The Research On Construction Of Phylogenetic Tree Based On Distance Matrix

Molecular phylogenetic analysis is one of the most important fields ofbioinformatics，the main task of which is to rccoustruct a phylogenetic tree from agroup of homologous DNA or protein sequences，showing the evolutionar relationshipbetween the species of those sequences．Usually, a phylogenetic tree is a binary tree，in which the leaf nodes stand for the species or the organisms，the tree topologyindicates the phylogenetic relationship，and the length of branches figure out theevolutionary distance between the species and their common ancestor. There arc twomain types of methods to reconstruct phylogenetic trees：based on distance matrixmethod，and based on feature method. Distance matrix method has wide applicationsbecause of its simplicity and solid theory．Based on distance matrix method is a commonly method in constructingphylogenetic trees, but the traditional distance matrix method is built to base onsequence alignment. Therefore, in order to solve this problem, we propose two newmethods, both of which are without sequence alignment, and relatively intuitive, lesscalculation, easy to understand. These methods are based on the distance marixmethod in the improved transitive closure of fuzzy clustering and based on thedistance marix method in the inproved k-nearest neighbor.Based on the distance marix method in the improved transitive closure of fuzzyclustering is based on the original dissimilarity distance matrix by inproving, and geta new similarity distance matrix, which is a direct reflection on the high similaritymatrix between species, and then using the transitive closure method of the fuzzyclustering to construct the phylogenetic trees.Based on the distance marix method in the inproved k-nearest neighbor isproposed for basing on the k nearest neighbor method and graph theory, which isbased on the original distance matrix to identify k most similar taxa of each line, andthen link up with the line, if there is loop, we remove the largest side of the loop, atlast, we get a minimum connected graph, which the phylogenetic trees constructed onbased. This algorithm is mainly the choice of K, k is not too small, otherwise, whichmake the map became a minimal connected graph, even the existence of isolatededges; k is not too large, otherwise,which make the map more complicated, even thecalculation will incease, correspondingly, the time complexity and space complexitywill increase. In order to assess the method feasibility of constructing the phylogenetic trees,we often use Neighbor.exe program of PHYLIP software to assess, we verify themethod feasibility by the experiment.