Data analysis of two-parameter flow cytometric measurements
Abstract:
A new procedure was developed to evaluate two-parameter flow cytometric data. These data usually show overlapping distributions of different subpopulations. To calculate the fractions of cells in the various subpopulations, the maximum likelihood method was used, which assumes a superimposition of Gaussian distributions in the histogram. The resolution quality was tested for both one-parameter and two-parameter data using simulated histograms with two superimposed subpopulations. Comparison of this procedure to existing one-parameter evaluations methods, using the same original histograms, yielded satisfactory results. For the case of one parameter as well as for two parameters, the subpopulations were well separated if the distance between the mean values exceeded 2 sigma ; the total count and the proportions of the subpopulations were of minor influence. The evaluation procedure also worked well when applied to simulated two-parameter histograms, similar to DNA protein measurements of cultured cells.
