Research On Phylogenetic Network Construction Algorithm Analysis And Platform Constrution

Phylogeny is the science of studying the evolution of species.The construction of phylogenetic trees,the construction of phylogenetic networks,and the calculation of the distance between phylogenetic networks are all important research directions of phylogeny.But so far,there is no effective integration of these algorithm platforms.As a result,phylogenetic researchers often need to use multiple phylogenetic trees or phylogenetic network construction software when constructing trees or networks,which consumes a lot of time and energy.This thesis implements the important improved algorithms of Neighbor-Joining method: INJ algorithm and ENJ algorithm in terms of constructing the phylogenetic tree.These two algorithms effectively reduce the time complexity of constructing the phylogenetic tree using the Neighbor-Joining method.In constructing phylogenetic network,this thesis implements the improved algorithms of Cass algorithm: LNetwork algorithm,BIMLR algorithm and Frin algorithm.On the problem of calculating the distance between the phylogenetic trees,this thesis implements the cluster algorithm,the segmentation algorithm,the triple algorithm and the equivalent algorithm.On the problem of calculating the distance between phylogenetic networks,this thesis implements the equivalent algorithm,semi-equivalent algorithm,three-division algorithm and vector algorithm.Through the in-depth analysis and implementation of the above algorithms,this thesis builds a visualization platform for the phylogenetic tree and phylogenetic network: PTNB software(Phylogenetic Tree/Network Builder),the visualization software implements the above algorithms,and visualizes the expression of the tree and the network as a graph,and realizes various forms of phylogenetic tree and phylogenetic network of rectangular graph,oblique line graph,and branch length ratio graph.In order to evaluate the phylogenetic tree construction algorithm,this thesis conducts Bootstrap experimental analysis.This thesis compares the INJ algorithm and the ENJ algorithm by sampling the nucleotide sequence of the real data H1N1 virus.The experimental results show that the two algorithms improve the speed of constructing the phylogenetic tree by the Neighbor-Joining method,but the bootstrap value does not decline.This thesis also analyzes the GC content of H1N1 data.The GC content of the data is about 40%,which proves that the density of DNA is not very high,and heat and alkali can easily make it denatured.Finally,this thesis compares the PTNB software with some phylogenetic network construction software.The comparison results show that PTNB is a relatively comprehensive software that covers multiple algorithms.