Applications of Bayesian Skyline Plots and Approximate Bayesian Computation for Human Demography

Bayesian methods have been adopted by anthropologists for their utility in resolving complex questions about human history based on genetic data. The main advantages of Bayesian methods include simple model comparison, presenting results as a summary of probability distributions, and the explicit inclusion of prior information into analyses. In the fijield of anthropological genetics, for example, implementing Bayesian skyline plots and approximate Bayesian computation is becoming ubiquitous as means to analyze genetic data for the purpose of demographic or historic inference. Correspondingly, there is a critical need for better understanding of the underlying assumptions, proper applications, and limitations of these two methods by the larger anthropological community. Here we review Bayesian skyline plots and approximate Bayesian computation as applied to human demography and provide examples of the application of these methods to anthropological research questions. We also review the two core components of Bayesian demographic analysis: the coalescent and Bayesian inference. Our goal is to describe their basic mechanics in an attempt to demystify them