scatR is an R package that implements smoothed and continuous assignment testing (SCAT) of genetic samples. These methods allow a researchers to infer smoothed allele frequencies across space using a Bayesian Gaussian process model fit to samples with known origin. Additionally, the package implements methods for the probabilistic estimatation of the spatial origin of samples using their genotypes.
The package can be installed in the standard way R CMD INSTALL scatR. Note that this package depends on a reasonably current install of Boost, anything >=1.47 should be fine. If you want to take advantage of the GPU accelerated features both Magma (>=1.1) and CUDA (>=4.3) must be installed along with a NVIDIA graphics card with a compute capability of 2.0 or better.
Documentation is forthcoming.
This software is based on the SCAT software written by Matthew Stephens. Our implementation was developed in collaboration with John Novembre and the Center for Tropical Research at UCLA. This work was supported by National Science Foundation grant 0933731.