Similarity Analysis Of DNA Sequences With The Mode Entropy

With the rapid development of the genome project in the global range,the knowledge system of modern biology is constantly being updated and enriched,more and more biological macromolecules are being excavated,and the important of biological research is slowly from data-accumulation to data-comparison and data-analysis,bioinformatics began to develop rapidly.Many specific gene databases and protein databases related to humans,animals,plants,bacteria and other life forms have been established.Faced with such large-scale biological data,deep analysis of the useful information contained in these biological sequences,and depth analysis of the similarity between these biological sequences has become the focus of current bioinformatics research.How to deeply analyze the useful information contained in these molecular data,how to use this information to reveal the correlation between with nucleic acids,DNA and proteins has become the research emphasis.In this paper,based on the DNA sequences of seven viruses,a method for analyzing the similarity of DNA sequence based on mode entropy was proposed.The main contents of the paper are as follows:(1)Introduced many common method for analyzing the similarity of DNA sequences.Especially approximate entropy and this algorithm characteristics,and carried out matlab simulation analysis,proposed a new method for analyzing the similarity of DNA sequence based on mutual mode entropy;Through numerical representations of DNA squences.The DNA basic sequence of the virus was transformed,the mutual model entropy between each sequence was calculated.The experimental datas show that it is feasible to analyze DNA sequence similarity by the mutual mode entropy method.Finally,the influence of different DNA representation methods on the mutual mode entropy values between DNA sequences are analyzed.Experiments show that the mutual mode entropy values between sequences based on the integer representation method can more accurately determine the degree of similarity between sequences.(2)The influence of coding length m on the mutual mode entropy of DNA sequences is discussed.The simulation results show that the mutual mode entropy calculated by using the appropriate m-value can more accurately reflect the similarity between different sequences;And compare the effects of mutual model entropy,approximate entropy,DTW algorithms to measure the similarity between DNA sequences.(3)Local similarity between DNA sequences is simulated dynamically by sliding window method.and obtain the "most matching" window size under different DNA representation methods.Based on this,the similar interval segments between homologous sequences are found out.