"Sephardic" is a word that describes a liturgy, a language and a history, and it is stretched over at least three different genetic stories. There are the communities that left Iberia after 1492 and settled in the Ottoman Empire, in Salonica, Istanbul, Izmir, Sofia and Sarajevo. There are the communities of Morocco, Algeria and Tunisia, where Iberian exiles joined much older Jewish populations. And there are people in Spain, Portugal and Latin America who descend from converts who stayed, and who often come to genetic testing with exactly that question. This post covers what ancient DNA can say about each, and what a qpAdm model does with a Sephardic genome. It is part of the series introduced in Jewish ancestry and ancient DNA.
As everywhere in the series: ancestry is not identity, and nothing here is a statement about who is Jewish.
The shared base#
The genome-wide studies of living Jewish communities found that Turkish, Bulgarian, Greek and Italian Sephardic Jews cluster with the other Mediterranean and Middle Eastern Jewish communities, close to Ashkenazi Jews and to Levantine populations, and not with Spaniards or Turks (Behar et al. 2010; Atzmon et al. 2010). The Levantine component of that ancestry is, as far as the ancient genomes can tell, the same Bronze Age Canaanite ancestry that underlies every Levantine population, a mixture of the local Levantine Neolithic lineage with a substantial Iranian- or Caucasus-related layer (Haber et al. 2017; Agranat-Tamir et al. 2020). It is described in the post on the Canaanite source population.
On that base, the different Sephardic communities carry different admixture. The Ottoman communities carry southern European ancestry that is difficult to separate between Iberian, Italian and Balkan origins, because all three are Mediterranean populations with largely Anatolian-farmer ancestry and modest steppe input. The North African communities carry, in addition, a Maghrebi component and are on average closer to Ashkenazi and Ottoman Sephardic Jews than to their Berber and Arab neighbours, which reflects the arrival of the Iberian exiles into older communities that had themselves stayed largely endogamous (Campbell et al. 2012).
The converso question#
A recurring reason people with Spanish, Portuguese, Mexican, Colombian or Brazilian ancestry order an ancient-DNA analysis is a family tradition of Jewish descent. It is worth being clear about what the genetics can and cannot support.
A Y-chromosome study estimated that around a fifth of Iberian paternal lineages could be assigned to a Sephardic-like origin (Adams et al. 2008), a figure that has been much debated because the haplogroups involved (chiefly J and E1b1b) are also common across the Mediterranean for reasons that have nothing to do with Jewish history. A haplogroup alone is never evidence of a share of ancestry, and it is certainly not evidence of a religious history. At the genome-wide level, a Levantine component in an Iberian genome is real and measurable in aggregate, but it has several possible sources: Phoenician and Punic settlement, the Roman-era eastern Mediterranean, the Islamic centuries, and Jewish communities, all of which drew on a similar Levantine pool. A model can tell you that your genome carries a Levantine share and how large it is. It cannot tell you which of those histories delivered it.
That is not a reason to skip the analysis. It is a reason to read the result as what it is: a measurement of ancestry, with a standard error, that is consistent with several histories. The family tradition then stands or falls on documents, not on DNA.
What a qpAdm model does with a Sephardic genome#
qpAdm writes your genome as a mixture of ancient source populations, tested against a set of outgroups, and returns a weight, a standard error and a Z-score for every source and a p-value for the model that can reject it. The method is described in Understanding qpAdm and the numbers in the reading guide. For a Sephardic genome the two eras of the report usually resolve like this.
Hunter-Gatherer and Neolithic Farmer era. The Anatolian and Iranian farmer sources and the Natufian source carry most of the weight, as they do for every Jewish community. The European admixture shows as Western Hunter-Gatherer and Western Steppe Herder shares, and for a North African community a North African Farmer or Iberomaurusian share can appear.
Classical Antiquity era. This is where the three Sephardic histories separate, and where the analyst's search is spent. The starting model is Canaanite beside a southern European source. The questions are then:
- Which European source fits: the Iron Age Iberian population, Imperial Italy, or the Aegean? These are close in genetic space, and separating them depends on the file's coverage. The report shows the standard error that tells you how well they were separated.
- Does an Indigenous Berber source earn a place? For a Moroccan, Algerian or Tunisian Jewish genome it very often does; for an Istanbul or Salonica one it usually does not, and a model that carries it anyway is rejected by the nested-model check.
- Does the Levantine share prefer the Canaanite source to the Iron Age Phoenician or Roman-era Eastern Mediterranean references? For most Jewish genomes it does, which is one reason the Canaanite source was added; for an Iberian genome with a Punic history the Phoenician source may fit as well or better, and that difference is itself informative.
Each of those comparisons is a separate run with its own p-value, and only the model that survives all of them is published. The Model Lab lets you re-run them yourself on your own merged sample afterwards.
What the model does not say#
- It does not tell you whether your Levantine ancestry came through a Jewish community, a Phoenician port or the Islamic centuries. The sources are ancient populations, and all three histories drew on the same ones.
- It does not identify a family, a surname or a town. Those are questions for records and for Ancient Matches and relative-finding tools, not for an admixture model.
- It does not measure Jewishness, and it is not evidence for or against anyone's status.
Related reading#
- Jewish ancestry and ancient DNA: the series introduction
- Ashkenazi Jewish DNA, the community Sephardic genomes are most often compared with
- Phoenician and Punic DNA across the Mediterranean, for the other Levantine history of Iberia
- The Canaanite source population
- Why qpAdm for Jewish genomes
References#
- Adams, S. M. et al. (2008). The genetic legacy of religious diversity and intolerance: paternal lineages of Christians, Jews, and Muslims in the Iberian Peninsula. American Journal of Human Genetics, 83(6), 725–736.
- Agranat-Tamir, L. et al. (2020). The genomic history of the Bronze Age Southern Levant. Cell, 181(5), 1146–1157.
- Atzmon, G. et al. (2010). Abraham's children in the genome era. American Journal of Human Genetics, 86(6), 850–859.
- Behar, D. M. et al. (2010). The genome-wide structure of the Jewish people. Nature, 466, 238–242.
- Campbell, C. L. et al. (2012). North African Jewish and non-Jewish populations form distinctive, orthogonal clusters. Proceedings of the National Academy of Sciences, 109(34), 13865–13870.
- Haber, M. et al. (2017). Continuity and admixture in the last five millennia of Levantine history from ancient Canaanite and present-day Lebanese genome sequences. American Journal of Human Genetics, 101(2), 274–282.
