Discrimination of Species with Disorder of DNA Sequences and its Application to Evolution in Introns

DNA序列紊乱物种的识别及其在内含子进化中的应用

基本信息

批准号：
17500146
负责人：
YOSHIHARA Ikuo
金额：
$ 2.43万
依托单位：
University of Miyazaki
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2005
资助国家：
日本
起止时间：
2005 至 2007
项目状态：
已结题

项目摘要

It is believed that evolutionary imprints have been accumulated in DNA sequence of chromosome and most of the DNA analyzing methods are based on pattern matching technology. However, it is doubtful to find almost all the significant schemes are known well in advance and also it is difficult to analyze precisely in case of getting only few fossil or sample.In such cases, a method to compare species in brief is probably welcome, even if accuracy is not enough high. Then we think out a novel method that employs disorder or irregularity of the DNA sequences to compare species. Species are compared each other from the viewpoint of irregularity of DNA sequences instead of conventional methods. The merit of the method is no need of the preliminary knowledge of the species or genes, because this method doesn't require characteristics or analyzed information of the sequence, beforehand.On the other hand, even if irregular indices are the same, it does not directly mean the species are the same. It is desirable to check the sequences from the view point some indices of irregularity. I tried to apply various methods, e.g. statistics, physical, statistical amounts like the long distance order degree etc. of several statistics, the 1/f fluctuation, the information entropy, and the quantum statistics mechanics like the occurrence rate etc. Many experiments lead us to the conclusion that the proposed method distinguishes if the living creature is comparatively near or far, but it is hard to obtain the taxonomical trees only with the indices of irregularity.

人们认为，染色体DNA序列中已经积累了进化印记，大多数DNA分析方法都是基于模式匹配技术。然而，令人怀疑的是，几乎所有重要的方案都被提前知晓，而且在只有很少的化石或样本的情况下很难精确分析。在这种情况下，一种简单比较物种的方法可能会受到欢迎，即使准确度不够高。然后我们想出了一种新方法，利用 DNA 序列的无序或不规则来比较物种。物种之间的比较是从DNA序列的不规则性角度而不是传统方法进行的。该方法的优点是不需要物种或基因的初步知识，因为该方法不需要事先要求序列的特征或分析信息。另一方面，即使不规则指数相同，它也不需要预先了解序列的特征或分析信息。并不直接意味着物种相同。需要从一些不规则性指标的角度检查序列。我尝试应用各种方法，例如统计学，物理，统计量，如几个统计的长距离有序度等，1/f涨落，信息熵，以及量子统计力学，如出现率等。许多实验使我们得出的结论是，该方法区分生物的距离较近或较远，但仅靠不规则性指数很难获得分类树。