[Bioc-devel] Bug in DEXSeq, me thinks

[Steve Lianoglou]

Hi,

It just so happened that DEXSeq, for reasons I couldn't explain, kept
throwing errors during a call to `estimateDispersions`. I couldn't
explain it because I really thought I didn't change much, but I've got
some magic combination of samples I'm currently trying to
estimateDispersion on that none of them throw any errors while
estimate `muhats` (line 245 of core_functions.R):

muhats <- sapply(testablegenes, function(gn) {
...
muhat
})

In this particular case, every muhat returned is a proper numeric
vector, so sapply simplifies into a matrix.

Then on line 280, you try to index muhats via a list-like accessor:

muhats[[genename]]

but this fails.

I've changes line 245's sapply to lapply, and all is well in the world
again, perhaps you folks could do similar?

I'm not sure why now none of the `muhats` returned in the current
sapply aren't `try-error`s when the were before (and thereby the
sapply returned a list), but ... there we have it.

(btw, it's DEXSeq v1.1.9)

Thanks,
-steve

-- 
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact