[BioC] strand-specificity for interval_overlap (girafe/genomeIntervals)

[karlerhard at berkeley.edu]

Hello all,

I've been using the girafe package to make counts tables for two nascent
transcript libraries (overlap between aligned reads and genes), which I am
comparing directly.  I would like to check if there's a difference between
"sense" and "antisense" tags within genic regions.

I thought I could do this easily by making the following object:

library1_sense <- detail(library1[strand(library1)=="+"])

and then using it in the interval_overlap with the list of annotated
genes.  However, the resulting object is a data frame and not an
AlignedGenomeIntervals object, which interval_overlap seems to need.

Is there some way around this?  Maybe a way to force strandedness in the
interval_overlap function?  Or another way to create a subset of "+" or
"-" strand reads while maintaining the AlignedGenomeIntervals format? 
I've been reading the manual but I can't find a way to do this.

thanks for any help you can offer!


Karl Erhard
Graduate Student
UC Berkeley