Plot estimated FDR etc. for BGmix output.

Share:

Description

Given a threshold on the posterior probabilities, genes are declared as null or differentially expressed. For any given threshold, the FDR (false discovery rate) and FNR (false non-discovery rate) can be estimated using the posterior probabilities. This function plots these quantities twice, once versus the threshold probabilities, and once versus the number of declared positives.

Usage

1
plotFDR(res, ylim = NULL, q.plotfnr = F, q.plotpcut = T, q.plotnpos = T, ...)

Arguments

res

list object output from 'calcFDR'

ylim

optional argument specifying limit for y-axis

q.plotfnr

Include FNR in plots?

q.plotpcut

Include the plot of error rates v. threshold on posterior probabilites?

q.plotnpos

Include the plot of error rates v. no. positives.

...

arguments passed to 'plot'

Value

No value is returned to R. Results from BGmix model are output to files.

Author(s)

Alex Lewin

Examples

1
2
3
4
5
6
7
8
## 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)
fdr <- calcFDR(params)
par(mfrow=c(1,2))
plotFDR(fdr)