I'm doing some work with exon expression arrays for the first time (Affy HuEx 1.0), and I was wondering what analysis packages or tools exist. Specifically, I'm looking to identify truncated genes, where the last few exons are missing.
I've found 'exonmap' in the bioconductor repository - are there are others? What experiences have you had with them?
I use Affymetrix Power Tools (APT), which are pretty nice when you have figured out how to use them (it took a while for me..) Before that, I was using exonmap for R, but I found it a bit slow on my puny laptop. I think it also required the installation of a huge index. Both APT and exonmap are nice in their own ways.
We produce a free tool called AltAnalyze which I feel will really make any exon array analysis (or any Affymetrix expression array analysis) as easy as possible and produce the most useful information. It should work out of the box for anyone, without the need to install anything else. Users need to just download the software, unzip it, open it, select the folder with your arrays, indicate which samples correspond to which groups and get results. AltAnalyze will detect which array you are using, automatically download the needed library and annotation files needed and call APT for running RMA.
The gene expression results you get will be provided with differential expression stats and a wide array of conventional and unconventional functional annotations (e.g., WikiPathways, microRNA binding sites, protein coding vs. ncRNAs, GPCR coupling). For any non-conventional array (Exon 1.0, Gene 1.0 or JAY array), AltAnalyze will allow you to obtain alternative exon statistics using a variety of algorithms, determine which alternative exons have known splicing annotations and see what the putative effect on protein domains or microRNA binding sites is due to alternative exon inclusion. This tool will produce even more informative results when dealing with RNA-Seq data, which can be directly compared to Affymetrix data (standardized exon identifiers used).
As an FYI to others who may stumble across this thread, I've ended upusing the aroma-affymetrix packages, along with some custom post-processing code.