How to use DESeq with non Integer values from mirdeep2

[cmorgan]

Hi,

I've recently started using mirdeep2 for miRNAseq data and would like to do a DE analysis between two conditions.

The normalised miRNA expressions were given in an output file by mirdeep2 called "miRNAs_expressed_all_samples_now.csv" which contained colums like this:

#miRNA read_count precursor total seq seq(norm)
mmu-let-7a-1-3p 571.00 mmu-let-7a-1 571.00 571.00 74.79
mmu-let-7a-5p 38731.00 mmu-let-7a-1 38731.00 38731.00 5073.08

The colum with my normalised scores (seq(norm)) has non integer values which I have been unable to use with DESeq to perform a DE analysis.

I was given the following error when i tried to attatch my conditions to the counts table.
> cds<-newCountDataSet(countsTable, conds)
Error in newCountDataSet(countsTable, conds) :
The countData is not integer.

Has anyone come across this before? Should i convert my seq(norm) colum to integer values or is there a way to force DESeq to accept non integer values.

Any help or advice you can give would be much appriciate.

[Wallysb01]

When I went through the mirdeep2 to DESeq, I just used the raw counts as output by mirdeep2 and not the mideep2 normalized counts. DESeq will normalize for you anyway.

[cmorgan]

Excellent! That worked like a charm.

Cheers

[dolphing]

When I went through the mirdeep2 to DESeq, I just used the raw counts as output by mirdeep2 and not the mideep2 normalized counts. DESeq will normalize for you anyway.

Hi, Wallysb01, if I use -W in quantifier.pl options which you mentioned in another thread then the raw counts will have non-integer numbers. How to deal with this situation? Is that appropriate to use function 'round' in R to convert these numbers to integer?
Thanks~