random occurrence of regulatory element in genome

[pirates.genome]

Hi all,

I have detected some regulatory sequences of 13 mer from genome sequencing study. The 13-mer has a consensus sequence of GNACANNNTGTNC. How can I calculate occurrence of this 13-mer in the human genome by chance? Is there any way to simulate human genome sequence and then search for the occurrence of such 13-mer by chance? Or as an alternative, mathematical way of calculating the same?

Thanks in advance

[GenoMax]

Wonder if this tool may do what you are looking for: http://bioinfo2.ugr.es/wordCluster/wordCluster.php

[pirates.genome]

Thanx GenoMax for quick reply.

wordCluster is doing what I already did. I have identified regulatory elements from my genome sequence. Now I want to find out the occurrence of such regulatory sequence in the genome by chance. May be some kind of probability calculation can be done? Or if in any way I can simulate genome sequence and then use simulated genome for calculating random occurrence of the regulatory element.

I already tried simple perl trick to simulate genome by creating array of NTs and calling random NT from array for number of times of genome length but that is a dirty trick and doesn't mimic real genome, like some genomes have different GC bias etc...

[BAMseek]

A back-of-the-envelope approach might be to model it like rolling a 4-sided die, where the probabilities of A,C,G,T are independent and based on the global GC content of that species. If you want to model the dependency of a base on an adjacent base, then you could look into Markov chains (something like this link).

Justin