Abstract : Sequencing is the process of establishing good relations between an inheritance. The general purpose of sequence similarity analysis is to present the possibilities of homology sequence: the probability that these sequences come from the same ancestors. The same signal is intended to measure the rate of occurrence between nucleotide numbers or protein ranges. There are many uses for measuring single sequences. Bioinformatics has been an emerging field of research for the past three decades. The ultimate goal of bioinformatics was to maintain and manage biological data, and to develop and analyze computer tools to increase their understanding. The size of the data accumulated under various sequence projects is increasing exponentially, which brings with it the difficulties of testing methods. To reduce the gap between recently followed proteins and proteins with known functions, many calculation methods involving protein analysis and integration of algorithms were proposed in the past. Analysis of the protein sequence of protein sequences in existing families helps to predict the formation and function of a large number of newly discovered proteins. The existing results of protein analysis are unsatisfactory due to the large number of features obtained by various encoding methods. In this exercise, the process of selecting a mathematical­based feature is proposed to reduce the vector size of the extracted element. The proposed method of protein similarity analysis shows significant improvement in terms of comparison metrics: (a). hydrophobicity, (b). normwaalsvolume, (c). polarizability, (d). secondarystruct and (e). solventaccess.