Investigation of DNA sequence and the representation for exon and intron

Abstract

The classification of gene sequences into exons and introns is a difficult problem in DNA sequence analysis. In this project, I tried to find an effective method for exon and intron prediction in a DNA sequence. Specifically, the effectiveness of DB (Dual-Base) Curve representation in detecting the three-base periodicity property in DNA coding (region of exon) sequences has been investigated. The presence of such three-base periodicity was used to distinguish an exon from intron. The Fourier spectrum analysis technique used in this project is one of the methods to classify coding (region of exon) and non-coding region (region of intron). The concept of time was given by Short Time Fourier Transform (STFT) so as to get the nucleotides' positions. Several combinations was analyzed and compared with the existing methods. The prediction using DB Curve representation is in general comparable to the existing methods. This implies DB Curve representation has the potential to be more effective in exon and intron prediction by combine the DB Curves with more complex combinations.