy Roman Empire and Norman conquests.

In the Medieval and early modern periods (n = 28 individuals), we observe an ancestry shift toward central and northern Europe in PCA (Fig. 3E), as well as a further increase in the European cluster (C7) and loss of the Near Eastern and eastern Mediterranean clusters (C4 and C5) in ChromoPainter (Fig. 4C). The Medieval population is roughly centered on modern-day central Italians (Fig. 3F).


This shift is consistent with the growing ties between Medieval Rome and mainland Europe. Rome was incorporated into the Holy Roman Empire (3), which spanned much of central and western Europe. The Normans expanded from northern France to a number of regions, including Sicily and the southern portion of the Italian Peninsula (and even sacked the city of Rome in 1084), where they established the Kingdom of Sicily (3, 36).


  • By the time Rome was founded, its people (Latins and Etruscans) were already genetically Mediterranean, overlapping with modern Southern Europeans from Spain to Sicily.
  • The empire brought migrants to Rome from the Eastern Mediterranean, but they were mostly of Greek ancestry and not that different than the Romans.
  • When the western half of the empire fell, the migrants disappeared as Rome's population declined drastically. The city was later repopulated by Italians and other Europeans.
  • By the Middle Ages, the inhabitants of Rome were back to resembling Iron Age people and identical to modern Italians from North to South, with a few outliers tending toward Western and Central Europe.

Antonio et al. "Ancient Rome: A genetic crossroads of Europe and the Mediterranean". Science, 2019.

Related: Pigmentation Through Roman History


Sarah Nikas said...

If you read the supplementary material on that study the authors acknowledges elevating the african component in that study by 10 fold by choosing a poorer fitting source population for it than what the admiture software recommended. This study is highly dubious at best because of author bias. Additionally the "Iranian" that they equate to middle eastern is almost certainly actually a caucasian component, and not in fact iranian. Other studies have confirmed no iranian admixture to be found in europe, including italy. Hannah Moots is a well known open borders advocate so it's advantageous for her to modify the results to protray Rome as some sort of multicultural paradise.

I posted this on racial reality with my conversation with aitor. Not sure if you saw it but if you'd like more specific proof I'd be happy to provide it. My email is

Italianthro said...

I noticed that stuff about the African component. It's weird but at least they acknowledge it. What's interesting is that in the original report about the study, it looks like they had used Levant_N instead of Morocco_HG (they're very close in PCA so I'm not sure how "African" that component really is anyway).

As for Iran_N and CHG, they're also very similar and have both been used as proxies for the "Middle Eastern" component of Indo-Europeans. This is not the first study to do that, and I don't think it's a big deal.

Anyway, I'm used to studies being biased in their conclusions. But the data is what counts, and here it clearly shows that "multicultural diversity" was very low and that there's been pretty strong genetic continuity.

If you want to send me more info, my email is linked in the copyright at the bottom of the page, but you should post as much as possible here in the comments so others can see too.

Sarah Nikas said...

Alright, so since I can't post pictures or PDFs directly, I've uploaded them to MEGA links for anyone interested to see and download.

Yes, I noticed that too about the levant_N in the original report. In the supplementary data (!4rwnSCLS!foePUcue8YEWLpEddSj3amHRuOz_A81FYCSfJCV3hX4), it shows they did include the levant as a possible source in the ADMIXTURE test, but the software rejected it outright in favor of "Neolithic moroccan". However they also acknowledge that by replacing "neolithic moroccan" with the poorer fitting source population "moroccan hunter gatherer", that their north african results were magnified by 10 times compared to the better fitting data set.(!1z4nzQCK!wuth57HLUGaXEO9v49y8jCKaHpQrP_GkZs-BNsAYTWI) They don't explain there being any reason for them using the poorer fitting population other than that it increases reported african ancestry by 10 fold - which - is pretty much the hard definition of cherry picking.

