基于大规模质谱数据的蛋白质水平质量控制方法研究

发布时间：2018-04-22 00:21

本文选题：蛋白质组学 + 数据库搜索　；参考：《华东师范大学》2017年硕士论文

【摘要】：人类基因密码的破译和图谱绘制的完成,极大地鼓舞了科学家对生老病死的规律、生命起源与进化的过程、生物个体间高矮胖瘦差异性的成因等谜题的研究热情。以此为基础,为了更全面的理解和掌握生命现象的本质和规律,作为生命活动和生物过程的直接影响者,蛋白质组学顺理成章的进入科学家们的视野,并在近几年逐渐成为生命科学科研工作者的研究热点和重点。而作为蛋白质组学最关键的支撑技术的生物质谱技术的飞跃式发展更是为蛋白质组学提供了强有力的技术支持。大规模蛋白质组学的研究也因此拉开帷幕。生物质谱技术与数据库搜索方法的联合促成了高通量大规模的蛋白质质谱数据的累积。数据库搜索策略虽然极大的提高了生物质谱技术鉴定蛋白质的效率,但是由于质谱实验生物样本的差异巨大、质谱仪器的类别繁多且性能差异显著、现有数据库搜索算法的不足以及不同实验的操作过程不同等等因素造成了生物质谱技术鉴定的蛋白质质量存在问题,我们的目的是在鉴定到尽可能多的蛋白质的情况下保证鉴定结果的准确性,也就是提高召回率的同时尽可能减少假阳性。因此生物质谱技术鉴定蛋白质的质量控制方法研究成为一个关键问题。本文研究的重点在于不同来源的大规模蛋白质质谱数据鉴定蛋白质的质量控制方法的研究。解决多个不同来源实验数据进行蛋白质质谱鉴定的情况下假阳性蛋白质过多鉴定的问题。使用来自proteomexchange数据库的四组不同仪器产生的酿酒酵母的蛋白质质谱实验数据,总共113个RAW文件,进行引入实验间权重作为特征的质量控制方法鉴定蛋白质,再分别与传统的过滤谱图的质控方法和根据蛋白打分过滤蛋白的质控方法鉴定蛋白质做比较。结果显示,蛋白质打分的方法明显优于传统方法,而本文引入新特征的方法进一步改善了蛋白质鉴定的质量控制结果。
[Abstract]:The decoding of human gene code and the completion of mapping greatly encourage scientists to study the rules of birth, old age and death, the origin and evolution of life, the causes of the difference between individuals, and so on. On this basis, in order to fully understand and understand the nature and laws of life phenomena, proteomics, as a direct influence of life activities and biological processes, naturally enters the field of vision of scientists. In recent years, it has gradually become the research hotspot and focus of life science researchers. The rapid development of mass spectrometry, which is the key technology of proteomics, provides powerful technical support for proteomics. The study of large-scale proteomics has thus begun. The combination of biological mass spectrometry and database search techniques has contributed to the accumulation of high throughput and large-scale protein mass spectrometry data. Although the database search strategy has greatly improved the efficiency of biological mass spectrometry in protein identification, but because of the huge differences in biological samples in mass spectrometry experiments, mass spectrometry instruments have a wide variety of types and significant differences in performance. The deficiencies of existing database search algorithms and the different operating processes of different experiments have caused the problems of protein quality identified by bio-mass spectrometry. Our aim is to ensure accuracy in identifying as many proteins as possible, that is, to increase recall rates while minimizing false positives. Therefore, the study of quality control method for protein identification by mass spectrometry becomes a key issue. The focus of this study is on the quality control methods for protein identification by mass spectrometry data from different sources. To solve the problem of false positive protein identification under the condition of multiple experimental data from different sources. The protein mass spectrometry data of Saccharomyces cerevisiae produced by four different instruments from proteomexchange database were used to identify proteins by using a quality control method which introduced the weight of the experiment to identify the protein, which consisted of 113 RAW files. Then compared with the traditional quality control method of filtration spectrum and the quality control method based on protein scoring filter protein. The results showed that the protein scoring method was superior to the traditional method, and the new feature method was introduced in this paper to further improve the quality control results of protein identification.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：Q51

【相似文献】