We developed a method for precisely determining statistically significant repeats of all types (direct,
inverse, complementary and non-complementary). 
The method can be applied to both
nucleotide and amino acids sequences as well as on the sequences with user defined alphabet. The program that
implements the method was tested on various nucleotide and amino acids sequences. 
It was noted that not all repeats found are statistically significant and that in some cases, 
counterintuitivly, shorter repeats are more statistically significant then longer ones.
For the large number of repeats that can be found in genomes, if the minimal repeat length is relatively low,
our method provides a significant gain in performance and quality of results compared to outputting all the
found repeats. 

