Fig. 2: A: Scatterplot showing PC1-2 of a principal component analysis, B: Individual ancestry inferences based on a 3-way qpAdm model.
From: Ancient DNA reveals diverse community organizations in the 5th millennium BCE Carpathian Basin
A Among the individuals in this study, outliers (all >50 k SNPs detected, tested by Mahalanobis distance calculation combined with χ² distribution, at α = 0.05; see Methods and Supplementary Data 3) are marked on the zoomed-in parts of panel A. The two zoom-in panels focus on the Neolithic and Copper Age samples from the Carpathian Basin with comparative data, whose chronology is indicated in the figure legend. Abbreviations in the legend: N Neolithic; MN Middle Neolithic; LN Late Neolithic; MLN Middle–Late Neolithic; CA Copper Age; ECA Early Copper Age; LCA Late Copper Age; EBA Early Bronze Age; GHP Great Hungarian Plain. Source data for the scatterplot are provided in the Source Data file. B Error bars represent qpAdm estimates as centers with one standard error, calculated using the block-jackknife approach implemented in qpAdm (see Supplementary Data 5A). Individual genotypes were the units of study, used as targets in qpAdm tests. Abbreviations are as follows: WHG Western Hunter-Gatherer; EHG Eastern Hunter-Gatherer. The Anatolian Neolithic ancestry component is omitted from the chart. Grave numbers and laboratory IDs are indicated on the individual barplots. Faint bars denote qpAdm tests with p < 0.05, indicating a lack of fit. For further acceptable models, p-values and component estimates, see Supplementary Data 5.
