[BioC] htseq and dexseq

Steve Lianoglou lianoglou.steve at gene.com
Wed Jun 26 18:27:27 CEST 2013

Hi Andreia,

It's hard to parse the read stats you put into the email due to some
forced wrapping, I guess, so I'm not quite sure what the numbers are
that you are showing, but you say regarding the htseq counting for

> But then the most weird thing is that the total is being much higher than the
> number of reads aligned,

While I've never actually used ht-seq to count for DEXSeq, from the
chatter on this list I'm pretty sure that it will likely double count
a read if it spans more than one exon, ie. one exon will assign a +1
to the count for every exon it spans. If that can explain the issue
that you are seeing, then this is expected behavior -- and moreover,
you actually want that.

As an aside, the vignette that comes with the parathyroidSE data
packge has a section that shows you how to do DEXSeq entirely using
R/bioconductor tools:




Steve Lianoglou
Computational Biologist
Bioinformatics and Computational Biology

More information about the Bioconductor mailing list