IntClust (version 0.0.2)

CharacteristicFeatures: Determining the characteristic features of a cluster

Description

The function CharacteristicFeatures requires as input a list of one or multiple clustering results. It is capable of selecting the binary features which determine a cluster with the help of the fisher's exact test.

Usage

CharacteristicFeatures(List,Selection=NULL,BinData,ContData = NULL,
Datanames=NULL,nrclusters=NULL,sign=0.05,topC=NULL,fusionsLog=TRUE,
WeightClust=TRUE,names=NULL)

Arguments

List

A list of the clustering outputs to be compared. The first element of the list will be used as the reference in ReorderToReference.

Selection

If differential gene expression should be investigated for a specific selection of compounds, this selection can be provided here. Selection can be of the type "character" (names of the compounds) or "numeric" (the number of specific cluster).

BinData

A list of the binary feature data matrices. These will be evaluated with the fisher's extact test.

ContData

A list of continuous data sets of the compounds. These will be evaluated with the t-test.

Datanames

A vector with the names of the binary data matrices.

nrclusters

Optional. The number of clusters to cut the dendrogram in. The number of clusters should not be specified if the interest lies only in a specific selection of compounds which is known by name. Otherwise, it is required.

sign

The significance level to be handled.

topC

Overrules sign. The number of features to display for each cluster. If not specified, only the significant genes are shown.

fusionsLog

To be handed to ReorderToReference.

WeightClust

To be handed to ReorderToReference.

names

Optional. Names of the methods.

Value

The returned value is a list with an element per method. Each element contains a list per cluster with the following elements:

Compounds

A list with the elements LeadCpds (the compounds of interest) and OrderedCpds (all compounds in the order of the clustering result)

Characteristics

A list with an element per defined binary data matrix in BinData and continuous data in ContData. Each element is again a list with the elements TopFeat (a table with information on the top features) and AllFeat (a table with information on all features)

Details

The function rearranges the clusters of the methods to a reference method such that a comparison is made easier. Given a list of methods, it calls upon ReorderToReference to rearrange the number of clusters according to the first element of the list which will be used as the reference.

Examples

# NOT RUN {
data(fingerprintMat)
data(targetMat)
data(geneMat)

MCF7_F = Cluster(fingerprintMat,type="data",distmeasure="tanimoto",normalize=FALSE,
method=NULL,clust="agnes",linkage="ward",gap=FALSE,maxK=55,StopRange=FALSE)
MCF7_T = Cluster(targetMat,type="data",distmeasure="tanimoto",normalize=FALSE,
method=NULL,clust="agnes",linkage="ward",gap=FALSE,maxK=55,StopRange=FALSE)

L=list(MCF7_T ,MCF7_F)

MCF7_Char=CharacteristicFeatures(List=L,Selection=NULL,BinData=list(fingerprintMat,
targetMat),Datanames=c("F","T"),nrclusters=7,topC=NULL,sign=0.05,fusionsLog=TRUE,WeightClust=TRUE,
names=c("F","T"))
# }