RNA-seq(tophat|cufflinks) on part of a genome: special considerations?

[brachysclereid]

Hi,
I was wondering if there are any particular considerations one must take into account when running RNA-seq on just a small part of the genome. Such as areas surrounding a few GWAS hits?

[Pepe]

You'll find matches for eads belonging to places outside your partial genomic reference. Mapping algorithms try to find the best hit in the reference provided. If the real place is not provided they'll match to whatever they can find.

[Simon Anders]

Exactly. So, map against the whole genome, and narrow down to your loci of interest only afterwards.