Phylogenetic Analysis Based On Selected Evolutionary Distance Of DNA Sequence

In the research of species evolution, phylogenetic analysis from a group of homologous protein or DNA sequences has become an important field in bioinformatics. Its main method is as follows: firstly, select a group of homologous protein or DNA sequences; then calculate the evolutionary distance among sequences; finally construct an evolution tree which shows the evolution relationship of those species. From 1980s,people usually construct evolution trees by the reasonable definition of distances among protein or DNA sequences and certain assumptions under the distances.This paper has done some exploratory researches and improvement about the evolutionary distance method.Multiple sequence alignment has been used in phylogenetic analysis of DNA sequences. So we briefly describe the concepts and the research methods of multiple sequence alignment. The last step is to construct evolution tree in phylogenetic analysis of DNA sequences, so we briefly describe mainly three tree-building methods:distance,maximum parsimony and maximum likelihood.The nucleotide substitution is the driving force of the gene evolution, so estimating of nucleotide substitution number in DNA sequences, the foundation of gene evolution research, is also a basic statistic method of calculating evolution distance between genes. However, there are many nucleotide substitution models, such as the Jukes-Cantor model, Kimura model, Equal-input model, Tamura model, Tamura-Nei model and so on. And each of them all contains p-distance, d-distance andГdistance dG. So we need to select evolutionary distance from three distances in order to construct evolution trees.? We briefly describe the nucleotide substitution model and p-distance, d-distance andГdistance dG of the above mentioned models, and establish a dynamic equation of DNA sequences evolution by the identicalness in nucleotide sequences, and then evolutionary distance dy (selected evolutionary distance) is obtained. dy, calculated in the above 5 models, indicates the relationship among p-distance, d-distance andГdistance dG. According to the characteristics of dynamic equations, selection model can be transformed into a linear model . Then both of the parameters b and r (the average substitution ratio of nucleotide each year) are obtained by the Least Square Method.Finally take the mitochondrial DNA sequences of 16 species for example. We obtain the selected evolutionary distance dy, and then compare the different evolutionary distances by evolution trees constructed on basis of them. The results indicate that the selected evolutionary distance dy is an efficient evolutionary distance to analyze DNA sequences.