Liu Zhichao, Cai Wensheng, Shao Xueguang
Research Center for Analytical Sciences, College of Chemistry, Nankai University, Weijin Road No. 94, Tianjin 300071, China.
J Chromatogr A. 2008 May 9;1190(1-2):358-64. doi: 10.1016/j.chroma.2008.03.014. Epub 2008 Mar 18.
For resolution of multicomponent overlapping GC/MS signals, a window independent component analysis (WICA) is proposed. In WICA, the window information, i.e., the existing interval along the retention time, of each component is obtained. Then, with the data in the selective (non-overlapping) window, the mass spectrum of the first component is extracted by using independent component analysis (ICA), and with the data of selective (non-overlapping) ions, chromatographic profile of the component is calculated. With a subtraction of the extracted information from the data matrix, the information of the next component can be obtained. In this way, mass spectrum and chromatographic profile of each component can be extracted sequentially along the retention time. Due to the usage of the window information, the limitation of ICA in resolution of multicomponent overlapping analytical signals is overcome. Two experimental GC/MS data are analyzed with the proposed WICA. Results show that both mass spectra and chromatographic profiles of the components in the overlapping GC/MS signal are accurately extracted.
为了解析多组分重叠的气相色谱/质谱（GC/MS）信号，提出了一种窗口独立成分分析（WICA）方法。在WICA中，获取每个组分的窗口信息，即沿保留时间的存在区间。然后，利用选择性（非重叠）窗口中的数据，通过独立成分分析（ICA）提取第一个组分的质谱，并利用选择性（非重叠）离子的数据计算该组分的色谱图。从数据矩阵中减去提取的信息，即可获得下一个组分的信息。通过这种方式，可以沿着保留时间顺序提取每个组分的质谱和色谱图。由于使用了窗口信息，克服了ICA在解析多组分重叠分析信号方面的局限性。利用所提出的WICA对两个实验性GC/MS数据进行了分析。结果表明，重叠GC/MS信号中各组分的质谱和色谱图均被准确提取。
