HAPLOPOOL: improving haplotype frequency estimation through DNA pools and phylogenetic modeling

In the context of disease association studies, haplotype frequencies are usually estimated from genotype data. In order to reduce genotyping costs, one can estimate the haplotype frequencies from DNA pools. Here we present a method, HaploPool, which estimates the haplotype frequencies in a population from a set of DNA pools of l individuals each, where l is a small number (typically two or three). HaploPool is based on a combination of a perfect phylogeny model, together with the EM algorithm on subsets of the SNPs. Using HaploPool, we study the trade-off between DNA pooling and the accuracy of haplotype frequency estimation. We show that the accuracy of frequency estimates obtained by HaploPool on a set of n DNA pools of two individuals each, is roughly the same as the accuracy

• Type: Journal
• Year: 2007
• Where: BIOINFORMATICS
• Authors: Bonnie Kirkpatrick, Carlos Santos Armendariz, Richard M. Karp, Eran Halperin
• Keywords: BIOINFORMATICS 2007, DNA Pools, Frequencies, Haplotype