Hi, I'm new in bioinformatics and I have had some trouble finding a binding site for a known transcription factor. To summarize, I'm interested in finding if any of my sequences (only 5 promoter sequences) have a certain motif, which has been reported for other similar organisms. I tried MEME discovery tool (web tool), but using just 5 sequences has given very weak results (e-values too high) which I guess are not reliable. I tried with FIMO and MAST using as input motifs some sequences reported in articles, but the results I got from both are quite different. For example, for one sequence that very likely has that motif, when using MAST I got at least 3 possibilities of the position of the motif, but when I use FIMO I got zero results.
Does anyone know what other tools can I use to confirm my results? Or do you have any advice on the strategy I can follow to find those motifs?
Thank you, and sorry for the long post!
Motif finding usually doesn't work. The signal is too weak and the actual binding of the TF to the promoter depends on other factors such as DNA structure. How prominent is the motif you are searching? How much information does it contain?
The motif is 20 bp long and has a dyad symetry. I'm not sure what are you refering to with how much information. When you say motif finding doen't work, are you refering to de novo discovery, or all approaches in general?
Information meaning information content, how strong is the motif. I mean that the whole concept that we can find motifs in the DNA and say with some confidence that a DNA binding protein interacts with this region is far fetched. I must be combined with DNA accessibility test.