Performs the two-step approach for detecting selection signatures using genomic data from diploid individuals and biallelic markers developed by Gianola et al. (2010) and modifications to carry out inference based on the Laplace approximation and to incorporate pedigree information using the likelihood derived by Martínez et al. (2017).
- Implements relatively simple Bayesian approaches to infer selection signatures via
Wright’s
$F_{ST}$ in diploid organisms using data from biallelic molecular markers. - Two models and two methods to compute de posterior mean of the are available, resulting in four approximations.
- The basic method is the original one developed by Gianola et al. (2010), the package also implements a variant using the Laplace approximation instead of Monte Carlo integration to compute the posterior mean.
- The other method extents the original Bayesian model to infer allele frequencies by incorporating pedigree information via a modification of the probability mass function derived by Martínez et al. (2017), for this approach, the posterior mean can be computed using the Laplace approximation or Monte Carlo integration.
- Allows using arbitrary model hyperparameters and even posing a different set of hyperparameters for each subpopulation.
To install the development version from GitHub repository:
# install.packages("devtools")
devtools::install_github("edserranoc/SSBayesHelena")The SSBayesHelena package as a whole is licensed under the GPLv3.0. See the LICENSE file for more details.
The Corporación Colombiana de Investigación Agropecuaria-Agrosavia and Universidad Nacional de Colombia provided financial support for this project, which is warmly acknowledged.