交互移动窗口因子分析法(AMWFA)用于二维数据重叠色谱峰的比较分析

AMWFA Method Applied to Comparative Analysis of Two-dimensional Data with Overlapped Peaks

采用交互移动窗口因子分析法(AMWFA),通过挖掘两个体系中的选择性信息,获得了不同样本间的共有组分数,还同时得到了各物质对应的光谱或质谱信息.详细阐述了本法的原理和计算方法,并用一个模拟的GC-MS数据对方法进行了验证.

In ＂OMICS＂ researches, comparison study of components present in different complex systems is very difficult because of the existence of many overlapped peaks. Herein, a new method, named alternative moving window factor analysis （AMWFA） , was proposed, which is the combination and development of multi- component： spectral correlative chromatography （MSCC） and subwindow factor analysis （SFA）. The method could be utilized to determine the number of common components of different samples and resolve overlapped peaks using the common selective information hiding in two data matrixes. The advantages of AMWFA are its extensive adaptability to the systems with no or only weak selective information, self-verified to the resolution results, no experienced user request. In this paper, the powerful ability of the proposed method has been proved by the group of simulated GC-MS data.