Background letter frequencies

[Ariana Treat]

Hi,

I have concatenated STREME output files from three separate runs on different lists of sequences (different phylogenetic order), and I used this file as both the input and query motifs for TOMTOM to compare all motifs to themselves after changing the motif numbers so there would not be duplicates. Should I make any changes to the background letter frequencies at the top of each file? Since the motifs are from different sequences, the background frequencies are different. I'm not sure if TOMTOM is only reading the first occurence of background frequencies in my combined file which has three different background frequencies, or if it is able to use all three for each set of motifs? I'm also unsure how much the background frequencies would change the results, since the TOMTOM output from my concatenated file looks reasonable with visually similar motifs having small e-values.

I'm also curious whether I should change my background frequencies from STREME output to match the background frequency of the sequences I am searching when I use STREME motifs from one set of sequences to search for occurences of these motifs in a different set of sequences.

Best,

Ariana