Gibbs3: Function obtains the MCMC chains for the parameters of...

Description Usage Arguments Details Value Author(s) See Also Examples

View source: R/Gibbs3.R

Description

This function provides the MCMC chains for the parameters of interest that will form their posterior distribution. This function is to obtain the gene sets that are differentially expressed among five phenotypes of interest, taking into account one as baseline.

Usage

1
Gibbs3(noRow,noCol,iter,GrpSzs,YMu,L0,V0,L0A,V0A,MM,AAPi,ApriDiffExp,result1,result2)

Arguments

noRow

Number of row of the dataset

noCol

Total number of subjects considered.

iter

Number of iterations for the Gibbs sampler.

GrpSzs

Vector with the sizes of the gene sets considered. Output from the function DataGeneSets.

YMu

Output y.mu from the MCMCDataSet

L0

Vector with the prior parameters.

V0

Vector with the prior parameters.

L0A

Vector with the prior parameters.

V0A

Vector with the prior parameters.

MM

Parameter of the prior.

AAPi

Parameter of the prior.

ApriDiffExp

Number of differentially expressed gene sets apriori

result1

Matrix for the MCMC chains for the parameter that identifies the difference in gene set expression from phenotype 1 in comparison with the phenotype chosen as baseline. The rows are for the gene sets and the columns for the number of iterations.

result2

Matrix for the MCMC chains for the parameter that identifies the difference in gene set expression from phenotype 2 in comparison with the phenotype chosen as baseline. The rows are for the gene sets and the columns for the number of iterations.

Details

This function provides the MCMC chains for the estimation of the posterior distribution for the parameters of interest for each gene set.

Value

This function returns a list with four items

alfa.1

A list with the MCMC chains for the estimation of the posterior distribution for the parameter associated with the comparison of phenotype 1 with respect to the reference phenotype.

alfa.2

A list with the MCMC chains for the estimation of the posterior distribution for the parameter associated with the comparison of phenotype 2 with respect to the reference phenotype.

Author(s)

A. Quiroz-Zarate aquiroz@jimmy.harvard.edu

See Also

See the BAGS Vignette for examples on how to use function Gibbs2. This function can also be used when the gene expression data has a time series experimental design. In this case, there will be three time points on the time course sampling. The assumption is that measurements between time points are independent. This assumption is reasonable when there is irregular and sparse time course sampling.

Examples

1
# Similar to the example on Gibbs2, but in this case there are three different phenotypes of interest. The user has to define which if the three is the reference group in order to obtain the gene groups that are differentially expressed.

BAGS documentation built on Nov. 8, 2020, 11:11 p.m.