ccParams: Read posterior means and classification probabilities from...

In BGmix: Bayesian models for differential gene expression

Description

Reads output files containing posterior means from BGmix AND reads posterior probabilities of each gene being classified in the null mixture component.

Usage

ccParams(filedir, q.beta = T, q.sig = T, q.z = T, quiet = T)

Arguments

filedir	character. The name of the output directory created by BGmix.
q.beta	logical. Read beta values?
q.sig	logical. Read gene variance parameters?
q.z	logical. Read z values?
quiet	logical. Parameter passed to 'scan'. (If false, 'scan' prints details of number of items read in.)

Value

mbeta	matrix no. genes x no. effects. Posterior means of gene effect parameters (usually gene means and log fold changes).
msig2	matrix no. genes x no. variances. Posterior means of gene variances.
mbb	vector of hyperparameters (b) for gene variances (posterior means).
maa	vector of hyperparameters (a) for gene variances (posterior means).
mtau	matrix no. genes x no. conditions. Posterior means of gene precisions.
mwtc	vector of posterior mean mixture weights
mzg	vector of posterior mean allocation for each gene
meta	vector of mixture parameters (eta)
mlambda	vector of mixture parameters (lambda)
pc	matrix no. genes x no. mixture components. Posterior probability for each gene of being classified into each mixture component.

Author(s)

Alex Lewin

Examples

## Note this is a very short MCMC run!
## For good analysis need proper burn-in period.
data(ybar,ss)
outdir <- BGmix(ybar, ss, c(8,8), nburn=0, niter=100, nthin=1)
params <- ccParams(outdir)

BGmix documentation built on Nov. 8, 2020, 4:54 p.m.