CummeRbund: subset Cuffdiff genes based >1 isoform or other features

[beeman]

I using cummeRbund to analyse some cuffdiff output. I'd like to be able to create a geneset that contains genes with >1 isoform (or other defined sig diff feature like >! tss or promoter etc).

I can create an object containing only sig DE genes as per the instructions in the manual using;
> mySigGeneIds<-getSig(cuff,alpha=0.05,level='genes')
> mySigGenes<-getGenes(cuff,mySigGeneIds)

But I was wondering if it's possible to filter these gids further to get those items with differential isoform expression between samples. At the moment I only want to get gids with >1 isoform, however I'd also like to apply this strategy/filter to other features such as TSS, promoter, etc etc..


Additionally, once such a list is created what would be the correct way of looping through the genes and plotting data for each gene?

eg;
1. get sig genes with >1 isoform
> foobar
2. For each gene in list, plot expression
> for gene_id in x_list:
> myGeneId<-gene_id
> myGene<-getGene(cuff,myGeneId)
> gl.iso.rep<-expressionPlot(isoforms(myGene),replicates=T


I'm not very experienced with R, but I've read all cummeRbund manual and some other documentation and I can't seem to figure it out. Any assistance would be warmly welcomed!!

Thanks,