Figure 6

Influence of admixture clines on isolation-by-distance patterns. The purest genotypes apparent from inferred admixture values in STRUCTURE were sorted into bins. Eels with extreme Q-values were selected as representatives of the two “pure” gene pools (Q < 0.1, and Q > 0.9 for American and European eels, respectively). The distribution of “admixed” Q-values was inferred from remaining intermediate genotypes, and served to inform a random sampling strategy to draw alleles, using multinomial sampling, from the respective “pure” gene pools according to abovementioned “admixed” Q-proportions. We augmented the proportion of virtually created “admixed” individuals in a stepwise process by (A) 4%; (B) 3%; (C) 2% and (D) 1% per population for a total of 12 virtual populations each. Significance of IBD was tested using the Mantel statistics for correlated genetic data [84]. To test our hypothesis, that IBD patterns can be generated in European eels by increasing levels of gene flow from South to North, the rectangular matrix of pairwise geographical distances from Wirth and Bernatchez [39] was superimposed on the genetic pairwise D _CE chord distances among the 12 virtual populations. Thus, assuming a linear increase of gene flow, we attributed the Southern-Eastern-most location (River Tiber) the lowest, and the Northern-Western-most locality (Iceland) the highest hybridization rate. Intermediate levels were attributed in ascending order along the European coastline.