DEXSeq error in fitDispersionFunction

[unicornich]

Hi to all
I have a problem with running fitDispersionFunction in DEXSeq.
I have 2 conditions, 3 replicates each.
Every step from the DEXSeq manual so far went smoothly, but when I run the fitDispersionFunction (of course after running estimateDispersion) I get an error saying:

Error in fitDispersionFunction(ecs):
no CR dispersion estimations found, please first call estimateDispersions function.

I thought it was because of the parallel processing issue as the manual says, since I tried to run it on 20 cores, but I get the same problem when running without nCores option.

Can anyone help with solving this problem?
Thanks!

Ana Marija

[dpryan]

You'll need to post the code you use (all of the relevant steps).

[areyes]

Hi @unicornich,

If you ar using R-3.0.*, have a look at this link,

https://stat.ethz.ch/pipermail/bioconductor/2014-April/059098.html

otherwise, could you include what @dpryan asked and the output off your sessionInfo()?
Alejandro

[unicornich]

@areyes
I work with R 3.0.3 and latest DEXSeq version.
And your suggestion worked! It just takes a huge amount of time because I'm still getting errors with nCores option. But I can wait, what matters is it works!
Thank you so much!

@dpryan
I followed basically the exact steps from the manual, but here they are anyway for future readers:

folder="/Volumes/Temp/Anna/counting/dexseq_counts"
countfiles=dir(folder)
countfiles=paste(folder,countfiles,sep="/")

ecs=read.HTSeqCounts(countfiles=countfiles[c(7:9,10:12)],design=factor(c("T23","T23","T23","T29","T29","T29")),"/Volumes/Temp/Anna/counting/dmel-all-transcript-r5.49-oregonR-forDEXseq-notaggr.gff")
ecs=estimateSizeFactors(ecs)
ecs=estimateDispersions(ecs)
ecs=fitDispersionFunction(ecs)
ecs=testForDEU(ecs)
ecs=estimatelog2FoldChanges(ecs)
res1=DEUresultTable(ecs)