Anthropology, University of Utah
Thursday All day, Park Concourse
Principal component analysis (PCA) is a method of reducing the dimensionality of complex datasets. For large, highly inter-correlated data such as the HapMap or the 1000 Genomes Project, PCA is an ideal method for revealing internal structures that cannot be seen using other methods. PCA has been widely used in population genetics to examine population history and relationships between individuals and populations across the genome. Here we combine PCA with Fst-like measures of genetic distance to examine across each chromosome in a rolling window comparison of divergence or convergence by comparing an admixed population with its source populations. The result is a chromosome-by-chromosome map that plots where the admixed population is more similar to one of the parent populations than the other. Regions of significant regional lengths of divergence may be useful in suggesting gene regions where selection has recently taken place, as the combination of two historically distinct populations is likely to lead to adaptation and selection.