【摘要】 DNA序列的相似性分析已成为当前生物信息学科中的研究热点,对分析算法的需求也逐步增加,基于样本熵的DNA序列相似性分析方法存在一定的效率问题。本文提出了一种基于多尺度熵的分析方法,以7种病毒DNA序列作为实验研究的对象,采用整数法将其分别表示为时间序列,而后通过对比多个时间尺度下序列之间样本熵互值大小来显示序列之间的相关性,并与原有的样本熵算法的分析结果进行比较。实验表明,本文提出多尺度熵分析方法是切实可行的。更多还原

【Abstract】 The similarity analysis of DNA sequences has become a research hotspot in the bioinformatics discipline,and the demand for analysis algorithms has gradually increased. There are certain efficiency issues based on analyzing the similarity of DNA sequences with sample entropy. This paper studies the application of multiscale entropy for similarity analysis of DNA sequences. The DNA sequences of seven viruses are used as experimental objects,which are converted into digital sequences by the numerical representation of DNA sequences. Then,by comparing the mutual values of sample entropy between DNA sequences at multiple time scales,the similarity of DNA sequences is analyzed. And compared with the results of sample entropy method,the experiments are designed.Experiments results strengthens the conclusion that it is feasible to analyze the DNA sequences similarity by multiscale entropy.更多还原