Hi All,
I was directed here from a similar question I posted on the Biostars website: https://www.biostars.org/p/180182/#180440
I'm analyzing some microarray data, and after applying the BH correction (which I understand to be the default) w/ topTable, I end up with a list of genes where all adjusted p-values are > .05.
When I use the p.adjust function also with the BH correction and the p-values from limma's top table though, I get adjusted p-values that are all < .05.
I'm sure there's a better way to display this but for simplicity's sake, my table looks like this: http://imgur.com/GXXmpdw
actual_adjusted p-values are the values I get using p.adjust.
Any thoughts? My understanding was that limma actually calls p.adjust, so it seems odd that these values would differ.
Thank you for recognizing my error so swiftly. I was definitely not including all genes when using p.adjust.
Perhaps I could co-opt you to answer my original question from Biostars then: how do I interpret the adj. p-value for many of my genes being the same value?
I also note that some of my top results are labeled as normgene->intron (aka not real genes). Could this be the issue? I used the oligo package for initial processing of my data. Thanks again.
This question comes up every once in a while. The repeated values are due to the part of the BH method that ensures that the p-value to FDR mapping is monotonic. See here, for example: p.adjust BH generates duplicate values