Feature Extraction Method of Protein Sequences Based on Power Spectrum

Abstract:

Based on the power spectrum，a new way for extracting the protein sequences feature was proposed by applying the hierarchical clustering and entropy evaluation. It contained the following three main parts. Firstly，the numerical expression of amino acid sequences was given by the classical HP model. Then，the characteristic spectrum of protein sequence was obtained by using the discrete Fourier transform，and a 12-dimensional feature vector was constructed to represent the protein sequence spectral. Finally，the hierarchical clustering method was used to obtain the structure of protein sequences. The way is a new extension from DNA sequence to the protein sequence. By testing and comparing on three sets of data，their hierarchical structures shown that the new method is better than the DNA sequence analysis method based on power spectrum for extracting the structural information of the data system. This method has important biological significance in determining the structure and function of the unknown genes.