Dear All,

I am using EdgeR package to extract the differential expression . I do not have replicates in my samples.

Now I want to know the normalized read count of my datasets.

I have tried with rpkm method but edgeR follow the TMM method for normalization so there is a variation at normalized read.

Could anyone suggest how can I get the normalized read count for all my reads.

Here is my code-

countdata <- read.table("CD.txt",header = T,sep = "\t",row.names = 1)
attach(countdata)
group <- factor(c(1,2))
y <- DGEList(counts = countdata, lib.size = c(126724,347238), group = group)
y$samples
keep <- rowSums(rpkm(y)>1) >= 2
y <- y[keep, , keep.lib.sizes=FALSE]
y <- calcNormFactors(y,method="TMM")
design <- model.matrix(~group)
y$samples
bcv <- 0.2
y <- DGEList(counts = countdata, lib.size = c(126724,347238), group = group,genes=data.frame(Length=GeneLength))
y <- calcNormFactors(y)
RPKM <- rpkm(y)
et <- exactTest(y, dispersion=bcv^2)
y1 <- y
y1$samples$group <- 1
y0 <- estimateCommonDisp(y1[c(CHI_ReadCnt,DEH_ReadCnt),])
y$common.dispersion <- y0$common.dispersion
fit <- glmFit(y, design)
lrt <- glmLRT(fit)
resSig <- res[ which(res$PValue < 0.05 ), ]

I shall be very thankful.

What you may be misunderstanding is that the TMM normalization factors are automatically taken into account when you compute the RPKMs.

I very much dislike the term "normalized read count" because I don't think there is any such thing. What you probably need is just the RPKM. And it is clear from your code that edgeR has already computed these for you.