computeCorrelation: Compute R and R^2 on a particular row of two...

In alussana/AffiXcan: A Functional Approach To Impute Genetically Regulated Expression

Description

Compute R and R^2 on a particular row of two SummarizedExperiment assays

Usage

computeCorrelation(geneName, realExpr, imputedExpr)

Arguments

geneName	A string. The row name in realExpr and imputedExpr objects that identifies the vectors between which R and R^2 have to be computed
realExpr	A SummarizedExperiment object containing expression data
imputedExpr	The returning object of affiXcanImpute()

Value

A list of two objects:

• rho: the pearson's correlation coefficient (R) between the real expression values and the imputed GReX for the cross-validation i on the expressed gene y, computed with cor()

• rho.sq: the coefficient of determination (R^2) between the real expression values and the imputed GReX for the cross-validation i on the expressed gene y, computed as pearson^2

• cor.test.p.val: the p-value of the cor.test() between the real expression values and the imputed GReX for the cross-validation i on the expressed gene y

Examples

if (interactive()) {
trainingTbaPaths <- system.file("extdata","training.tba.toydata.rds",
package="AffiXcan")

data(exprMatrix)
data(regionAssoc)
data(trainingCovariates)

assay <- "values"

training <- affiXcanTrain(exprMatrix=exprMatrix, assay=assay,
tbaPaths=trainingTbaPaths, regionAssoc=regionAssoc, cov=trainingCovariates,
varExplained=80, scale=TRUE)

imputedExpr <- affiXcanImpute(tbaPaths=trainingTbaPaths,
affiXcanTraining=training, scale=TRUE)
realExpr <- exprMatrix

geneName <- "ENSG00000256377.1"
imputedExpr <- SummarizedExperiment::assays(imputedExpr)$GReX
realExpr <- SummarizedExperiment::assays(realExpr)[[assay]]

correlation <- computeCorrelation(geneName, realExpr, imputedExpr) 
}

alussana/AffiXcan documentation built on Oct. 20, 2020, 12:08 a.m.