Deciphering Ancient DNA
Making sense of ancient human migrations through genetic archaeology is a fun and very challenging task. We are quickly learning that the world was far more complex and interconnected for much longer than we previously thought. The number of known ancient cultures has expanded greatly in recent decades, and the intricate web of the back-and-forth flow of culture and DNA is getting more complicated by the day.
That said, there is a huge volume of work available to the public now, thanks to unbelievable advances in technology. DNA testing is widespread, precise, and relatively cheap now. And all the data with all the various tools out there is available and affordable for hobbyists like myself to tinker with.
I’ve been digging through my own family history for some time now, and along the way learning everything I can about our species’ history. I guess you could call what I do a “meta-analysis” using multiple sources and tools, trying various combinations, and triangulating.
Before jumping into some of the approaches and findings, here is a list of websites I have found to be the most useful for my purposes.
First, the maps I’ll be using here come mostly from this wonderful Atlas:
Second, the best resource in genetic archeology these days is probably:
I recently found the Illustrative DNA will give you your G25 coordinates (along with multiple analyses and visualizations of their own):
Once armed with those, you can head over to:
There, you can run all kinds of models to find proximity to ancient or modern DNA samples, thought the tools are not user friendly enough for many…
Fortunately, there’s Genoplot:
They won’t give you your exact G25 coordinates, but you can simulate them from Euro K13 or Harappa World calculators, which will get you pretty close (depending on your ethnic mix, of course…) for free :-)
The great thing about Genoplot is that their tools are visually appealing and generally more user friendly than Vahaduo’s. I’ve written a few overviews of the tools available there to show how to run Admixture and PCA analyses.
Primarily, I use all of these to explore my own family history, including my wife’s Ashkenazi Jewish side. I’m a Ukrainian / Baltic mix, which turns out is a combination of ancient Northeastern European and Eurasian Steppe heritage. Balto-Slavs mixed with Scythian-Sarmatians and then got a doze of Germanic (and some proto-Celtic British Isles) blood through the Vikings between 800 and 1100 AD. Ashkenazi Jews, on the other hand, are a fascinating puzzle that remains to be fully solved, but generally they are a crazy mix of ancient Mediterranean and Middle Eastern hiritage with varying dozes of Central European, Balkan, and Steppe Nomad blood.
I generally start at the end of the Stone Age, when Eurasia looked like this:
Europe was dominated by the Corded Ware culture, but Bell Beakers were already emerging from the Iberian peninsula. Minoan culture was sturting to prosper in Creete, while my ancient Baltic ancestors were spreading the Battle Axe culture to Scandinavia.
Bell Beaker culture completely took over eventually, leaving only Battle Axe and Minoans in Europe. Meanwhile, Anatolia was a hot mess:
Bronze Age is where I like to start looking at more defined populations we are used to and building models from them. Unetice culture of Central Europe is where much of my Slavic ancestry likely originated:
This is where you can find plenty of ancient DNA samples that can be reasonably clustered into groups. I made a few calculators on Genoplot for these and ran my own DNA, as well as a number of later (Iron and Middle Age) samples to see their ancient breakdown.
The problem I see with most commercial websites, including Illustrative DNA, My True Ancestry, and Your DNA Portal, for instance, is that they will analyze DNA ancestry using an incompatible mix of various ages… You can’t put Bronze, Iron, and Middle ages together in one model, since the samples from the latter are themselves derivatives of the former!
For example, here’s mine from IDNA:
That’s not bad in the big picture, but Caucasian group is a couple millennia older than the other three… My wife’s is far worse:
Instead, I spend the time to filter the lists from Vahaduo or Genoplot to create age-specific models and calculators:
This starts to make more sense. With a bit of refinement, I can combine proper samples into familiar groups:
I like that the calculators in Genoplot will show you both the groups as well as the matching samples from them. And the maps are cool, though a bit buggy.
Let’s see what happened next in the world. Early Iron Age = Hellenistic world. Hallstat and La Tene cultures in Cental and Western Europe.
The major development was of course the rise of the Roman Empire and it’s struggle against the northern Germanic Tribes of “barbarians”. To the east, Scythians were being partially replaced by the Sarmatians, as further waves of nomadic invasions, the Huns and the Alans, were approaching. And to the south, Greek culture was dominating Anatolia, the Levant, and North Africa.
And so we come to the period of Late Antiquity and early Middle Ages, which saw the fall of Rome, the rise of Byzantium and Islam, and a great web of intertwined migrations / invasions all across Eurasia.
It’s fascinating to run some of the historical samples from Iron Age or later Middle Ages against a Bronze Age ancestrial model. Here’s a slice of Eastern Europe with Baltic and Scythian-Sarmatian samples. Also some of the most fascinating people in history — Szolad site in Hungary, where Slavs, Avars, Alans, and Lombards all came together during the period of Great Migrations.
It’s fascinating to see the old Bronze Age European components in these! While Ingrians were mostly isolated, you can see the Iron Age Estonians already had a large influx of Eastern European blood from an earlier era. You can also see that most of the nomadic people had also already mixed with europeans, with only the earliest pre-scythian samples showing a large steppe component. And Moldovan “Scythians” were actually mostly locals.
Next is another fascinating sample of Levantine DNA from the Iron Age:
What I love about these is the earlier Anatolian component in almost all of them, likely a mix of native and Greek heritage. Also the Italian blip for one of them.
And finally, we come to the Viking Age.
As if the prior period of migratory invasions wasn’t enough, these Scando-Balto-Germanics really took changing Europe quickly to a while new level. Sailing the norht seas and the vast network of Eastern European rivers down to the Black and Caspian Seas, they established many of the most succesful nations / empires of today, including the UK and Russia, and played a central role in all European politics, including those of the Byzantine Empire.
And of course they left a genetic trace everywhere they went. But what heritage did they themselves already carry? Let’s take a look:
Through the Bronze Age lens, we see something of a split between ancient British and Nordic heritage, with varying degree of other European admixtures. Except, of course, for the Foggia sample, which looks a lot more like the Ashkenazi Jews of today… I thought they’d be Lombards, but looks like they are something else altogether.
When you use the Iron Age lens, you see much more concentrated Nordic origins for most, though Britain still lingers. Eastern Europe and Steppe Nomads are now added to the mix. And the Foggia folks are now an Iberian, Italian, Balkan, Levantine mixture. I guess they are “Italian” after all, and not Lombard at all.
Interesting to look at both BA and IA models for myself and my AJ wife:
And might as well look at the Viking Age and beyond:
Using Vahaduo PCA tool, we can also look at the overlaps, as well as the world as it is today. Here’s me vs modern samples and Viking Age overlay:
Hellow Gotland (Rus?) Vikings and Germanic-Slavs (Sorbs?)
And here’s my AJ wife:
Ah, there those pesky Foggia “Vikings” are :D
Playing with all of these has been a ton of fun and highly educational. Keep in mind, the samples are limited, and the techniques keep evolving, so everything is subject to change. But the world’s history, though ever more complex and intertwined, is starting to make a lot more sense to me, as is my own family’s little part.
