模式识别技术在筛选溶脲脲原体和人精子交叉抗原中的应用研究

发布时间：2018-01-16 05:31

本文关键词：模式识别技术在筛选溶脲脲原体和人精子交叉抗原中的应用研究　出处：《上海交通大学》2007年硕士论文　论文类型：学位论文

【摘要】： 免疫生物学研究表明,人精子细胞和溶脲脲原体(简称UU)存在着交叉反应抗原(又称交叉抗原)。筛选鉴定这些交叉抗原对阐述UU致男性不育机理以及研制避孕疫苗具有重要意义。传统的筛选交叉抗原的方法大多基于生物学实验,过程非常复杂,其中的蛋白质提纯是一个技术难点,有时候由于量少而无法提纯,从而导致实验无法继续而失败。生物学实验筛选人精子和UU间的交叉抗原便存在着这样的问题。当今,UU的基因组测序已经完成,其基因组编码的蛋白质(包括实验鉴定以及预测工具预测的蛋白质)均可以在蛋白质数据库中检索到。另外,蛋白质数据库中还可以检索到特异性表达在人精子细胞中的蛋白质。利用这些数据集,根据交叉抗原的结构序列特性,应用模式识别的技术筛选潜在的交叉抗原是一个值得研究的课题,这也是论文研究的主要内容。发生免疫交叉反应的抗原之间含有相同或者相似的B细胞表位。利用这一基本原理,论文设计了一种从UU和人精子蛋白质数据集中筛选出候选交叉抗原及其B细胞表位的方法。首先,搜索SWISS-PROT数据库,查找出所有UU蛋白质和特异性表达在人精子中的蛋白质,建立相应数据集;其次,在UU和人精子细胞蛋白质数据集中查找出给定长度的相同或相似子序列;再者,对这些子序列进行B细胞表位的预测,选择B细胞表位可能性大的子序列,这些子序列即候选的交叉反应B细胞表位,其所在的抗原即候选的交叉抗原。这里需要解决两个关键的问题:一是如何在两个海量数据集中有效快速地查找给定长度的相同或者相似子序列;二是如何有效地进行B细胞表位的预测。论文对以上两个问题
[Abstract]:Immunobiological studies have shown. Human sperm cells and Ureaplasma Urealyticum (UUU) have cross-reaction antigens (also known as cross-antigens). Screening and identification of these cross antigens is of great significance in explaining the mechanism of male sterility induced by UU and in developing contraceptive vaccine. Traditional methods for screening cross antigens are mostly based on biological experiments. The process is very complex, in which protein purification is a technical difficulty, sometimes due to a small amount of can not be purified. This led to the failure of the experiment. This is the problem of screening cross-antigens between human sperm and UU in biological experiments. Today, the genome sequencing of UU has been completed. The proteins encoded by its genome (including those identified by experiments and predicted by predictive tools) can be found in the protein database. Proteins specifically expressed in human sperm cells can also be found in the protein database. The application of pattern recognition in screening potential cross-antigens is a subject worthy of study, which is also the main content of this paper. The same or similar B cell epitopes are present between antigens that are immune cross-reactive. Use this basic principle. A method of screening candidate cross-antigens and their B cell epitopes from UU and human sperm protein data sets was designed. Firstly, the SWISS-PROT database was searched. All UU proteins and proteins specifically expressed in human spermatozoa were found and the corresponding data sets were established. Secondly, the same or similar subsequences of a given length were found in the UU and human sperm cell protein datasets. Furthermore, the B cell epitopes were predicted and the B cell epitopes were selected, which were the candidate cross reaction B cell epitopes. Two key problems need to be solved here: first, how to find the same or similar subsequences of a given length effectively and quickly in two massive datasets; The second is how to predict the B cell epitopes effectively.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2007
【分类号】：R392

【相似文献】