maigesPack (version 1.36.0)

hierM: Function to do hierarchical cluster analysis

Description

This is a function to do hierarchical clustering analysis for objects of classes maiges, maigesRaw and maigesANOVA. Use the function hierMde for objects of class maigesDEcluster.

Usage

hierM(data, group=c("C", "R", "B")[1], distance="correlation", method="complete", doHeat=TRUE, sLabelID="SAMPLE", gLabelID="GeneName", rmGenes=NULL, rmSamples=NULL, rmBad=TRUE, geneGrp=NULL, path=NULL, ...)

Arguments

data

object of class maigesRaw, maiges, maigesANOVA or maigesDEcluster.

group

character string giving the type of grouping: by rows 'R', columns 'C' (default) or both 'B'.

distance

char string giving the type of distance to use. Here we use the function Dist and the possible values are 'euclidean', 'maximum', 'manhattan', 'canberra', 'binary', 'pearson', 'correlation' (default) and 'spearman'.

method

char string specifying the linkage method for the hierarchical cluster. Possible values are 'ward', 'single', 'complete' (default), 'average', 'mcquitty', 'median' or 'centroid'

doHeat

logical indicating to do or not the heatmap. If FALSE, only the dendrogram is displayed.

sLabelID

character string specifying the sample label ID to be used to label the samples.

gLabelID

character string specifying the gene label ID to be used to label the genes.

rmGenes

char list specifying genes to be removed.

rmSamples

char list specifying samples to be removed.

rmBad

logical indicating to remove or not bad spots (slot BadSpots in objects of class maiges, maigesRaw or maigesANOVA).

geneGrp

numerical or character specifying the gene group to be clustered. This is given by the columns of the slot GeneGrps in objects of classes maiges, maigesRaw and maigesANOVA.

path

numerical or character specifying the gene network to be clustered. This is given by the items of the slot Paths in objects of classes maiges, maigesRaw and maigesANOVA.

...

additional parameters for heatmap function.

Value

This function display the heatmaps and don't return any object or value.

Details

This function implements the hierarchical clustering method for objects of microarray data defined in this package. The default function for hierarchical clustering is the hclust.

See Also

somM and kmeansM for displaying SOM and k-means clusters, respectively.

Examples

## Loading the dataset
data(gastro)

## Doing a hierarchical cluster using all genes, for maigesRaw class
hierM(gastro.raw, rmGenes=c("BLANK","DAP","LYS","PHE", "Q_GENE","THR","TRP"),
      sLabelID="Sample", gLabelID="Name", doHeat=FALSE)

## Doing a hierarchical cluster using all genes, for maigesNorm class
hierM(gastro.norm, rmGenes=c("BLANK","DAP","LYS","PHE", "Q_GENE","THR","TRP"),
      sLabelID="Sample", gLabelID="Name", doHeat=FALSE)

## If you want to show the heatmap do
hierM(gastro.norm, rmGenes=c("BLANK","DAP","LYS","PHE", "Q_GENE","THR","TRP"),
      sLabelID="Sample", gLabelID="Name", doHeat=TRUE)

## If you want to show the hierarchical branch in both margins do
hierM(gastro.summ, rmGenes=c("BLANK","DAP","LYS","PHE", "Q_GENE","THR","TRP"),
      sLabelID="Sample", gLabelID="Name", doHeat=TRUE, group="B")

## If you want to use euclidean distance only into rows (spots or genes)
hierM(gastro.summ, rmGenes=c("BLANK","DAP","LYS","PHE", "Q_GENE","THR","TRP"),
      sLabelID="Sample", gLabelID="Name", doHeat=FALSE, group="R", distance="euclidean")