Ren F, Tanaka H, Gojobori T
Department of Information Medicine, Medical Research Institute, Tokyo Medical and Dental University, Japan.
Comput Methods Programs Biomed. 1995 Feb;46(2):121-30. doi: 10.1016/0169-2607(94)01609-j.
Ever since the discovery of a molecular clock, many methods have been developed to reconstruct the molecular evolutionary phylogenetic trees. In this paper, we deal with the problem from the viewpoint of an inductive inference and apply Rissanen's minimum description length principle to extract the minimum complexity phylogenetic tree. Our method describes the complexity of the molecular phylogenetic tree by three terms which are related to the tree topology, the sum of the branch lengths and the difference between the model and the data measured by logarithmic likelihood. Five mitochondrial DNA sequences, from the human, the common chimpanzee, the pygmy chimpanzee, the gorilla and the orangutan, are used for investigating the validity of this method. It is suggested that this method might be superior to the traditional method in that it still shows good accuracy even near the root of phylogenetic trees.
自从发现分子钟以来，已经开发了许多方法来重建分子进化系统发育树。在本文中，我们从归纳推理的角度处理这个问题，并应用里斯annen的最小描述长度原理来提取最小复杂度的系统发育树。我们的方法通过与树拓扑、分支长度总和以及由对数似然测量的模型与数据之间的差异相关的三个项来描述分子系统发育树的复杂度。使用来自人类、普通黑猩猩、侏儒黑猩猩、大猩猩和猩猩的五个线粒体DNA序列来研究该方法的有效性。结果表明，该方法可能优于传统方法，因为即使在系统发育树的根部附近，它仍然显示出良好的准确性。
