lfc {IntEREst}R Documentation

Log fold change

Description

Log fold change estimation and normalized log fold change using edgeR package.

Usage

lfc(x, fcType="edgeR", sampleAnnoCol=c(), sampleAnnotation=c(), 
	silent=TRUE, group=c(), rejection.region="doubletail", 
	pseudoCnt=1, log2=TRUE, ...)

Arguments

x

Object of type SummarizedExperiment.

fcType

Available as "scaledRetention" or "edgeR" (as default) corresponding to either log fold change of scaled retention values or degeR normalized log fold change values.

sampleAnnoCol

Which colummn of colData of x to consider for the analysis.

sampleAnnotation

A vector of size 2 which cotains values from colData of SummarizedExperiment object; e.g. if getAnnotation(x)[, sampleAnnoCol]= c("test", "test", "ctrl","ctrl", ...) , and the goal is to compare "test" and "ctrl" samples, sampleAnnotation should either be c("test", "ctrl") or c("ctrl", "test").

silent

Whether run exactTestInterest silently, without warnings.

group

Vector to manually define the sample groups (or annotations). It is ignored if sampleAnnoCol is defined.

rejection.region

The rejection.region parameter in exactTest, considered only if fcType is "edgeR".

pseudoCnt

Pseudo count for log transformation (default=1).

log2

Logical value either TRUE (default) or FALSE indicating whether the foldchanges should be log 2 transformed.

...

Other parameter settings from the exactTestInterest function.

Value

Vector including fold change values.

Author(s)

Ali Oghabian

See Also

exactTestInterest, u12DensityPlotIntron

Examples


geneId<- paste("gene", c(rep(1,5), rep(2,5), rep(3,5), rep(4,5)), 
	sep="_")
readCnt1<- sample(1:100, 20)
readCnt2<- sample(1:100, 20)
readCnt3<- sample(1:100, 20)
readCnt4<- sample(1:100, 20)
fpkm1<- readCnt1/(tapply(readCnt1, geneId, sum))[geneId]
fpkm2<- readCnt2/(tapply(readCnt2, geneId, sum))[geneId]
fpkm3<- readCnt3/(tapply(readCnt3, geneId, sum))[geneId]
fpkm4<- readCnt4/(tapply(readCnt4, geneId, sum))[geneId]

# Creating object using test data
interestDat<- data.frame( 
		int_ex=rep(c(rep(c("exon","intron"),2),"exon"),4),
		int_ex_num= rep(c(1,1,2,2,3),4),         
		gene_id= geneId,
		sam1_readCnt=readCnt1,
		sam2_readCnt=readCnt2,
		sam3_readCnt=readCnt3,
		sam4_readCnt=readCnt4,
		sam1_fpkm=fpkm1,
		sam2_fpkm=fpkm2,
		sam3_fpkm=fpkm3,
		sam4_fpkm=fpkm4
)
readFreqColIndex<- grep("_readCnt$",colnames(interestDat))
scaledRetentionColIndex<- grep("_fpkm$",colnames(interestDat))

scalRetTmp<- as.matrix(interestDat[ ,scaledRetentionColIndex])
colnames(scalRetTmp)<-gsub("_fpkm$","", colnames(scalRetTmp))

frqTmp<- as.matrix(interestDat[ ,readFreqColIndex])
colnames(frqTmp)<-gsub("_readCnt$","", colnames(frqTmp))


InterestResultObj<- InterestResult(
	resultFiles=paste("file",1:4, sep="_"),
	rowData= interestDat[ , -c(readFreqColIndex, 
		scaledRetentionColIndex)],
	counts= frqTmp,
	scaledRetention= scalRetTmp,
	scaleLength=TRUE, 
	scaleFragment=FALSE,
	sampleAnnotation=data.frame(
		sampleName=paste("sam",1:4, sep=""),
		gender=c("M","M","F","F"),
		health=c("healthy","unhealthy","healthy","unhealthy")
		, row.names=paste("sam", 1:4, sep="")
	)
)

lfcFpkm<- lfc(InterestResultObj, fcType="scaledRetention",
	sampleAnnoCol="health", 
	sampleAnnotation=c("unhealthy", "healthy"), 
	silent=TRUE, group=c(), pseudoFpkm=1, log2=TRUE)
lfcFpkm2<- lfc(InterestResultObj, fcType="scaledRetention", 
	group=c("healthy","unhealthy","healthy","unhealthy"),
	sampleAnnotation=c("unhealthy", "healthy"), 
	silent=TRUE, pseudoFpkm=1, log2=TRUE)

lfcEdgeRFpkm<- lfc(InterestResultObj, fcType="edgeR",
	sampleAnnoCol="health", 
	sampleAnnotation=c("unhealthy", "healthy"), 
	silent=TRUE, group=c(), pseudoFpkm=1, log2=TRUE)
[Package IntEREst version 1.4.1 Index]