European American DNA: the ancient origins behind a colonial-era genome

What are the ancient origins of European American DNA? The same three deep layers as any European genome, western hunter-gatherer, Anatolian farmer and steppe pastoralist, blended in the ratios of the regions that populated colonial and nineteenth-century America: the British Isles, the Rhineland, the Netherlands and Scandinavia first, then Ireland, Italy and Poland. Ancestrify models each era from your raw DNA.

A consumer test hands a large share of American customers the same three labels: "British and Irish", "Germanic" and "Scandinavian", in shifting proportions, with a sliver of something else. The labels describe the last four centuries. This guide is about what sits underneath them: which European coasts built the colonial population, what the ancient genomes of those coasts contain, and why a genome assembled from several countries is a perfectly ordinary target for a formal ancestry model.

The short answer: A European American genome is a European genome whose sources happen to be spread across several countries. Every one of those countries carries the same three prehistoric layers, so the deep-time story is shared: foragers, then farmers from Anatolia, then a steppe reset in the third millennium BC that reached Britain by 2400 BC. The colonial blend added no new ancient source. It only changed the mixing ratios, and in many genomes added a small African or Native American share that the ancient layers do not explain.

The coasts that built the colonial population#

The seventeenth-century settlers of New England and the Chesapeake came overwhelmingly from England, with Dutch settlement on the Hudson and a Swedish colony on the Delaware. The eighteenth century widened the intake: Ulster Scots and Lowland Scots into the Appalachian backcountry, Palatine and Rhineland Germans into Pennsylvania, a steady Welsh and Huguenot trickle. The nineteenth century then remade the population with Catholic Irish after the Famine, a second and far larger German wave, Scandinavians into the upper Midwest, and, after 1880, Italians, Poles and other eastern and southern Europeans into the industrial cities.

Every one of those sources is a modern population with a published ancient-DNA transect, which is why the deep history of a European American genome can be told with unusual confidence: the ancient genomes were dug up on the source coasts, not in Massachusetts.

What the three labels are made of#

"British and Irish" is the Beaker-descended population of the islands. Olalde and colleagues showed that the arrival of Beaker-associated people after 2400 BC replaced roughly 90 percent of Britain's gene pool within a few centuries, carrying steppe-derived ancestry over the Neolithic farmers who had built Stonehenge. Early medieval England then received a large continental contribution from the North Sea coast, measured by Gretzinger and colleagues in 2022, and the Viking Age added a Norse layer to the same islands. Ireland and western Britain kept more of the earlier Beaker-era profile.

"Germanic" is the label for the Rhineland, the Low Countries and northern Germany, the very coast the Anglo-Saxon settlers sailed from. Genetically it is the Dutch and northwest German profile: heavy steppe-related ancestry over a farmer base, essentially stable since the Bronze Age. That is also why "British and Irish" and "Germanic" trade percentages so freely between test companies: the two sources overlap, and the boundary a company draws between them is a modelling choice.

"Scandinavian" is the northern end of the same continuum, with slightly more hunter-gatherer ancestry preserved and a Viking Age history of movement in every direction.

The three deep layers, applied to any European-descended genome#

Strip away the country labels and every European source reduces to three ancient populations whose ratios vary by region. The hunter-gatherers who recolonised Europe after the Ice Age survive as a minority everywhere, largest in the Baltic and Scandinavia. Farmers of Anatolian descent spread across the continent after 7000 BC and dominate southern European genomes to this day. Then, after 3000 BC, pastoralists from the Pontic steppe pushed west: Haak and colleagues found that the Corded Ware people of Germany traced about three quarters of their ancestry to the Yamnaya, and that steppe-related ancestry is now largest in northern and northwestern Europe and smallest in Sardinia and the southern Mediterranean.

An English, Dutch, German or Swedish source therefore hands a colonial genome a high steppe share, a moderate farmer share and a small forager share. An Italian or Polish source shifts the ratios, more farmer in the first case, a touch more forager and steppe in the second. Blending them produces a genome whose deep-era proportions sit between its sources, which is exactly what a steppe-percentage estimate shows.

Founder effects and the New World cline#

Colonial populations were small and grew fast, and small founding groups leave signatures. Communities that stayed endogamous for generations, the Pennsylvania Amish and Mennonites, French Acadians in Louisiana, some Appalachian valleys, carry measurable founder effects and a raised frequency of specific rare variants. In the population at large, the mark of the colonial era is subtler: Novembre and colleagues showed that within Europe, genes mirror geography closely enough to place a genome within a few hundred kilometres of its origin. A genome assembled from England, the Palatinate and Norway has no such single origin, and on a PCA it lands in the empty water between those three countries. That is not an error. It is the correct position of a genome with three sources, and it is why a "closest population" ranking for a European American kit is so often a coin toss among neighbours.

