Search
Question: cummeRbund sigMatrix wrong counts
0
2.4 years ago by
Till0
Till0 wrote:

Hi,

I'm using cummRbund to analyze some data. However the sigMatrix function seems to produce a graph with incorrect counts of DEGs, in my case they are much higher than they should be (as reported by getSig with same alpha).This is the same error as already reported at Seqanswers (http://seqanswers.com/forums/showthread.php?t=51230) it seems.

Is that actually a bug or am I missing something?

Cheers

modified 21 months ago by r.chereji0 • written 2.4 years ago by Till0
0
21 months ago by
United States
r.chereji0 wrote:

I noticed the same problem: the number of DEGs reported by sigMatrix is higher than the numbers reported by getSig using the same alpha. I would appreciate if someone could clarify why these 2 functions report different numbers and how one could use these 2 functions to get identical results.

Here is a short example:

library(cummeRbund)

mySigMat = sigMatrix(cuff,level='genes',alpha=0.05)
mySigMat

setOfGenes = getSig(cuff,x='iPS',y='hESC',alpha=0.05,level='genes')
length(setOfGenes)

In the previous example, sigMatrix reports 120 DEGs when comparing hESC vs iPS, while getSig reports 117 DEGs. Why are these numbers different?

Thanks!