Figure 5.

Effect of the scoring parameters on the PSMix clustering analysis. Scatterplot of the number of bins (nbin, log-scaled y-axis), according to the mean homoplasy rate (HR, x-axis). The mean homoplasy rate is defined as the average number of peaks belonging to the same individual that are affiliated within the same bin. The eleven datasets are displayed as dots (in dark grey, GeneMapper datasets labelled with "GM" as prefix; in light grey, RawGeno datasets labelled with "RG" as prefix and in white, manual dataset) and as maps on the scatterplot. The maps represent C. uniflorum populations with pie-charts representing assignment probabilities obtained with PSMix (K = 5 groups). On the upper-right corner: geographical location of the sampling and names of the identified phylogenetic groups (violet – Meridional cluster, blue – Occidental cluster, red – Central cluster, green – Oriental cluster and yellow – Cryptic cluster). The mean homoplasy rate could not be calculated for the GeneMapper datasets.