The goal of poolABC is to help users perform model choice and parameter inference using Pool-seq data integrated into an Approximate Bayesian Computation (ABC) framework. This package provides functions to simulate Pool-seq data under models of ecotype formation and to import Pool-seq data from real populations. Two ABC algorithms are implemented to perform parameter estimation and model selection using Pool-seq data. Cross-validation can also be performed to assess the accuracy of ABC estimates and calculate model misclassification.


You can install the development version of poolABC from GitHub with:

# install.packages("devtools")

Alternatively, you can install the current stable version of poolABC from CRAN: