class.mergeExpressionSet: Class mergeExpressionSet, a class for merged microarray data,...
In MergeMaid: Merge Maid

Description Details Slots Methods See Also Examples

This is class representation for merged Microarray Data.

The mergeExpressionSet class is conceived as an extension of the ExpressionSet class provided in Biobase for the storage of expression array data. A mergeExpressionSet object is primarily a list of ExpressionSet objects, along with an incidence matrix indicating which genes appear in which studies. A mergeExpressionSet object with a single study reverts to the ExpressionSet class. A number of accessor functions are defiined for this class, as well as a few convenient analysis and plotting functions.

We assume there are K studies, representing a total of M unique genes.

data: A list of ExpressionSet objects.
geneStudy: Binary incidence matrix with M rows and K columns. Each column represents a study, and each row represents a gene. If study "s" contains gene "g", then geneStudy[g,s]=1, otherwise geneStudy[g,s]=0.
notes: Object of class "character" This slot is available for storage of descriptive information.

Derived from ExpressionSet:

exprs(mergeExpressionSet): An accessor function for the data slot.
exprs<- (mergeExpressionSet): A replace function for the data slot
notes (mergeExpressionSet): An accessor function for the notes slot.
notes<- (mergeExpressionSet): A replace function for the notes slot.
geneNames (mergeExpressionSet): Accessor function for union of gene ids in all studies.
geneNames<- (mergeExpressionSet): A replace function for gene ids.

Class-specific methods:

geneStudy (mergeExpressionSet): Accessor function for the geneStudy slot.
phenoData (mergeExpressionSet): Accessor function for phenodata in ExpressionSet's. Returns a list, one phenodata matrix per study.
phenoData<- (mergeExpressionSet): A replace function for phenodata in ExpressionSet's. Returns a list, one phenodata matrix per study.
intersection (mergeExpressionSet): Represent data for genes common to all studies as a single ExpressionSet object.
modelOutcome (mergeExpressionSet): Calculate regression coefficients for each study/gene.
intCor (mergeExpressionSet): Calculate the integrative correlation coefficients for mergeExpressionSet data.
intcorDens (mergeExpressionSet): Plot the distribution of the integrative correlation coefficients and the null distribution obtained by permutation here we use the approximate method to calculate the integrative correlation.

Standard generic methods:

length (mergeExpressionSet): Function returning the number of studies in the mergeExpressionSet.
names (mergeExpressionSet): Function returning study names.
names<- (mergeExpressionSet): A replace function for study names.
[ (mergeExpressionSet): A subset operator. Returns a mergeExpressionSet containing a subset of the studies. A mergeExpressionSet with only one study is returned as a single ExpressionSet.
summary (mergeExpressionSet): Obtain the basic information for 'mergeExpressionSet'.
plot (mergeExpressionSet): Draw scatterplots to compare integrative correlations for genes, here we use the approximate method to calculate the integrative correlation.

mergeExprs, intCor, modelOutcome, intcorDens, ExpressionSet

  if(require(Biobase) & require(MASS)){
  data(mergeData)
  merged  <-mergeExprs(sample1,sample2,sample3)

  merged[1:2]
  i<-c(1,3)
  merged[i]

  exprs(merged)

  names(merged)<-c("study1","study2","study3")

  length(merged)

  summary(merged)

  plot(merged)

  plot(merged[1:2])

  intcorDens(merged)

  inter  <- intersection(merged)
  }