ASP.Bayesian: Samples in the posterior distribution of the frequencies and...

In ASPBay: Bayesian Inference on Causal Genetic Variants using Affected Sib-Pairs Data

Description

Samples using Metropolis-Hasting Algorithm in the posterior distribution of the four haplotype frequencies and OR

Usage

 ASP.Bayesian(N, Tem_Gen, Index_Gen, IBD, snp, thin = 1, sd.freq = 0.05,
                     sd.psi = 0.05, p0 = c(rep(1/4, 4), 1), psi.prior = 0) 

Arguments

N	Number of Metropolis-Hastings iterations
Tem_Gen	Genotypes of controls (denoted by the number of alternative allele)
Index_Gen	Genotypes of index cases
IBD	IBD states for each affected sib-pair
snp	Names or number column of the SNP to consider
thin	Thinning parameter (keep only every thin-th draw)
sd.freq	Random walk standard deviation of the frequency logarithms
sd.psi	Random walk standard deviation of the OR
p0	The initial point of random walk
psi.prior	Precision of gaussian log(OR) prior (0 = improper flat prior)

Details

Samples using Metropolis-Hasting and likelihood defined by data. More precisely, give the frequency samples of haplotypes for observed SNP and unobserved causal SNP and give the sample of the odds ratio associated to the causal SNP.

Value

List of 5 vectors of length N/thin with components:

f_ab	Sample of the haplotype composed by the two alternative alleles
f_Ab	Sample of the haplotype composed by the reference allele for the causal (unobserved) locus and the alternative alleles for the observed locus
f_aB	Sample of the haplotype composed by the alternative allele for the causal (unobserved) locus and the reference alleles for the observed locus
f_AB	Sample of the haplotype composed by the two reference alleles
OR	Sample of the OR

Author(s)

Claire Dandine-Roulland

References

Dandine-Roulland, Claire and Perdry, Herve. Where is the causal variant? On the advantage of the family design over the case-control design in genetic association studies. Submitted to Eur J Hum Genet

See Also

ASP.Selection, Graphs.Bayesian

Examples

data(ASPData)
B <- ASP.Bayesian(1e5, ASPData$Control, ASPData$Index,
                  ASPData$IBD, 15 )

ASPBay documentation built on May 2, 2019, 8:55 a.m.