Different Motifs predicted for the same set of sequences

[yasamanrezvani]

Describe the bug Gimme motif predicts different motifs for a set of sequences in different runs.

I have a set of sequences that are stored in a fasta file. I ran Gemme motif to find the presented motifs in the sequences for two times and each time I am getting different motifs. Is this related to the randomness of selected sequences as training set? I have set the parameter -f to 0.7 as I have small number of sequences to include 70% of my sequences for prediction of motifs and only 30% for validation.

Expected behavior I expected to get at least same number of detected motifs with very similar patterns from 2 separate runs on a single fasta

Is there anything like setting the seed when running the motif search. In case that the issue comes from randomness of training set. Please let me know if this can be fixed.