Spectral Methods for Classifying Short Exon and Intron Sequences

In the paper, the performances on FFT-based, FIR and IIR digital filter-based (DF-based) period-3 classifications of short exon and intron sequences of the human and eleven model organisms are studied. For each organism, 8,000 exon sequences and 8000 intron sequences of 150-base length are used for training and testing. Two numerical representations and three threshold values are considered. For comparison, FIR and IIR digital filter orders are chosen to have a similar computational complexity as that of the FFT-based method. Simulation results indicate that the FIR DF-based method and the IIR DF-based method can serve as effective methods for spectral analysis of nucleotide sequences, and the K-Quaternary Code I is attractive in terms of classification precision.