how to extract normalized expression from DESeq2 object (dds)?

Question

[Sandeep]

Hi I have multiple groups to compare using DESeq2 and used contrast argument in results function to find DE. But now i would like to extract the normalized expression based on compared group for the heatmap and the further merging it to the results of corresponding comparison.

I was thinking the same logic as the "results function" but it seems not working.

counts(dds, normalized=T, c("Group", "A1", "A2"))

counts(dds, normalized=T, c("Group", "A1", "A3"))

counts(dds, normalized=T, c("Group", "A1", "A4"))

counts(dds, normalized=T, c("Group", "B1", "A2"))

...

etc

Answer 1

[swbarnes2]

Here's the description of how to use the counts function in DESEq

https://www.rdocumentation.org/packages/DESeq2/versions/1.12.3/topics/counts

It does not anywhere allow for you to subset using elements of colData or the language of contrasts. So stop trying. Just get the whole matrix and then subset it.

[Sandeep]

I get it there is no arguments in counts function to use contrast type group selection but i got 28 groups and 378 paired wise combinations DE. now tell me how to subset it, without automate it ?

[swbarnes2]

Counts will give you the normalized counts for each sample. That's all.

[Sandeep]

Thats what i need but by the groups i select.