retrieve normalised count data from DESeq2

Question

[Rößti]

Hello, sorry if this is a really stupid question but I have been struggling with this for a week or so now.

Basically I carried out a bacterial RNA-seq on 2 conditions in triplicates. After doing some analysis using DESeq2 it is looking like there is no differential expression between the 2 conditions however, is there a way that I can save the normalised count data to a cvs file so that I can check the quality of the replicates?

Thanks

[Nicolas Rosewick]

FYI you can do a PCAplot to check wheter your replicates cluster together .. or not. Check DESeq2 vignette for PCAplot

Answer 1

[andrew.j.skelton73]

foo <- counts(object, normalized = TRUE)
write.csv(foo, file="norm_counts.csv")

where object is a DESeqDataSet object

[Rößti]

Cheers for that :o)

[pablo61991]

And to obtain the count matrix before the normalization just change normalized = TRUE for normalized = FALSE?

[andrew.j.skelton73]

Yes, or don't include the normalized parameter as when you look at the function's help page (?DESeq2::counts), you'll see that it's FALSE by default.

[pablo61991]

Oh, thank you I missed the default parameter.

[LuisNagano]

Thank you, andrew, your script worked out!

But when I extract the DESeqDataSet object (dds), normalized counts table don't have Gene Ids, only numbers, like that:

gene id         BSR111-Med-46                   BSR112-Med-58
1               2.87573335679571                2.58809911711063
2               31.6330669247528                51.7619823422126

How can I get the normalized count data with gene ids?

Thank you again

[andrew.j.skelton73]

Check the rownames on your input matrix. If they're not set, then they'll default from 1:n