James Johnson
Dec 10, 2012, 7:45:20 PM12/10/12
to meme-...@googlegroups.com
I would look at the E-values of the motifs found with just JASPAR (in the CentriMo results). I expect you'll find a few of them very close to the cut-off threshold. The p-values should be the same.
All the programs in the MEME Suite attempt to correct for multiple testing which is often done through the E-value. In most cases in the MEME Suite the E-value is simply the p-value multiplied by the number of things tested. That means the more motifs you test the larger the multiplier used to calculate the E-value and hence the fewer motifs that reach the cutoff and are reported.
All the programs in the MEME Suite attempt to correct for multiple testing which is often done through the E-value. In most cases in the MEME Suite the E-value is simply the p-value multiplied by the number of things tested. That means the more motifs you test the larger the multiplier used to calculate the E-value and hence the fewer motifs that reach the cutoff and are reported.
Reply all
Reply to author
Forward
0 new messages