Hi, I am working on a multifactor analysis using DESeq. When I call estimateDispersions I get the following error: cdsFull <- estimateDispersions( cdsFull, method="pooled-CR", modelFormula = count ~ sample+allele+allele:diet) Error in glm.fit(mm, y, family = MASS::negative.binomial(initialGuess), : NA/NaN/Inf in foreign function call (arg 1) Could someone please tell me what the meaning of this error is. My design works well on my mock dataset, but with other experimental datasets having the same structure as the mock it throughs this error. Thus, it must result from something in the data. Thank you for your help. best wishes, Chris [[alternative HTML version deleted]]

Hi Chris thank you. What's your output of sessionInfo()? Are you using the most recent release version? Best wishes Wolfgang May/24/12 10:15 PM, chris_utah scripsit:: > Hi, > > I am working on a multifactor analysis using DESeq. When I call estimateDispersions I get the following error: > > cdsFull<- estimateDispersions( cdsFull, method="pooled-CR", modelFormula = count ~ sample+allele+allele:diet) > Error in glm.fit(mm, y, family = MASS::negative.binomial(initialGuess), : > NA/NaN/Inf in foreign function call (arg 1) > > Could someone please tell me what the meaning of this error is. My design works well on my mock dataset, but with other experimental datasets having the same structure as the mock it throughs this error. Thus, it must result from something in the data. > > Thank you for your help. > > best wishes, > Chris > > > > > > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- Best wishes Wolfgang Wolfgang Huber EMBL http://www.embl.de/research/units/genome_biology/huber

Hi On 2012-05-24 22:15, chris_utah wrote: > cdsFull<- estimateDispersions( cdsFull, method="pooled-CR", modelFormula = count ~ sample+allele+allele:diet) > Error in glm.fit(mm, y, family = MASS::negative.binomial(initialGuess), : > NA/NaN/Inf in foreign function call (arg 1) As just discussed off-list with Chris: This was due to a missing "try" in the code to 'estimateDispersion'. I have just fixed the issue in DESeq version 1.8.3 (release) and 1.9.7 (devel). Cheers Simon