Is there a parameter that is able to display the motif identified in both ZOOPS and the ANY occurrence together?

[Carolina Mendoza]

Hello everyone, 

The sequences that I working with have multiple motifs for my protein of interest. The problem is that even when I used a low threshold, I get different numbers and locations for these motifs when I used the parameters ZOOPS and ANY. Is there any way I can identify all the motifs together?

If this is naive question, sorry I am quite new withing this area of research project 

Carolina 

[CharlesEGrant]

MEME needs to know what statistical model to use for the distribution of motifs in your sequences. Specifying different statistical models is definitely going to produce different results! 

If you think that each of your sequences contains exactly one copy of a given motif, you should use the "only one occurrence per sequence" (OOPS) model. If you think that each of your sequences will contain either zero or one copies of a particular motif, you should use the ZOOPS model. If you think your sequences might contain multiple copies of a given motif you should the "any number of repetitions" (ANR) model. Think about each kind of motif separately. For example, you might suspect that your sequences contain instances of motifs A, B, and C. If you think your each of your sequences will each contain exactly one copy of motif A, choose the OOPS model. If you think each of your sequences contains multiple copies of motif A, use the ANR model. 

You should choose the model based on your understanding of the biology of your experiment, but if you really have no idea, the ZOOPS model is a good place to start. You can try running the ANR model, and by looking at the statistical strength of the motifs discovered, and the distribution of the sites for the motif you can see if there really is evidence for multiple copies of a motif per sequence..

If you think that your sequences contain multiple kinds of motifs, you may want to increase the number of motifs MEME is asked to find. Keep in mind that MEME uses a greedy algorithm. Once MEME gathers enough evidence to estimate the statistical confidence for the motif, it will report that motif and begin looking for other motifs. It will keep this up until it has reported the number of motifs you requested, or it runs out of time. MEME will tend to report the most statistically significant motifs first, but this isn't guaranteed. You may want to start by asking MEME to find 10-20 motifs. At some point, the statistical significance  of the reported motifs should drop off dramatically.