Early European Farmers and the Anatolian DNA Trail

Twenty-one of the first 26 Anatolian farmers to have their genomes studied came from a single mound: Barcın Höyük, south of the Sea of Marmara in north-western Turkey. The DNA did not come from teeth or long bones. It came from the cochlea, the snail-shaped cavity inside the petrous bone of the skull, which can yield up to a hundred times more DNA than teeth.

That technical trick, reported in 2015, helped settle one of archaeology’s oldest arguments. The early European farmers who built the first villages from Greece to Spain were, genetically, the descendants of people like those buried at Barcın.

The argument the genomes ended

Farming began in south-west Asia after the last Ice Age and reached Europe thousands of years later. There were two classic explanations. In one, farmers migrated and brought crops and animals with them, a process sometimes called demic diffusion. In the other, local hunter-gatherers learned the new way of life from neighbours, so the ideas moved but the people mostly stayed put.

Pottery, house plans and crops could not settle it, because both models predict similar artefacts. DNA could, because the two models predict very different genomes.

Four studies, one chain of evidence

1. 2014: three ancestral populations. Iosif Lazaridis, Johannes Krause and colleagues sequenced a 7,000-year-old farmer from Stuttgart, Germany, and eight 8,000-year-old hunter-gatherers from Luxembourg and Sweden, and compared them with 2,345 people living today. Most modern Europeans, they found, descend from at least three populations: western hunter-gatherers, Ancient North Eurasians, and early European farmers of mainly Near Eastern origin.
2. 2015: the source population. Iain Mathieson and colleagues analysed 230 West Eurasians who lived between 6500 and 300 BC, including the first genome-wide data from Anatolian Neolithic farmers. Those farmers, they showed, belonged to the population that was the source of Europe’s first farmers.
3. 2016: the Aegean link. Zuzana Hofmanová, Joachim Burger and colleagues sequenced five Neolithic people from both sides of the Aegean: Bar8 and Bar31 from Barcın, Rev5 from Revenia in northern Greece, and Pal7 and Klei10 from Paliambela and Kleitos, about 2,000 years later. All five were strikingly similar to each other and to early farmers across Europe.
4. 2016: deeper roots. A second Lazaridis study added 44 ancient Near Easterners from about 12,000 to 1,400 BC. It placed the Anatolian-related farmers within a wider mosaic of the first farming populations.

Hofmanová’s team did not hedge. Their study, they wrote, provides “the coup de grâce” to the idea that farming spread across Europe through ideas alone, without significant migration. A chain of ancestry ran from central and south-western Europe back to Greece and north-western Anatolia.

The first farmers were not one people

The 2016 Near East study added an important twist. The earliest farmers of the southern Levant (modern Israel and Jordan) and of the Zagros Mountains (modern Iran) were strongly different from each other genetically. Each descended from its own local hunter-gatherers. Farming, in other words, was taken up by several distinct populations, not spread by a single founding group.

Their descendants then went in different directions. Farmers related to those of Anatolia spread west into Europe. Levant-related farmers spread south into East Africa. Iran-related farmers moved north into the Eurasian steppe, where their ancestry later turns up in the Yamnaya herders.

These early Near Eastern populations also carried a mysterious component called Basal Eurasian ancestry. It comes from a lineage that split from other non-Africans before they diverged from each other and that shows little sign of Neanderthal mixing. The 2014 study estimated that early European farmers had about 44% of it. One hypothesis, still debated, is that this dilution helps explain why Europeans today carry slightly less Neanderthal DNA than East Asians.

What our atlas data shows

The AADR Genome Atlas on this site draws on release v66.p1 of the Allen Ancient DNA Resource. Filtering for Neolithic and Pre-Pottery Neolithic labels from Turkey returns 110 individuals, with mean dates from about 10,700 to 5,600 years ago.

• Sites: Çatalhöyük (37), Barcın Höyük (25), Boncuklu (22), Aşıklı Höyük (6), Menteşe (5) and smaller samples elsewhere.
• Men with a Y-chromosome call: 44, of whom 19 carry haplogroup G2a, the most common lineage, with a few C1a and J2a.

Two caveats. The earliest of these people, at sites like Boncuklu and Aşıklı, lived long before farmers left Anatolia, so they are better seen as ancestors of the migrants than as migrants themselves. And sample counts track excavation effort, so Çatalhöyük’s lead says more about archaeology than about ancient population size. Our migration map rates this expansion, from Çatalhöyük towards the Danube, the Mediterranean and Iberia, as a high-confidence event.

Two roads west, and a mix along the way

Haak’s 2015 analysis found that early farmers in Spain, Hungary and Germany descended from a common stock, despite reaching those places by different routes: along the Mediterranean coast and up the Danube. A 2019 study of Neolithic Britain traced most of its farmers to the Mediterranean branch; they replaced the hunter-gatherer population of people like Cheddar Man.

The same study found that early farmers from across Europe cluster genetically with present-day Sardinians, echoing earlier work. Sardinia is the modern population closest to Europe’s first farmers, which makes the island a useful living reference point for an ancestry that has been diluted almost everywhere else.

The farmers did not stay genetically unchanged. By about 6,000 to 5,000 years ago, farmers across much of Europe carried more hunter-gatherer ancestry than their predecessors had. How much, and how quickly, varied by region, and this is where active debate continues. For a later case from Iron Age central Italy, see our Etruscan genome project.

What early European farmers carried in their genes

Hofmanová’s Aegean farmers were probably not able to digest milk as adults, but four of the Aegean individuals carried two copies of the SLC24A5 variant associated with lighter skin, and four carried at least one copy of a lighter-skin SLC45A2 variant. As the authors note, reduced pigmentation was not solely a high-latitude phenomenon. That makes farmers a major source of the light-skin variants common in Europe today. You can read how those variants spread in our article on the evolution of human skin colour, and why milk digestion came much later in the evolution of lactose tolerance.

Their lack of milk tolerance is unsurprising. The main European lactase-persistence variant only reached appreciable frequency in Europe around 4,000 years ago, and it remains below 15% in modern Greek, Turkish and Sardinian populations.

Mathieson’s scan also detected natural selection at genes linked to diet, pigmentation and immunity, plus two separate episodes of selection on height. The farming way of life, it seems, reshaped bodies as well as landscapes.

One preserved body shows the result. Ötzi the Iceman, who died in the Alps about 5,300 years ago, carried about 90% early Anatolian farmer ancestry, as our article on Ötzi the Iceman explains. To see how this recent chapter fits into seven million years of human evolution, open the interactive family tree.