library(edgeR)
files <- dir(pattern="*\\.tab$")
RG <- readDGE(files, skip=3)
table<-RG$counts; head(table); 

nn<-colnames(table)
nn<-sub(".ReadsPerGene.out","",nn)
colnames(table)<-nn
colnames(RG$counts)<-nn

group <- factor(rep(c("T0","T1","T6"),2))
counts<-RG$counts
> dim(counts)# 48526     6
nonzero<-counts[rowSums(counts)>0,]; dim(nonzero)# 30772     6
y <- DGEList(counts=nonzero, group=group)
keep <- rowSums(cpm(y)>1) >= 4; #por lo menos el 66% de las muestras tengan 1 cpm
dim (y[keep,])#  13467     6
y <- y[keep, , keep.lib.sizes=FALSE]
y <- calcNormFactors(y)
y <- estimateDisp(y)
y <- estimateTrendedDisp(y)
y <- estimateTagwiseDisp(y)
png("bcv.png")
plotBCV(y)
dev.off()

png("smear.png")
plotSmear(y)
dev.off()


geneNames<-read.table("Hsa.ID.names.txt")