Incosistency between contrast and coef parameters in lfcShrink function
1
0
Entering edit mode
יהב • 0
Last seen 29 days ago
Israel

I encountered with inconsistency while using lfcShrink and I would like to consult with you.

There is option to use contrast or coef parameters with this function. contrast can receive a list of 2 character vectors, while coef can receive only elements of resultsNames(dds). This is issue for me because with contrast switch between the elements affect the sign of log2foldchange, and with coef I can't control it.

Do you have solution for this issue?

DESeq2 • 218 views
0
Entering edit mode
@mikelove
Last seen 1 day ago
United States

This is discussed in the extended section on shrinkage in the vignette.

0
Entering edit mode

Can you point out to this discussion? I can't find answer for this issue there

0
Entering edit mode

For all Bioconductor packages, not only DESeq2, you can either find these on the package homepage, or you can directly type into your R console:

vignette(“DESeq2”)

0
Entering edit mode

Yes I know, but I can't find anything there regarding the problem I mentioned here

0
Entering edit mode

Extended section on shrinkage (or something like this) should be a header in the table if contents. You can search for “Extended section”