论文部分内容阅读
细胞穿膜肽(cell penetrating peptides,CPPs)是一种小分子多肽,能够容易地穿过细胞膜.这类分子,尤其是具有靶向功能的CPPs为高效率投送药物到靶细胞带来希望.因此,对其展开研究对于生物医学有着一定的意义.本工作主要从序列水平对具有不同穿膜活性的CPPs进行研究,试图找出影响CPPs穿膜活性的因素,以及不同活性CPPs与非穿膜肽(Non CPPs)序列上的差异,并引入一种分析生物序列的方法.我们基于CPPsite数据库和不同的文献获取CPPs和Non CPPs序列,并进一步从CPPs序列中提取具有高、中、低穿膜活性的穿膜肽(HCPPs、MCPPs、LCPPs)用于构建数据集.基于这些数据集,开展了以下研究:首先,利用方差分析的方法,对不同活性的CPPs以及Non CPPs的氨基酸及二级结构组成进行分析,发现氨基酸的静电与疏水相互作用对CPPs的穿膜活性起到了重要影响,同时螺旋结构和无规卷曲也会影响CPPs的穿膜活性;其次,使用理化性质与长度将不同活性的CPPs展示在二维平面上,发现在某些特殊的性质下不同活性的CPPs与Non CPPs可以产生聚簇现象,HCPPs、MCPPs以及LCPPs和Non CPPs被分成了三簇,这种现象显示了它们之间的差异;最后,本文引入了生物序列理化质心的概念,将组成序列的残基看作质点,进而把序列抽象成质点系进行研究,并将此方法应用到CPPs的分析中,通过PCA方法将不同活性的CPPs投射到三维平面上,结果发现绝大部分CPPs聚在一起,部分LCPPs与Non CPPs聚在一起.此工作对于CPPs的设计,以及理解不同活性CPPs序列上的差异具有一定的意义.另外,本文引入的生物序列理化质心的分析方法也可以用于其他生物问题的分析,同时它们可以作为某些生物分类问题的输入参数,在模式识别中起到一定的作用.
Cell penetrating peptides (CPPs) are small molecule peptides that easily cross cell membranes, and these molecules, especially CPPs with targeting functions, promise to deliver drugs to target cells with high efficiency. Therefore, it is of great significance to study biomedicine.In this work, CPPs with different transmembrane activities are mainly studied at the sequence level in order to find out the factors influencing the transmembrane activity of CPPs and the relationship between different active CPPs and non-transmembrane (Non CPPs) sequences and introduce a method to analyze biological sequences.We obtained CPPs and Non CPPs sequences based on CPPsite database and different literatures, and further extracted CPPs with high, medium and low transmembrane (HCPPs, MCPPs, LCPPs) were used to construct data sets.Following these data sets, the following studies were carried out: Firstly, the amino acid and secondary structure of CPPs and non-CPPs with different activities were analyzed by analysis of variance (ANOVA) The results showed that the electrostatic interaction between amino acids and hydrophobic interaction played an important role in the transmembrane activity of CPPs. At the same time, helical structure and random coils also affected the CPPs Secondly, using different physicochemical properties and length to display CPPs with different activities on a two-dimensional plane, it was found that CPPs and Non CPPs with different activities could produce clustering, HCPPs, MCPPs, LCPPs and Non CPPs under some special properties Is divided into three clusters, this phenomenon shows the difference between them; Finally, this article introduced the concept of physical sequence of biological sequence of the centroid, the residues of the composition sequence as a particle, and then the sequence of the abstract into a series of studies and Applying this method to the analysis of CPPs, the different active CPPs were projected onto the three-dimensional plane by the PCA method, and found that most of the CPPs clustered together and some of the LCPPs clustered with Non-CPPs.This work was very helpful for the design, As well as to understand the differences in the sequences of different active CPPs.In addition, the analysis method of physical and chemical centroid of biological sequences introduced in this paper can also be used for the analysis of other biological problems, and at the same time they can be used as input parameters for some biological classification problems Pattern recognition plays a role.