African and Native American ancestry inside European American genomes#

The colonial population did not mix only with itself. Bryc and colleagues, working with a large consumer dataset, found that roughly 3.5 percent of self-identified European Americans carried 1 percent or more African ancestry, with the highest rates in South Carolina and Louisiana, and that a smaller group carried Native American ancestry, concentrated in the same southern states and in the Southwest. The reverse is also true: African American genomes average about a quarter European ancestry. The Catoctin Furnace study by Harney and colleagues showed how concretely those connections can be traced, linking 27 people enslaved at a Maryland ironworks to more than 40,000 living relatives in a consumer database.

For an ancient-DNA model this matters in a specific way. A small African or Native American share is a source that no European ancient population can absorb. A model built only from European sources will either reject the genome or push the missing ancestry into whichever source is least wrong. The correct response is to add the source and test again, which a formal model can do and a percentage calculator cannot.

How qpAdm models a genome mixed from several countries#

Here is the part that surprises people. A qpAdm model of a European American genome does not need to know which countries contributed, because it does not model countries. It models the deep sources: a hunter-gatherer population, a Neolithic farmer population, a steppe population, each represented by ancient genomes, and it asks whether the target genome is a mixture of those sources within statistical error. An English, German and Swedish blend and a pure Danish genome can both pass that test, with different proportions and honest standard errors.

The later eras work the same way. A proximal model offers sources like Iron Age Britain, Roman-era continental Europe, early medieval Saxons and Norse, and the model reports which combination is required. A genome with a Sicilian grandparent will demand a southern source that a New England genome does not, and the p-value says so. Country labels are not inputs. They emerge, approximately, from the weights.

Reading your own results#

• G25 distances: the modern era usually lists several northwestern European averages within a hair of each other. Read the cluster, not the winner; the rank order among close neighbours is noise.
• Admixture fits: the deep-era trio in northwest European ratios for most colonial genomes, drifting southward with Italian ancestry and eastward with Polish. Within-era splits between English, Dutch and German sources trade weight freely.
• qpAdm: a well-sampled target. The productive questions are the marginal ones: is a Scandinavian source required, does a southern European source improve the fit, does a small non-European source need to be added. Each has a formal answer with error bars, and the European model sets cover the standard cases.

Frequently asked questions#

What are the ancient origins of European American DNA?#

The same three prehistoric populations as any European genome: hunter-gatherers who recolonised Europe after the Ice Age, farmers who spread from Anatolia after 7000 BC, and steppe pastoralists whose descendants reset northern and western Europe after 3000 BC. The colonial blend changed the ratios, not the sources.

Which European countries do European Americans descend from?#

Historically England, Scotland, Ulster, the German Rhineland, the Netherlands and Scandinavia in the colonial period, then Ireland, Germany, Italy, Poland and the rest of eastern and southern Europe in the nineteenth century. Any one genome is a subset of that list, and a formal model recovers the deep sources rather than the passport.

Why does my ethnicity estimate change between companies?#

Because "British and Irish", "Germanic" and "Scandinavian" describe one genetic continuum split three ways, and each company draws the lines differently. Ethnicity estimates versus ancient-DNA models explains why the ancient-era answer is more stable than the modern one.

Do European Americans have Native American or African ancestry?#

Some do. Bryc and colleagues found roughly 3.5 percent of self-identified European Americans with 1 percent or more African ancestry, most often in the South, and a smaller share with Native American ancestry. Whether a particular genome does is a testable question, and a formal model either requires that source or does not.

Terms used here are defined in the glossary.

Sources and further reading#

1. Bryc, K. et al. (2015). The genetic ancestry of African Americans, Latinos, and European Americans across the United States. American Journal of Human Genetics, 96.
2. Novembre, J. et al. (2008). Genes mirror geography within Europe. Nature, 456.
3. Haak, W. et al. (2015). Massive migration from the steppe was a source for Indo-European languages in Europe. Nature, 522.
4. Olalde, I. et al. (2018). The Beaker phenomenon and the genomic transformation of northwest Europe. Nature, 555.
5. Gretzinger, J. et al. (2022). The Anglo-Saxon migration and the formation of the early English gene pool. Nature, 610.
6. Harney, É. et al. (2023). The genetic legacy of African Americans from Catoctin Furnace. Science, 381.
7. Han, E. et al. (2017). Clustering of 770,000 genomes reveals post-colonial population structure of North America. Nature Communications, 8.