Error in microbial count object: Error in hatmatrix/non-conformable arguments

[bugbitten]

I imported a phyloseq object to DESEQ2 for obtaining the log2fold change in the number of OTU's per sample. My condition is the disease_stat: cancer, healthy, adenoma

Following is the code i used:

phyloseq_to_deseq2(biom_otu_tax, ~disease_stat)
cts=counts(DESeq2BIOM)
geoMeans = apply(cts, 1, function(row) if (all(row == 0)) 0 else exp(mean(log(row[row != 0]))))
DESeq2BIOM = estimateSizeFactors(DESeq2BIOM, geoMeans = geoMeans)
#It stopped working after this point
DESeq2BIOM = estimateDispersions(DESeq2BIOM)
DESeq2BIOMRES = DESeq(DESeq2BIOM,test = "Wald", betaPrior = "FALSE")
DESeq2BIOMRES = DESeq(DESeq2BIOM,test = "Wald")

The error I get is Error in t(hatmatrix %*% t(y)) :
"error in evaluating the argument 'x' in selecting a method for function 't': Error in hatmatrix %*% t(y) : non-conformable arguments"

I browsed a few other solutions before posting a question: http://seqanswers.com/forums/archive...p/t-46009.html
They did not address my issue.

What does this error indicate? Help in troubleshooting much appreciated!

[Brian Bushnell]

I moved this to Bioinformatics where you'll have better luck.

[bugbitten]

Error in hatmatrix %*% t(y) : non-conformable arguments"

Thanks for moving it here. New to the community and would appreciate some direction to troubleshoot the issue!

[Michael Love]

What version of DESeq2 software are you using? Type: sessionInfo(). The latest release is 1.8, so I'd recommend first checking if updating helps.

Note that DESeq() includes estimateDispersions() as a step, so you can just proceed from your custom size factor calculation to DESeq().

[Gorgon]

Hello,

I happen to have a similar problem to what the original poster had:

> DiffAbund <- DESeq(DiffAbund, test="Wald", fitType="parametric")
estimating size factors
estimating dispersions
gene-wise dispersion estimates
Error in t(hatmatrix %*% t(y)) :
error in evaluating the argument 'x' in selecting a method for function 't': Error in hatmatrix %*% t(y) : non-conformable arguments



The DESeq2 version I'm running is 1.8.2.

Any suggestions on what might be the problem and how to solve it? I have a bunch of variables in the model, so I guess taking a few out could solve the issue with the matrix operations, but I'm trying to avoid that. Any help would be very mucho appreciated.

[cam687]

I know this is an old thread, but in my case the error was because I had collapsed technical replicates