nerichment analyses for a new species

[Lapo Ragionieri]

Hi,

I would like to run an enrichment analyses for one TFBS (CTCF) in insect genome using meme.

I used fimo with drosophila jaspar TFBS meme file and I got a TFBS gff file for the genome of my species.

Now I would like know to compare Drosophila with my species and check if the two species are differently enriched in that TFBS.

I tested ame but it is not working. Can you suggest a good strategy to run enrichment analyses in a de novo species?

Best

  

[cegrant]

Using the MEME Suite to perform an enrichment analysis comparing two entire eukaryotic genomes is probably ill-advised.

The problem is that none of the MEME Suite tools can distinguish between an actual, active TFBS and a site that matches the motif entirely by chance. The MEME Suite tools all rely on statistical analysis to either detect an unusual number of motif matches in a limited set of sequences, or an unusually good match to a motif for the number of sequence positions considered. At the scale of a full genome, chance matches to a motif are generally going to overwhelm the number of true TFBS and your results will be unreliable. See section 3 of the original FIMO paper, on the limitations of FIMO when scanning full eukaryotic genomes (in particular for CTCF). 

In projects like this the MEME Suite is typically used in conjunction with wet-lab techniques like ChIP-Seq to identify actual, active TFBS.