TBEST (version 5.0)

PartitionTree: Find the most detailed partition of a tree into tight branches.

Description

Description: The function finds the most detailed partition of a hierartchical tree into tight branches, given a level of significance for tightness.

Usage

PartitionTree(x,siglevel=0.05,statname="fldc",
        sigtype=c("raw","corrected","fdr"))

Arguments

x

An object of class best, such as computed by function SigTree.

siglevel

Threshold of significance for tightness of branches. Default is 0.05.

statname

A character string specifying the name of measure of tighness whose is significance is to be used for partition. The choices are "fldc"(default),"bldc","fldcc".

sigtype

A character string specifying how the significance threshold siglevel should be interpreted. If "raw", the threshold will be applied directly to the p-values tabulated for each tree node in x. With "corrected" chosen, the threshold will be applied to the p-values corrected for multiplicity: p_cor = 1 - (1 - p)^(N - 2), where N is the number of leaves in the tree. of significance. If "fdr", siglevel is interpreted as a threshold on false discovery rate.

Value

An object of class partition. See ?partition for details.

See Also

SigTree,partition,best

Examples

Run this code
# NOT RUN {
data(leukemia)
mytable<-SigTree(data.matrix(leukemia),mystat="all",
        mymethod="ward",mymetric="euclidean",rand.fun="shuffle.column",
        distrib="Rparallel",njobs=2,Ptail=TRUE,tailmethod="ML")
mypartition<-PartitionTree(x=mytable,siglevel=0.001,statname="fldc",
        sigtype="raw")
partition1<-mypartition$partition
sigmatrix1<-mypartition$sigvalue
fix(partition1)
fix(sigmatrix1)
# }

Run the code above in your browser using DataCamp Workspace