SummaryClassLabel: Local FDR method for summarizing class labels
In BANFF: Bayesian Network Feature Finder

Description Usage Arguments Value Examples

The finalized class label can be either by method "Major vote", or "under local fdr control".

1 2	SummaryClassLabel(net,test.stat,zTrack,method=c("MajorVote","LocfdrControl"), plot=FALSE,nburn=50,nskip=2,locfdr.alpha=0.2)

`net`	The adjacent matrix with 0/1 indicating "connected" or "not directly connected
`test.stat`	The observed test statistics. Missing values are represented as NAs.
`zTrack`	The trace for z. It is a num_of_genes by num_of_iterations matrix.
`method`	A char. Either "MajorVote" or "LocfdrControl".
`plot`	Logical. Default=FALSE, means the plot should be drawn.
`nburn`	Default=50.
`nskip`	Default=2
`locfdr.alpha`	Default=0.2

A vector, where each value is the summarized class label in c(1,2,3) representing "down-regulate", "null", and "up-regulate".

## Not run: 
data(net)
data(test.stat)
res=FeatureClassFinder(net,test.stat,niter=100,na.action="NN")
classLabelEst=SummaryClassLabel(net,test.stat,res$zTrack,method="MajorVote")
print(table(classLabelEst))

## End(Not run)