Simultaneous parentage and dispersal estimation in monoecious plants: A Bayesian approach, combining genetic and spatial data

In plants, pollen movement and seed dispersal have a critical influence on population dynamics and levels of gene flow.  Reproductive processes, including dispersal, can be difficult observe; consequently, ecologists have begun to embrace tools such as genetic parentage analyses which promise to shed light on these processes.  While parentage analysis is often done prior to the estimation of the population-level parameters, it has been shown that joint estimation of parameters increases the power of parentage assignment.  Microsatellites are highly variable genetic markers which can often be used to distinguish all individuals in a population.  Such genetic tools can be very powerful, but there are several sources of error associated with microsatellites that must be considered.  This model explicitly treats three types of genotyping error: mistyping, allelic dropout, and null alleles.  In microsatellites, PCR errors lead to “stutter” that can cause a misreading of the allele length.  Therefore, unlike previous models, we assume that mistyping errors are more likely to occur between alleles of similar lengths.  The method does not require strong priors. For simulated test data sets, the model successfully recreated the simulated pedigrees and dispersal parameters for pollen and seed.  In a preliminary analysis of a population of northern red oaks (Quercus rubra) less than 10% of sampled seedlings had an in-plot parent.  Simulations indicated that at least 80% of sampled seedlings should have one or more parent within the 5 ha stand, suggesting that seed and pollen dispersal distances may be considerably longer than expected.