Abstract
Bayesian Skyline Plots (BSPs) are representations of the posterior probability density of the effective population size in function of time; i.e. a graphical representation of the fluctuation of the effective population size with time based on the estimates obtained from Bayesian inference. The interest of BSPs is that they allow to infer gradual changes of the effective population size without the need of a specific mathematical function determining the shape of the demographic change. For instance, a population expansion can be well characterized by a BSP either if the change had ocurred instantaneously, exponentially or logistically.. Currently, the only implementation of this analysis has been done within the MCMC-based estimation of likelihood approach and is restricted to non-recombining DNA sequence data. We have explored how to implement BSP within the approximate Bayesian computation (ABC) framework, with promising preliminary results. An implementation in ABC allows to obtain BSP from any multilocus molecular data, e.g. recombining DNA sequence data, microsatellites, AFLPs or SNPs.