Combining genetic, historical and geographical data to reconstruct the dynamics of bioinvasions: application to the cane toad Bufo marinus

摘要: We developed a spatially explicit model of bioinvasion and used an approximate Bayesian computation (ABC) framework to make various inferences from combination genetic (microsatellite genotypes), historical (first observation dates) geographical (spatial coordinates introduction sampled sites) information. Our method aims discriminate between alternative scenarios estimate posterior densities demographically relevant parameters the invasive process. The performance our landscape-ABC is assessed using simulated data sets differing in their information content (genetic and/or data). apply methodology recent spatial expansion cane toad, Bufo marinus, northern Australia. find that, at least context toad invasion, are more informative than for discriminating scenarios. However, provides most accurate estimates demographic parameters. For we some evidence strong bottleneck prior introduction, small initial number founder individuals (about 15), large population growth rate 400% per generation), standard deviation dispersal distance 19 km generation high invasion speed equilibrium (50 year). approach strengthens application ABC field by allowing statistical be made on dynamics species sources