Dipole entropy based techniques for segmentation of introns and exons in DNA

We have used superinformation, which is a measure of the disorder of the entropy content of different portions of a sequence, to analyze the structural variations of the introns and exons in DNA. We have computed superinformation for the angles of the dipole moments of the base-pairs and nucleotides in the double and single-stranded forms of DNA, respectively. We show that the computed dipole-angular superinformation of the introns are significantly higher than those of the exons and that these techniques could be used for intron-exon segmentation. They also yield more accurate and computationally faster results than the previously reported methods.