[BioC] Plotting Gene Expression Values

[Sean Davis]

----- Original Message ----- 
From: "McGee, Monnie" <mmcgee at mail.smu.edu>
To: <bioconductor at stat.math.ethz.ch>
Sent: Friday, March 04, 2005 3:49 PM
Subject: [BioC] Plotting Gene Expression Values


> Dear BioC,
>
> I have completed an analysis of a set of microarray data (affy data), and 
> I would like to plot the gene expression values.  The purpose is to 
> compare the genes in my four groups that are differentially expressed. 
> I've done histograms and boxplots, but what I really want is a 
> gene-by-gene comparison.  How do I plot an expression set and how do I 
> make sure that each of the expression values on the plot is labelled with 
> the proper gene?

Monnie,

You may want to look at ?heatmap.  If you have your differentially expressed 
genes so that you can index your expression set, you can do:

heatmap(exprs(MyExprs[diff.expressed.genes,]))

This will give you the typical two-way clustering.  You can use colors from 
the geneplotter package (dChip.colors or greenred.colors) to make a prettier 
picture.  I prefer the heatmap.2 function from the gplots package (bundle 
gregmisc), but this will hopefully get you started.

Sean