Enhancements to the ADMIXTURE algorithm for individual ancestry estimation.

摘要: The estimation of individual ancestry from genetic data has become essential to applied population genetics and epidemiology. Software programs for calculating estimates have tools in the geneticist's analytic arsenal. Here we describe four enhancements ADMIXTURE, a high-performance tool estimating ancestries allele frequencies SNP (single nucleotide polymorphism) data. First, ADMIXTURE can be used estimate number underlying populations through cross-validation. Second, individuals known exploited supervised learning yield more precise estimates. Third, by penalizing small admixture coefficients each individual, one encourage model parsimony, often yielding interpretable results datasets or with large numbers ancestral populations. Finally, exploiting multiple processors, analyzed even rapidly. described make accurate, efficient, versatile estimation.