compareAn2graphfile: compareAn2graphfile

In MineICA: Analysis of an ICA decomposition obtained on genomics data

Description

This function builds a correlation graph from the outputs of function compareAn.

Usage

  compareAn2graphfile(listPairCor, useMax = TRUE,
    cutoff = NULL, useVal = c("cor", "pval"), file = NULL)

Arguments

listPairCor	The output of the function compareAn, containing the correlation between several pairs of objects of class IcaSet.
useMax	If TRUE, the graph is restricted to edges that correspond to maximum score, see details
cutoff	Cutoff used to select pairs that will be included in the graph.
useVal	The value on which is based the graph, either "cor" for correlation or "pval" for p-values of correlation tests.
file	File name.

Details

When correlations are considered (useVal="cor"), absolute values are used since the components have no direction.

If useMax is TRUE each component is linked to the most correlated component of each different IcaSet.

If cutoff is specified, only correlations exceeding this value are taken into account during the graph construction. For example, if cutoff is 1, only relationships between components that correspond to a correlation value larger than 1 will be included.

When useVal="pval" and useMax=TRUE, the minimum value is taken instead of the maximum.

Value

A data.frame with the graph description, has two columns n1 and n2 filled with node IDs, each row denotes that there is an edge from n1 to n2. Additional columns quantify the strength of association: correlation (cor), p-value (pval), (1-abs(cor)) (distcor), log10-pvalue (logpval).

Author(s)

Anne Biton

See Also

compareAn, cor2An

Examples

dat1 <- data.frame(matrix(rnorm(10000),ncol=10,nrow=1000))
rownames(dat1) <- paste("g", 1:1000, sep="")
colnames(dat1) <- paste("s", 1:10, sep="")
dat2 <- data.frame(matrix(rnorm(10000),ncol=10,nrow=1000))
rownames(dat2) <- paste("g", 1:1000, sep="")
colnames(dat2) <- paste("s", 1:10, sep="")

## run ICA
resJade1 <- runICA(X=dat1, nbComp=3, method = "JADE")
resJade2 <- runICA(X=dat2, nbComp=3, method = "JADE")

## build params
params <- buildMineICAParams(resPath="toy/")

## build IcaSet object
icaSettoy1 <- buildIcaSet(params=params, A=data.frame(resJade1$A), S=data.frame(resJade1$S),
                          dat=dat1, alreadyAnnot=TRUE)$icaSet
icaSettoy2 <- buildIcaSet(params=params, A=data.frame(resJade2$A), S=data.frame(resJade2$S),
                          dat=dat2, alreadyAnnot=TRUE)$icaSet

resCompareAn <- compareAn(icaSets=list(icaSettoy1,icaSettoy2), labAn=c("toy1","toy2"),
                         type.corr="pearson", level="genes", cutoff_zval=0)

## Build a graph where edges correspond to maximal correlation value (useVal="cor"),
compareAn2graphfile(listPairCor=resCompareAn, useMax=TRUE, useVal="cor", file="myGraph.txt")


## Not run: 
#### Comparison of 2 ICA decompositions obtained on 2 different gene expression datasets.
## load the two datasets
library(breastCancerMAINZ)
library(breastCancerVDX)
data(mainz)
data(vdx)

## Define a function used to build two examples of IcaSet objects
treat <- function(es, annot="hgu133a.db") {
   es <- selectFeatures_IQR(es,10000)
   exprs(es) <- t(apply(exprs(es),1,scale,scale=FALSE))
   colnames(exprs(es)) <- sampleNames(es)
   resJade <- runICA(X=exprs(es), nbComp=10, method = "JADE", maxit=10000)
   resBuild <- buildIcaSet(params=buildMineICAParams(), A=data.frame(resJade$A), S=data.frame(resJade$S),
                        dat=exprs(es), pData=pData(es), refSamples=character(0),
                        annotation=annot, typeID= typeIDmainz,
                        chipManu = "affymetrix", mart=mart)
   icaSet <- resBuild$icaSet
}
## Build the two IcaSet objects
icaSetMainz <- treat(mainz)
icaSetVdx <- treat(vdx)

## Compute correlation between every pair of IcaSet objects.
resCompareAn <- compareAn(icaSets=list(icaSetMainz,icaSetVdx),
labAn=c("Mainz","Vdx"), type.corr="pearson", level="genes", cutoff_zval=0)

## Same thing but adding a selection of genes on which the correlation between two components is computed:
# when considering pairs of components, only projections whose scaled values are not located within
# the circle of radius 1 are used to compute the correlation (cutoff_zval=1).
resCompareAn <-  compareAn(icaSets=list(icaSetMainz,icaSetVdx),
labAn=c("Mainz","Vdx"), type.corr="pearson", cutoff_zval=1, level="genes")

## Build a graph where edges correspond to maximal correlation value (useVal="cor"),
## i.e, component A of analysis i is linked to component B of analysis j,
## only if component B is the most correlated component to A amongst all component of analysis j.
compareAn2graphfile(listPairCor=resCompareAn, useMax=TRUE, useVal="cor", file="myGraph.txt")

## Restrict the graph to correlation values exceeding 0.4
compareAn2graphfile(listPairCor=resCompareAn, useMax=FALSE, cutoff=0.4,  useVal="cor", file="myGraph.txt")


## End(Not run)

MineICA documentation built on Nov. 8, 2020, 5:35 p.m.