基于质谱技术筛选差异表达蛋白的统计学策略研究进展  

Statistical Strategies for Selection of Differentially Expressed Proteins Based on Mass Spectrometry Technology

作　　者：王锦霞[1] 常乘[2] 马洁[2] 吴松锋[2] 庄举娟[1] 朱云平[2] 

机构地区：[1]大连海事大学数学系,大连116026 [2]军事医学科学院放射与辐射医学研究所,北京蛋白质组研究中心,蛋白质组学国家重点实验室,北京102206

出　　处：《中国科学：生命科学》2015年第4期347-358,共12页Scientia Sinica(Vitae)

基　　金：国家重点基础研究发展计划(批准号:2013CB910800);国家自然科学基金(批准号:21105121,21475150);国家高技术研究发展计划(批准号:2012AA020201);蛋白质组信息学关键技术及分析系统研发国际合作项目(批准号:2014DFB30010);中央高校基本科研业务费(批准号:3132015159)资助

摘　　要：随着质谱技术的快速发展,蛋白质组学已成为继基因组学、转录组学之后的又一研究热点,寻找可靠的差异表达蛋白对于生物标记物的发现至关重要.因此,如何准确、灵敏地筛选出差异蛋白已成为基于质谱的定量蛋白质组学的主要研究内容之一.目前,针对该问题的研究方法众多,但这些方法策略的适用范围不尽相同.总体来说,基于质谱技术筛选差异蛋白的统计学策略可以分为3类:基于经典统计学派的策略、基于贝叶斯学派的统计检验策略和其他策略,这3类方法有各自的应用范围、特点及不足.此外,筛选过程还将产生部分假阳性结果,可以采用其他方法对差异表达蛋白的质量进行控制,以提高统计检验结果的可靠性.With rapid development of mass spectrometry, proteomics has become a popular research field following genomics and transcriptomics. Searching for reliable differentially expressed proteins will be crucial to biomarker discovery. Therefore, how to find out differentially expressed proteins accurately and sensitively has become one of the most important subjects in quantitative proteomics research. At present, there are a few research strategies focusing on this issue, but these methods have different applicable scopes and limitations. In general, the statistical strategies for selection of differentially expressed proteins based on mass spectrometry technology have three categories： the statistical strategies based on classical school, the statistical test strategies based on Bayesian school and the others. These methods differ in the application scopes, features and disadvantages. In addition, some false positive results will be generated during the process of selection. To improve the reliability of the results, new methods are needed with the development of quantitative proteomics.

关 键 词：质谱 蛋白质组学 差异表达蛋白 统计学原理 多重假设检验 

分 类 号：Q51[生物学—生物化学]