Figure 4.

Effects of facultative but not very low rates of sex (i.e., $\sigma \gg 1/N$) on estimates of meiotic crossing over. (A) Decay of linkage disequilibrium over 900 sites, as a function of distance between two sites. Different colors denote individual rates of sex, as shown in the legend. Solid line is the mean value over 1000 simulations; fainter curves represent 95% confidence intervals. A total of 50 paired samples were simulated (100 samples in total), $N=10,000,$ scaled mutation rate $\theta =4N\mu =10,$ scaled crossover rate during sex $R=40.$ (B) As in (A) but instead shows results from obligate sex simulations ran using ms, using a crossover rate equal to $40\sigma$ as shown in the legend. Due to binning of samples, $r^2$ is shown for distances between 25 and 875 sites apart in (A and B). (C) Estimates of R using LDhat, as a function of the effective crossover rate used in the facultative-sex coalescent simulation. Points are mean estimates from 1000 simulations, bars are 95% confidence intervals. Black line denotes $y=x;$ dashed red line is the linear regression fit. (D) Plot of all simulation results in (A) but instead as a function of the rescaled recombination rate $4Nc\sigma$ (plotted on a natural log scale) and after omitting polymorphisms with minor allele frequency <10%. Dotted lines show analytical expectations (Equation 6). Note the different y-axis scale compared to (A).