蛋白质定量研究已成为蛋白质组学的热点,它是疾病相关生物标志物发现的重要途径。基于稳定同位素标记的质谱分析技术是蛋白质定量最常用的方法之一。随着实验方法的发展和改进,定量数据处理方法也在不断更新与完善。一般来说,定量数据处理包括四步:搜库鉴定、图谱定量信息提取与计算、肽段丰度比计算和蛋白质丰度比计算及差异显著性分析,其中后三步是数据处理的核心。目前,后三步中每步都有多种可选算法,这些算法一般都是针对特定实验技术而提出的,缺乏深人的工作对它们进行系统比较和优化。为此,在总结目前主要实验技术方法的基础上,论述了定量算法的现状和存在问题,并针对一些问题提出了可行的解决办法。
Protein quantitative research has became a hot topic in proteomics and a main approach to discover disease relative biomarkers. The technique based on stable isotope labeling and mass spectrometry is one of the main supporting techniques for protein quantification. With the advance of experimental methods, the corre- sponding quantitative data processing methods are updating and improving continuously. Generally speaking, the quantitative data processing comprises of 4 steps, including database searching for peptide and protein identification, the extraction and calculation of basic quantitative information from the mass spectra, peptide abundance ratio calculation, protein abundance ratio calculation and significance analysis, where the last 3 steps are the core of quantitative research. Currently, there are multiple optional algorithms for these last three steps and many of them were proposed only for special experiment technique. The lack of comprehensive research on the optimization of each step of the quantitative data processing and the comparison of them, motivate us to discuss the status quo and problems of the quantitative algorithms, and propose some useful suggestions in this review.