I have a set of genes from an RNA-seq dataset generated from Wild type and mutant samples. As expected there are a number of genes that significantly differ in gene expression as called by Cufflinks/cuffdiff.

I am trying to create a scatterplot where the genes that significantly change in expression are colored differently than the genes that do not (ex.: Up = red, down = blue, NS = black).

However I can only seem to figure out how to create individual plots in a single color (like only showing the points that are significantly up-regulated in red, but no other points).

I would like to make a single graph with all of the genes plotted, and with the significantly changing genes in specific colors (perhaps even point sizes!). I'm fairly new to using R so any help would really be appreciated.

I have lists of these genes and have generated gene lists using:

>myDOWNGenesIds <- read.table("down>2.txt")

>myDOWNGenes <- getGenes(cuff,myDOWNGenesIds[ ,1])

And I can generate individual scatterplots using:

>s <- csScatter(genes(cuff), "Wild_Type", "Mutant")

>sDown <- csScatter(myDOWNGenes, "Wild_Type", "Mutant")

I can also change aspects of these individual graphs:
>sDown + geom_point(size=3, color="red")

  I am not familiar with this, but would this be sensible to try?

  First, you need to combine all 3 kinds of gene list files, up, down, and no change. There must be one column that indicates the direction of change, and then you can translate this and get a vector colors. Then, instead of using color="red", you give the color vector to the parameter "color".