edge (version 2.4.2)

plotQLDisp: Plot the quasi-likelihood dispersion

Description

Plot the genewise quasi-likelihood dispersion against the gene abundance (in log2 counts per million).

Usage

plotQLDisp(glmfit, xlab="Average Log2 CPM", ylab="Quarter-Root Mean Deviance", pch=16, 
       cex=0.2, col.shrunk="red", col.trend="blue", col.raw="black", ...)

Arguments

glmfit

a DGEGLM object produced by glmQLFit.

xlab

label for the x-axis.

ylab

label for the y-axis.

pch

the plotting symbol. See points for more details.

cex

plot symbol expansion factor. See points for more details.

col.shrunk

color of the points representing the shrunk quasi-liklihood dispersions.

col.trend

color of line showing dispersion trend.

col.raw

color of points showing the unshrunk dispersions.

...

any other arguments are passed to plot.

Value

• A plot is created on the current graphics device.

Details

This function displays the quarter-root of the quasi-likelihood dispersions for all genes, before and after shrinkage towards a trend. If glmfit was constructed without an abundance trend, the function instead plots a horizontal line (of colour col.trend) at the common value towards which dispersions are shrunk. The quarter-root transformation is applied to improve visibility for dispersions around unity.

Examples

nbdisp <- 1/rchisq(1000, df=10)
y <- DGEList(matrix(rnbinom(6000, size = 1/nbdisp, mu = 10),1000,6))
design <- model.matrix(~factor(c(1,1,1,2,2,2)))
y <- estimateDisp(y, design)

fit <- glmQLFit(y, design)
plotQLDisp(fit)

fit <- glmQLFit(y, design, abundance.trend=FALSE)
plotQLDisp(fit)