close
Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Dec 8;185(25):4703-4716.e16.
doi: 10.1016/j.cell.2022.11.002. Epub 2022 Nov 30.

Genome-wide data from medieval German Jews show that the Ashkenazi founder event pre-dated the 14th century

Affiliations

Genome-wide data from medieval German Jews show that the Ashkenazi founder event pre-dated the 14th century

Shamam Waldman et al. Cell. .

Abstract

We report genome-wide data from 33 Ashkenazi Jews (AJ), dated to the 14th century, obtained following a salvage excavation at the medieval Jewish cemetery of Erfurt, Germany. The Erfurt individuals are genetically similar to modern AJ, but they show more variability in Eastern European-related ancestry than modern AJ. A third of the Erfurt individuals carried a mitochondrial lineage common in modern AJ and eight carried pathogenic variants known to affect AJ today. These observations, together with high levels of runs of homozygosity, suggest that the Erfurt community had already experienced the major reduction in size that affected modern AJ. The Erfurt bottleneck was more severe, implying substructure in medieval AJ. Overall, our results suggest that the AJ founder event and the acquisition of the main sources of ancestry pre-dated the 14th century and highlight late medieval genetic heterogeneity no longer present in modern AJ.

Keywords: Ashkenazi Jews; IBD sharing; admixture; ancient DNA; demographic inference; founder event; mitochondrial DNA; pathogenic variants; population structure; runs of homozygosity.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests S.C. is a paid consultant and holds stock options at MyHeritage. D.B. is an employee and shareholder at The Janssen Pharmaceutical Companies of Johnson & Johnson. É.H. is an employee of 23andMe.

Figures

Figure 1.
Figure 1.. The medieval Jewish cemetery at Erfurt
(A) The layout of the cemetery. The inner city wall and the outer city wall are at the bottom and top of the map, respectively. Family members are marked in red ellipses (see next). (B) The pedigrees of the two families identified based on first degree relationships. Black symbols represent individuals with DNA; gray symbols represent inferred family members. Circles: females; squares: males. For each individual, we indicate the ID, the number of genotyped SNPs, the estimated age at death, and the 14C date (95.4% probability intervals). See also Figure S1.
Figure 2.
Figure 2.. Principal components analysis
We learned the principal components (PCs) using West Eurasian populations (Lazaridis et al., 2014) and projected the Erfurt individuals (filled red circles) onto the inferred axes. Modern Ashkenazi Jews (green squares), Jews of non-Ashkenazi origin (pink shapes), and Mediterranean populations (teal shapes) are highlighted. The inset zooms in on the region that contains AJ individuals. See also Figure S2.
Figure 3.
Figure 3.. Models for the ancestry of Erfurt Ashkenazi Jews
(A) Each qpAdm model for the ancestry of Erfurt Jews includes a Middle Eastern, a Southern European, and an Eastern European (Russians) source. The Southern European source was either South or North Italians, as indicated at the top of each panel. The Middle Eastern source is indicated in the x axis labels. Only models with qpAdm p value >0.05 in the main analysis and in the robustness tests are shown (Table S3). Error bars represent one standard error in each direction. qpAdm p values are presented above each model. (B) The ancestry of single Erfurt individuals, labeled by their IDs. We used qpAdm with Russian, Lebanese, and South Italian sources. The individuals are labeled by their Erfurt subgroup (EU/ME). qpAdm p values are shown for each individual. Results are not shown for low-coverage individuals (<50k SNPs), and for an additional individual who could not be modeled using these sources (p < 0.05). (C) A plot of δ13Cenamel and δ18Oenamel stable isotope ratios for a subset of 20 Erfurt individuals with >200k SNPs. The Erfurt subgroup affiliation (EU/ME) is color-coded (legend). See also Figure S3.
Figure 4.
Figure 4.. Models for AJ demographic history based on ancient and modern haplotypes
(A) A single-population model for the demographic history of AJ, inferred based on modern IBD sharing (Table S5, model [A]). In the diagram, the y axis represents the time in generations before present (gbp) and the width is (schematically) proportional to the effective population size. The 95% confidence intervals (CI) were computed using bootstrapping and are indicated near each parameter. (B) The mean number of IBD and ROH segments (per pair of haploid autosomal genomes) in modern AJ vs. segment length (11 bins between 4 and 15 cM). Each symbol (circles for IBD, triangles for ROH) is placed at the middle of its corresponding bin. The red line shows the expected number of segments per bin based on the demographic model of (A). (C) The total length of ROH segments in 16 EAJ individuals with >400k SNPs. The bars are colored proportionally to the contribution of segments of different lengths (legend). (D) ROH counts in EAJ (circles) and the expected number based on various models (lines). The model inferred using modern IBD (panel [A]) is in red. The same model but allowing consanguinity in EAJ (Table S5, model [D]) is in green. Both models fit poorly to short ROH segments. A model similar to (A) but with either a narrower or a longer bottleneck (Table S5, models [E] and [F], respectively) are in purple and teal, respectively. (E) A two-population model inferred jointly using IBD in MAJ and ROH in EAJ (Table S5, model [H]). According to the model, the ancestral population split Tb = 46 generations ago into one population of effective size Nb = 627 (representing Erfurt, indicated in red) and another of size NaNb. At the end of the bottleneck, the two populations merged with proportions f = 52% and 1 − f, respectively, and expanded exponentially as in the single-population model. The time of sampling of the Erfurt population is shown at 26 generations ago (assuming 25 years per generation). (F) Counts of IBD segments in MAJ (green circles; same data as in [B]) and ROH in EAJ (black circles; same data as in [D]), and the expectations (lines) based on the two-population model of (E) (MAJ, red; EAJ, pink). See also Figures S4 and S5.

References

    1. Agranat-Tamir L, Waldman S, Martin MAS, Gokhman D, Mishol N, Eshel T, Cheronet O, Rohland N, Mallick S, Adamski N, et al. (2020). The genomic history of the bronze age southern levant. Cell 181, 1146–1157.e11. - PMC - PubMed
    1. Alexander DH, Novembre J, and Lange K (2009). Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 19, 1655–1664. - PMC - PubMed
    1. Amorim CEG, Vai S, Posth C, Modi A, Koncz I, Hakenbeck S, La Rocca MC, Mende B, Bobo D, Pohl W, et al. (2018). Understanding 6th-century barbarian social organization and migration through paleogenomics. Nat. Commun 9, 3547. - PMC - PubMed
    1. Aneli S, Caldon M, Saupe T, Montinaro F, and Pagani L (2021). Through 40, 000 years of human presence in Southern Europe: the Italian case study. Hum. Genet 140, 1417–1431. - PMC - PubMed
    1. Antonio ML, Gao Z, Moots HM, Lucci M, Candilio F, Sawyer S, Oberreiter V, Calderon D, Devitofranceschi K, Aikens RC, et al. (2019). Ancient Rome: A genetic crossroads of Europe and the Mediterranean. Science 366, 708–714. - PMC - PubMed

LinkOut - more resources