A parameter to quantify the degree of genetic mixing among individuals in hybrid populations

Hybridization between genetically distinct taxa is a complex evolutionary process. One challenge to studying hybrid populations is quantifying the degree to which non-native genes have become evenly mixed among individuals in the population. In this paper, we present a variance-based parameter, m(d), that measures the degree to which non-native genes are evenly distributed among individuals in a population. The parameter has a minimum value of 0 for populations in which individuals from multiple taxa are present but have not interbred, and a maximum value of 1 for populations in which all individuals have the same amount of non-native ancestry. A recurrence equation showed that relatively few generations of random mating are required for md to approach 1 (indicating a well-mixed population), and that md is independent of initial amounts of non-native ancestry. The parameter is mathematically equivalent to F-ST and we show how existing formulae for F-ST can be used to estimate md when diagnostic loci are available. Computer simulations showed this estimator to have very little bias for realistic amounts of data.