RNA-Seq Data Analysis    ◾    189

5.3.7.7  Exploring the Data

Up to this point, we have prepared the count data for fitting the negative binomial general-

ized log-linear model. However, before that step, we can explore the data by visualizing the

library size and the distribution of cpm (log2 counts per million) of each sample.

png(file=”libsizeplot.png”)

x<-barplot(yNorm$samples$lib.size/1e06,

names=colnames(yNorm),

las=2, ann=FALSE,

cex.names=0.75,

col=”lightskyblue”,

space = .5)

mtext(side = 1, text = “Samples”, line = 4)

mtext(side = 2, text = “Library size (millions)”, line = 3)

title(“Barplot of library sizes”)

dev.off()

FIGURE 5.15  Library sizes.