Quick question on DESeq output baseMeanA and baseMeanB

[tdyo]

Hi all,

I may be over thinking this a bit, but I wanted to confirm the distinction between baseMeanA and baseMeanB in the DESeq output.

I assigned categories as "S, S, S, R, R, R" to my six samples, and I have assumed that baseMeanA refers to the S's and baseMeanB refers to the R's. Then I got worried that the categories may get alphabetized in the process of DESeq's analysis and the group results are flipped.

I looked through the DESeq manual, but couldn't find anything specifying a "first set is A, second set is B" rule or otherwise. Thanks in advance for clearing it up.

[chadn737]

It depends on how you list them when you use nbinomtest.

If you do res <- nbinomtest(cds,"S","R")

then baseMeanA should be "S".

If you do res <- nbinomtest(cds,"R","S")

then baseMeanA should be "R".

[tdyo]

If you do res <- nbinomtest(cds,"R","S")

then baseMeanB should be "R".

Just to clear up a potential typo, I'm guessing that was meant to be "baseMeanA"?

[chadn737]

Just to clear up a potential typo, I'm guessing that was meant to be "baseMeanA"?

Yes, I'll correct it