pca in DESeq2

[Tom2013]

Hello,

I run successfuly using the following code in DESeq2:
plotPCA(rld, intgroup=c("population", "treatment"))

However, I would like to use ggplot2 to draw a nice PCA plot as showed in the vignette. I run the following code and encounter a problem:
data <- plotPCA(rld, intgroup=c("population", "treatment"), returnData=TRUE)

The error is that the "returnData=TRUE" was not used. Does anybody know how to fix it?

Thank you in advance.

[blancha]

Check the version of DESeq2 that you're running?

packageVersion("DESeq2")

This argument appeared in DESeq2 1.5.27.

[Tom2013]

Thank you for your answer.

> packageVersion("DESeq2")
[1] ‘1.4.5’

It's 1.4.5, although I re-installed DESeq2 from Bioconductor yesterday.

[blancha]

You should check your Bioconductor version then.
You may have to update R first to be able to use the latest Bioconductor version.
So, I would update R first, then Bioconductor, and finally DESeq2.
It's a bit of work, but DESeq2 has evolved quite a bit over the different versions.
It may actually be worth repeating the entire analysis with the latest version of DESeq2.

I do agree that it is a common source of frustration that you only get the latest version of the package corresponding to your Bioconductor version when you install a package.

[Tom2013]

You should check your Bioconductor version then.
You may have to update R first to be able to use the latest Bioconductor version.
So, I would update R first, then Bioconductor, and finally DESeq2.
It's a bit of work, but DESeq2 has evolved quite a bit over the different versions.
It may actually be worth repeating the entire analysis with the latest version of DESeq2.

I do agree that it is a common source of frustration that you only get the latest version of the package corresponding to your Bioconductor version when you install a package.

After installed the latest R, I can install the latest DESeq2. The code works now.
Thank you very much for your help.