iBMQ (version 1.12.0)

eqtlMcmc: Bayesian Multiple eQTL mapping using MCMC

Description

Compute the MCMC algorithm to produce Posterior Probability of Association values for eQTL mapping.

Usage

eqtlMcmc(snp, expr, n.iter, burn.in, n.sweep, mc.cores, write.output = TRUE, RIS = TRUE)

Arguments

snp

SnpSet class object

expr

ExpressionSet class object

n.iter

Number of samples to be saved from the Markov Chain

burn.in

Number of burn-in iterations for the Markov Chain

n.sweep

Number of iterations between samples of the Markov Chain (AKA thinning interval)

mc.cores

The number of cores you would like to use for parallel processing. Can be set be set via `options(cores=4)', if not set, the code will automatically detect the number of cores.

write.output

Write chain iterations to file. If TRUE, output for variables will be written to files created in the working directory.

RIS

If TRUE, the genotype needs to be either 0 and 1. If FALSE the genotype need to be either 1,2 and 3.

Value

A matrix with Posterior Probability of Association values. Rows correspond to snps from original snp data objects, columns correspond to genes from expr data objects.

Details

The value of mc.cores may be ignored and set to one when the iBMQ installation does not support openMP.

References

Scott-Boyer, MP., Tayeb, G., Imholte, Labbe, A., Deschepper C., and Gottardo R. An integrated Bayesian hierarchical model for multivariate eQTL mapping (iBMQ). Statistical Applications in Genetics and Molecular Biology Vol. 11, 2012.

Examples

data(phenotype.liver)
data(genotype.liver)
#PPA.liver <-  eqtlMcmc(genotype.liver, phenotype.liver, n.iter=100,burn.in=100,n.sweep=20,mc.cores=6, RIS=FALSE)