基于线粒体全基因组的非比对方法比较

在系统发育分析等分子生物学的研究问题中，传统上要用到基于多序列比对的方法，然而这类方法有一定的局限性。本文介绍了一类新的研究系统发育问题的方法——非比对方法，该类方法可以克服基于比对距离方法在计算规模和主观因素等方面的局限性以及由基因重组导致的比对失效。同时，本文选取新方法中的几种分子序列相似度度量，用于94种哺乳动物的线粒体全基因组序列的系统发育分析研究中。由比较结果看来，非比对方法中的FDOD度量得到的结果与传统的分类学结果最为一致；并进一步和由基于比对的距离方法对哺乳动物细胞色素b的系统发育分析得到的结果作比较，发现FDOD不仅不逊色于传统的基于比对的距离方法，而且在对哺乳动物纲各个目的整合能力上普遍优于传统方法。

Comparison of alignment-free methods based on mitochondrion complete genome

In molecular biology, the Alignment-based methods are widely applied to phylogemc analysis. However, there are some limits for this kind of methods. In this paper we introduce a new kind of Alignment-free methods, which is with less influences of subjective factors and gene reshuffle, and the most important, its computational complexity is O(L) where L is the length of sequences. The FDOD method, which is one of the Alignment-free methods, has been proposed for phylogemc analysis by us recently. We use the new method in the study of phylogenic analysis of mitochondrion complete genome and Cytb of mammalian to compare it with other methods including both Alignment - base and Alignment-free. The result of the new method FDOD is more consistent with the one in traditional taxonomy than those of other methods. In addition, we also find that FDOD do better than traditional Kimura-2-parameters Distance to unite members of the same order into one or few branches.