meaning of DESeq PCA plot axis?

[mbk0asis]

What does DESeq PCA plot axis mean?

If one sample has higher x- or y-value than others in the plot, does that mean it's better or worse than other?

Can someone explain to me?

[Simon Anders]

When interpreting a PCA plot, your only supposed to look at the distances between points, which tell you how similar the represented objects (here: samples) are. In DESeq, the point is that it allows you to check whether your replicates are really more similar to each other than to the other samples.

PCA is nothing specific to DESeq but a general concept that is used very frequently in exploratory statistics, and hence useful to know about.

For a very easily understandable introduction, see the following book. Just looking through chapter 1 will give you an idea:

Michael Greenacre: Biplots in Practice (Fundación BBVA, 2010), available on-line here

[mbk0asis]

So I can just ignore numbers on the axis in this case...

Thank you for quick reply and your great software!!!

[mbk0asis]

One more question, can you make the plot to show the spots' names?

[Wolfgang Huber]

Dear mbk0asis
this is really an R question, nothing specific to DESeq. At the most basic level, have a look at the functions text, identify and locate in R's default graphics package. Then,

• http://www.omegahat.org/SVGAnnotation/
• http://www.rforge.net/doc/packages/iplots/iplot.html
• http://lamages.blogspot.co.uk/search/label/googleVis
• http://www.rstudio.com/shiny/

and there are many more options - just try a web search for 'interactive scatterplots in R'.