The Gibbs and splitmerge sampler for population mixture analysis from genetic data with incomplete baselines

摘要: Although population mixtures often include contributions from novel populations as well baseline previously sampled, unlabeled mixture individuals can be separated to their sources genetic data. A Gibbs and split–merge Markov chain Monte Carlo sampler is described for successively partitioning a sample into plausible subsets of each the extra-baseline present. The are selected satisfy Hardy–Weinberg linkage equilibrium conditions expected large, panmictic populations. number present inferred distribution counts per partition drawn by sampler. To further summarize sampler's output, co-assignment probabilities same computed partitions used construct binary tree relatedness. graphically displays clusters together with quantitative meas...