Figure 2.

Population structure for European, Jewish, and Middle Eastern populations, inferred with unsupervised clustering. The number of predefined clusters (K) is indicated to the left of each plot. Each individual is represented by a thin vertical line that is partitioned into K colored components according to the inferred membership in K genetic clusters. For each K, only the major mode is shown (obtained in 39 of 40 replicates for K = 2, 40 of 40 replicates for K = 3, 15 of 40 replicates for K = 4, 40 of 40 replicates for K = 5, and 20 of 40 replicates for K = 6).

Kopelman et al. BMC Genetics 2009 10:80   doi:10.1186/1471-2156-10-80
Download authors' original image