htseq-count: how to use multiple features?

[nasobema]

From the documentation I can only find how to select a certain type of feature for counting, like "exon" (default) or "ncRNA".

Is there also a way to select multiple features?
I'm using it currently for bacterial transcriptomics and ncRNA are of increasing importance here. Currently, I still need to run the script twice on different features and merge the results afterwards.

While this actually works fine, it would be simpler to just tell htseq-count to look for multiple features at the same time.

Any hints on how to do this?

[dpryan]

There isn't, but it would be simple to just modify the annotation file that you're using to just collapse exon and ncRNA into a single feature. Something like the following would work:

awk '{if($3 == "exon" || $3 == "ncRNA") {$3="interesting_feature"}; print $0}' foo.gtf > foo.modified.gtf

[nasobema]

There isn't, but it would be simple to just modify the annotation file that you're using to just collapse exon and ncRNA into a single feature. Something like the following would work:

awk '{if($3 == "exon" || $3 == "ncRNA") {$3="interesting_feature"}; print $0}' foo.gtf > foo.modified.gtf

OK, sure. But then, there wouldn't be the option to separate the feature types any more. Unless I maintain multiple annotation files, of course.

Seems like I have to stick to workarounds.

[dpryan]

A work around might prove simpler, yes. featureCounts also doesn't seem to support this, though I suppose you could just alter the source code for either of them to add this functionality.