Association of genetic traits to estimated haplotypes from SNP genotypes using EM algorithm and Markov chain Monte Carlo techniques

Item Type:	Article
Title:	Association of genetic traits to estimated haplotypes from SNP genotypes using EM algorithm and Markov chain Monte Carlo techniques
Creators Name:	Rohde, K. and Fuerst, R.
Abstract:	In order to find association of genetic traits to some haplotypes from closely spaced multilocus phase-unknown genotypes we use a three-stage approach. Haplotype frequencies and the most likely haplotype pair for each individual are estimated from random samples of individual or small (nuclear) family genotypes via an EM algorithm. If the most likely haplotype pair configuration of the whole sample outweighs the less likely ones, we may consider the estimated haplotypes as alleles of a multi-allelic marker and carry out the conventional statistics, TDT or ANOVA for quantitative traits. If the most likely haplotype pair configuration and the less likely ones do not differ much in their weight, we sample the TDT or ANOVA statistic over all haplotype pair configurations using the Metropolis-Hastings algorithm. Applications of our method to simulated data as well as real data are given.
Keywords:	Association to Quantitative Traits, Haplotype Estimation, Markov Chain Monte Carlo, Multi-locus SNP Haplotypes
Source:	Human Heredity
ISSN:	0001-5652
Publisher:	Karger
Volume:	56
Number:	1-3
Page Range:	41-47
Date:	1 January 2003
Official Publication:	https://doi.org/10.1159/000073731
PubMed:	View item in PubMed

Repository Staff Only: item control page