popanc

This program uses discriminant analysis and Markov chain Monte Carlo to infer local ancestry frequencies in an admixed population from genomic data. This is done in the context of a continuous correlated beta process model that accounts for expected autocorrelations in local ancestry frequencies along chromosomes. This method is best-suited for the analysis of admixed populations that experience little or no on-going gene flow from the parental populations, but that are still segregating for ancestry segments.