DNA sequence classification using power spectrum and wavelet neural network

Abstract

In this paper, we present a new method to cluster DNA sequence. The proposed method is based on using the Power Spectrum and the Wavelet Neural Network (WNN). The satisfying performance of the Wavelet Neural Networks (WNN) depends on an appropriate determination of the WNN structure. Our approach uses the Least Trimmed Square (LTS) to select the wavelet candidates from the Multi Library of the Wavelet Neural Networks (MLWNN) for constructing the WNN. The LTS has been able to optimize the wavelet neural network. The LTS algorithm is to find the regressors, which provide the most significant contribution to the approximation of error reduction. This wavelet can reduce the approximation error. In this study, the DNA sequence is coded by using a binary format. The Fourier transform is applied to attain respective Power Spectra (PS) by using the binary indicator sequence. The PS is applied to construct the mathematical moments which be used to build the vectors of real numbers, which are applied to compare easily the sequences with different lengths. Our aim is to construct classifier method that gives highly accurate results. This classifier permits to classify the DNA sequence of organisms. The classification results are compared to other classifiers. The experimental results have shown that the WNN-PS model outperformed the other classifier in terms of both the running time and clustering. In this paper, our approach consists of three phases. The first one, which is called transformation, is composed of three sub steps; binary codification of DNA sequences, Fourier Transform and Power Spectrum Signal Processing. The second section is the approximation; it is empowered by the use of Multi Library Wavelet Neural Networks (MLWNN). Finally, the third section, which is called the classification of the DNA sequences. The Euclidean distances is used to classify the signatures of the DNA sequences.