Spatial genetic patterns generated by two admixing genetic lineages: a simulation study

Two formerly geographically separated lineages of the zebra mussel Dreissena polymorpha had been given the opportunity to mix extensively across the newly built German Main–Danube canal. We had monitored this admixture of mussel lineages and had described spatial patterns in different genetic measures [Müller et al. 2001 (Heredity, 86: 103); 2002 (Proc. R. Soc. Lond., 269: 1139)]. Here, we present an individual-based model to assess the potential of spatial genetic patterns of detecting and quantifying admixture of mussel lineages. Genetic measures studied are (1) allele frequencies, (2) deviations from Hardy–Weinberg expectations of loci (deficit of heterozygotes, HWD) and (3) linkage disequilibria between unlinked loci (LD). For allele frequencies, we observed a cline over the zone of admixture in all simulations of mixing mussel lineages suggesting that these are appropriate for verification of their mixture. The point of the first contact between lineages was always detectable from their intermediate allele frequencies. LD and HWD were only spatially informative for diagnostic loci or loci with very strong differences in allele frequencies of lineages. For such loci, the probability of disequilibria was highest where lineages had met and decreased towards both sources of lineages Main and Danube. The overall probability of detecting any disequilibrium was higher for LD than for HWD and increased with an increasing rate of genetic interchange. Our simulation results are corroborated by our zebra mussel data and studies from literature. They are applicable to any case of two known linearly mixing genetic lineages.