DLTree: A Web Server For Phylogeny Reconstruction

The rapid development of DNA sequencing technology has resialted to explosive increase of genome sequences in the big data era. These genome sequences data makes it possible to resolve the origin and evolution of species by genome phylogeny reconstruction.However, the classical alignment-based methods do not work for this due to various reasons, including high time complexity of the aligument-based algorithms, the huge size of genome sequences and the existence of remote homology, etc.A number of alignment-free methods have been proposed for phylogeny reconstruc-tion over the past three decades. But there are some long-standing challenges in these methods, including requirement of huge computer memory and CPU time, and existence of duplicate computations. In this thesis, we address these challenges with the idea of compressed vector, fingerprint and scalable memory management. With these ideas, we developed the DLTree algorithm for whole genome-based phylogenetic analysis. The DLTree algorithm was compared with other methods on some benchmarked data sets.The results demonstrate that it is more efficient and accurate for phylogeny reconstruction.