The first part of the command is this:

geom_point(aes(A, log2(M), colour = factor(ifelse(log2(M) < 2, 1,2))), size = 0.8)

And the part that sets the color is

colour = factor(ifelse(log2(M) < 2, 1,2))

So I obviously missed a step, because you want the absolute value of log2(M). Sometimes it is hard to find the function you want, so in this case you should use apropos(). Also note that it is better to use shorter arguments rather than longer for apropos():

> apropos("absolute")
character(0) ## no joy!
> apropos("abs")
[1] "abs"                      "normalizeMedianAbsValues"
[3] "xtabs"   

And I'll let you take it from there.

As for question (b), the answer is always yes. This is R after all. Unless of course you mean 'is there an argument that I can use to make that happen', in which case the answer appears to be no. However, as I said, this is R so you can do pretty much whatever you want, if you are willing to work for it.

The problem  here is that MAplot() doesn't compute a p-value, so you need to supply one. Also, MAplot() just returns a 'gg' object, so we have to get all hackalicious. I'll use example(MAplot) for this next step.

>example(MAplot)

This creates data that I can now use, and also creates a 'gg' object (called 'd') that we are going to hack on. MAplot() doesn't output the M and A values directly - they are in the gg object, which is just a fancy list.

> class(d)
[1] "gg"     "ggplot"
> names(d)
[1] "data"        "layers"      "scales"      "mapping"     "theme"      
[6] "coordinates" "facet"       "plot_env"    "labels"     
> head(d$data)
      gene_id         A         M
1 XLOC_000001 0.6818857 0.1049131
2 XLOC_000002 0.0000000       NaN
3 XLOC_000003 0.0000000       NaN
4 XLOC_000004 4.9984325 1.4068705
5 XLOC_000005 2.2671843 1.6819516
6 XLOC_000006 0.0000000       NaN

We don't really want to jam all sorts of things into d, but instead want to use the functions in ggplot2 to do it for us. So we extract the data object from d, and modify:

> z <- d$data
> z$p <- runif(nrow(z))

Note that the second step I am doing is not what you want to do. You instead want to put the p-values in z, in the correct order. I assume you have a data.frame floating about that has the transcript IDs as well as p-values? If so, let's say you called it d.f. I further assume this data.frame has a column called 'gene_id' that has the gene IDs, and a column called 'p.value' that has the p-values. This is probably not true, so you need to modify to suit. Then you want to do something like

EDIT - did this backwards the first time

z$p <- d.f[match(d.f$gene_id, z$gene_id),"p.value"]

You should look at ?match to figure out what I am doing there. Now you can plot the MAplot.

> d + geom_point(data = z, aes(A, log2(M), colour = factor(ifelse(p > 0.05, 1, 2)))) + theme(legend.position = "none") + scale_colour_manual(values = c("black","red"))