Our group has a new paper out in a collective volume of Springer's series on Lecture Notes on Computer Science. The work, a collaboration with researchers from the Institutes of Biology and Computer Science of NCSR "Demokritos", describes for the first time the application of N-Gram Graphs in genomic sequences. Using this -not so conventional- approach allowed us to effectively distinguish between conserved coding (exons) and non-coding (constraint non-coding) sequences, based simply on their nucleotide composition. Given the increasing interest on constrained sequences, we believe that such approaches are very likely to become highly relevant in the search of hidden sequence functionality.
You can get a glimpse of the paper at the Publications page, while a full-text version is available upon request via e-mail.