I am new to motif identification area. From a set of sequences (~200), I want to search for a known motif and filter out sequences where the motif does not exist. My motif is 8bp long. A simple python search gave me too many random hits. Someone quickly suggested to me to use meme-tools and instead of simple pattern search use position-specific scoring matrix.
I think FIMO (from meme-suite) looks for known motif in given set of sequences. I ran it and it does not give me too many random hits. Python pattern search gave me as many as 9 hits on a 1000 bp long sequence. What I am struggling to understand is how FIMO works. How it calculates the log-odds scores on each sequence? Suggestions for any other tools are also welcome.