Figure 1.
Ancestry estimates using STRUCTURE program for Kuwaiti samples.
(Best model at K= 3). The three groups comprise 138, 63, and 72 individuals, respectively. For simplicity, we label the groups as Kuwait 1, Kuwait 2 and Kuwait 3 (from left to right). Mantel test correlation between pairs of all 4 separate runs (see Materials and Methods) is >0.99 with p-value = 1e-04 over 9999 replicates.
Figure 2.
Ancestry composition of Kuwaiti groups using HGDP data.
STRUCTURE results for the combined data set of three Kuwaiti groups and representative HGDP populations from West Asia (Brahui), Middle East (Bedouin, Druze and Palestinian), Mozabite (North Africa), Europe (French), and sub-Saharan Africa (Yoruba). Best Model for the combined data set is at K = 9. Structure results for the combined data set of three Kuwaiti groups and all of the HGDP populations are given in Figure S1. Red: French_Basque (Europe), Green: Bedouin (Arabs), Dark Green: Kalash (Asia), Cyan: Yoruba (sub-Saharan Africa), Blue: Druze (Persian) and Gray: Brahui as inferred from Figure S1. Black lines partition the groups.
Figure 3.
Length distributions of IBD and ROH segments shared among all of the Kuwaiti individuals (with Froh > 0.0625 or Fibd > 0.0625, as the case may be), and expected length distributions at different levels of inbreeding (6 or 9 generations since common ancestor).
Considered are only those segments of length ≥ 1 cM. Assuming Haldanes’s recombination model, the length of segments should follow an exponential distribution with the mean as [1 / (2 × Number of generations since common ancestor)] in Morgans. The figure illustrates that the Kuwaiti population, as a whole, share a recent common ancestor within 6-9 generations. Figure S2 gives the distributions of IBD and ROH segments shared among individuals within each of the three Kuwaiti groups.
Figure 4.
Plot of total amount of ROH versus total number of ROH segments in the three Kuwaiti groups and in representative HGDP populations.
Kuwaiti groups exhibit a range of homozygosity showing characteristics of both consanguineous and non-consanguineous mating patterns. See Figure S5 for a representation for all of the Kuwait individuals in the background of all of the HGDP populations.
Figure 5.
Scatter plot representing the first two principal components of merged data sets of the three Kuwaiti groups and representative HGDP populations
The first principal component value ranges from 3.505-3.537 and the second Component value ranges from 2.688-2.710 in multiple iterations of PCA. See Figure S6 for similar plot derived by including Yoruba population.
Figure 6.
Plot (Phylogenetic tree) of pairwise FST distances for the three Kuwaiti groups with a set of representative HGDP populations that cover Europe, Asia, and Middle East.
Bootstrap confidence – Red: 70% to 80%; Blue -80% to 90%; Green -90% to 100%.
Figure 7.
Linkage disequilibrium (LD) decay across the genomes of the Kuwaiti groups and other representative populations from the HGDP data set.