Inference of ancestral information in recently admixed populations, in which every individual is composed of a mixed ancestry (e.g., African Americans in the US), is a challenging problem. Several previous model-based approaches have used hidden Markov models (HMM) to model the problem, however, the Markov Chain Monte Carlo (MCMC) algorithms underlying these models converge slowly on realistic datasets. While retaining the HMM as a model, we show that a combination of an accurate fast initialization and a local hill-climb in likelihood results in significantly improved estimates of ancestry. We studied this approach in two scenarios-the inference of locus-specific ancestries in a population that is assumed to originate from two unknown ancestral populations, and the inference of allele frequencies in one ancestral population given those in another. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Sankararaman, S., Kimmel, G., Halperin, E., & Jordan, M. I. (2008). On the inference of ancestries in admixed populations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4955 LNBI, pp. 424–433). https://doi.org/10.1007/978-3-540-78839-3_37
Mendeley helps you to discover research relevant for your work.