And yeah, Iran_N and Caucasian hunter gatherer are often conflated because the ancestries are thought to be similar, however, again, it has been studied that the neolithic iranian genome has had no impact on Italy, and instead only the caucasian hunter gatherer ancestry is reprsentative of this ancestry for the region of europe. (!svgRAALa!sjcnF0bPFbOKbHHxPTXCIVkjSrr8hMn4LXE6oUupOXc) The increase in this component really represents the largest change from "iron age/republic era romans" to "imperial" and modern day central italians - which the author attempts to describe as "near eastern immigration". In reality, based off the sicilian bell beaker samples, we know that south italians largely had this CHG component already elevated in them by the bronze age (in contrast to central/north italians who seemed to have recieved it around the time of the Iron age to Roman Empire). The implication to me is that this shows a high degree of immigration from either southern italy or greece/anatolia to Rome - and that the authors are purposefully labelling it "iranian" to decievingly conflate it with the "near east" - even though levantine admixture wasn't found in the studied Roman population. They even briefly acknowledge twice in the study that it could entirely be a Caucasian component (!1z4nzQCK!wuth57HLUGaXEO9v49y8jCKaHpQrP_GkZs-BNsAYTWI),(!wnhTEQwJ!evMgAmxZt6kwG_vGKALEQ_qj4va0C_NwTygPfQ2f_ac), but refer to it as iranian to further a narrative of "near eastern admixture".

This brings me to my final point in that the terms and phrases the author used were quite different than your own. I don't think they described any degree of continuity throughout the whole study and chiefly referred to Rome as a "melting pot" and genetic product of whoever it was politically associated with, which certainly wasn't the truth that the admixture results showed. If anything, the admixture results for modern central italians looks quite similar to the imperial sample, which implies the resettlement of Rome by outlying rural italians rather than some a-historic holy roman empire settlement of Germans - as both the elevated anatolian and caucasian component stay roughly the same. This is why I'd like to point out independent author bias and objectives in this study as their advocacy for open borders and multiculturalism is not something they try to hide. (!Ay43HKib!HfKTwjsyWDz78r2ERNmdQ0d1Xpkiv2pKRdAXQK6gbF0)

Palermo Trapani said...

This article actually is consistent with the Raveane et al (2019) article Population structure of modern-Italians..." The study is the most comprehensive on Italian populations and uses 1,616 individuals from the 20 government regions and it compares the Italian samples to some 140 world wide populations

Figure 2 is the interesting Chart. The paper again consistent with prior literature (Parolo et al. 2015; Sazzini et al. 2016) shows 3-4 related Italian Clusters, Sardinia, Northern-Central Italy, Central Italy and Southern Italy-Sicily. Central Italians, except for Tuscany, clusters closer with Southern Italy and Sicily, which means Lazio-Rome.

You can see the source population. AN (yellow) is Anatolian Farmer DNA (Early European Farmer), WHG (Brown) is Western European Hunter Gatherer, CHG (Green) is Caucus Hunter Gatherer, EHG (Blue) is Eastern European Hunter Gatherer, IN (Black) is Iranian Neolithic, and then you see North Africa (Pink) and EAsia (purple). I looked at the population samples in the spread sheet provided by the authors. First, it is clear than all Italian Samples are majority Anatolian Farmer DNA, which the study on Rome shows was the dominate DNA source, along with WHG before the Steppe Migration. The ranges are 56% from SItaly1 (Basilicata sample) to NItaly2 (72%). Sardinia is > 80% AN which is why they are the closest to Otzi the Iceman. Sicily, where my ancestors come from is about 60% AN. Second, note the CHG DNA from Sicily to NCItaly1-NCItaly3 which includes Tuscany and SCItaly1 which has Lazio (Rome) in it. Recent work by suggest that the Steppe Migration can be modeled as 57% EHG-43%CHG. However, I suspect as you move North in Europe the Steppe is almost exclusively EHG whereas the Southern Steppe into Italy and Balkans has more CHG. Third, even in the North Italian Samples, which has the highest AN DNA at 72, the EHG is never more than 25%. It is likely therefore that the EHG you see in Northern Italy came about via the Germanic Invasions as the Roman empire in West Collapsed (i.e. Vandals,Lombards, Visigoths) and these tribes also invaded Sicily and the South at the same time. Later, the Normans would invade Sicily and move up the boot and sack Rome as well after driving out the Saracens and wanting control over Sicily and Southern Italy and the Pope not giving them total control. Fourth, there is also some Iranian neolithic in Southern Italy and Sicily documented as well.

So this study on ancient Rome in reality shows modern Italians and the Romans are the same people.