Figure 7.

Admixture proportions in chromosomal regions of partial sweeps

Mean admixture proportions of ECHs (purple) and in remaining haplotypes (black) are computed separately for each 100 kb window. Error bars with wide caps designate the standard 95% confidence intervals obtained from 10,000 bootstrapping iterations. Error bars with narrow caps designate Jeffrey’s binomial confidence interval, which better represents confidence when frequencies are zero or very low (in computing this interval, we assume that the sample size equals the number of 100 kb sequence windows). The dotted line represents zero admixture. The dashed line shows the mean admixture proportion on the X chromosome.

(A) Mean admixture proportions of haplotypes from each geographical region shown as labels on the x axis.

(B) Admixture proportions at each individual ECH. Here, x axis labels represent chromosomal positions where each ECH has the highest frequency (peaks in Figure 4). For legibility, one outlier is shown separately.