Flow cytometry (Fluorescence-Activated Cell Sorting, FACS) is a method for the analysis and preparation of particles in mixtures of substances based on scattered light and fluorescence properties. The high analysis speed and sensitivity as well as the objective quantification and multiparametric correlations (relationship of at least two variables) open up an almost unlimited field of applications for flow cytometry in research and diagnostics. The focus of interest is always the individual cell. In contrast to traditional biochemical and cytochemical methods, no average values of a cell preparation are obtained, but the correlation of the result with the individual cell is maintained. Analysis techniques rely on representations using one-dimensional (e.g. histograms), two-dimensional (e.g. dot-plot) figures, and even higher-order graphs (plots) (3D-plots, SPADE trees, etc.).
The most common and well-known evaluation representation of flow cytometry is the histogram. It represents a frequency distribution of the measured signals of a parameter. Typically, figures with data from different conditions are shown in one diagram. The horizontal axis represents the intensity of the individual measurements and the vertical axis represents the number of cells. In this way, the Gaussian distribution of a parameter is obtained, which is called the population. When cells are stained with fluorescently labeled antibodies, the fluorescence intensity is directly proportional to the number of binding sites (antigens) present, i.e., the more binding sites there are, the brighter the cell glows. Flow cytometry thus shows the distribution of different fluorescence intensities using a relative scale. The user must specify which areas should be considered “positive” or “negative” to properly evaluate a given population. These ranges are defined with control. Histograms are useful for cell cycle and proliferation analyses but are less useful for plotting data for several reasons. First, relationships between different markers will not be detected, i.e., double-positive cells cannot be identified. Second, small populations are lost in larger distributions, thus rare events are not noticed.
If two different parameters are recorded during one measurement, a histogram is not sufficient and a two-dimensional representation is used. It allows to show the correlation distribution, i.e. the relationship between two different characteristics, and thus to identify more complex phenotypes. Thus, the populations in demand can be isolated using gating. The original two-dimensional plot is also known as a “dot plot,” a graph that showed the relationship between two traits but lacked detail in terms of the intensity of the number of events in a given region. Therefore, two-dimensional plots have some utility in showing how populations of interest are identified.
Another form of representation is a scatter graph, which can show information about dependency structures of two defined characteristics. The data are shown as scatter graphs, in which distributed characteristics can overlap if the same values are present several times. From scatter graphs, various focal points can be shown, such as the number of experiments performed to generate the data, or the mean, dispersion, and significance of the data.
In flow cytometry, frequency distributions (populations) of cells are defined by their scattered light parameters – forward scattered light (FSC) and side scattered light (SSC) – and by their fluorescence. Since this is a relative measurement, controls must be used to define what is considered “positive” or “negative”. In the subsequent analysis, first, a pre-selection of the raw data is made (FSC versus SSC gate) and then the boundary between a negative and a positive population is defined, often based on a negative control (“threshold method”). A gate (a defined region) can of course also be set on the fluorescence parameters, for example, to select a specific lymphocyte population. All gates can be linked together. Thus, the gates act like filters connected in series.
Compensation is the proportional subtraction of a noticeable neighboring fluorescence in overlapping fluorescence spectra. In compensation, a relative amount is subtracted from a fluorescence signal by calculation and the difference in light quantity is referred to as the compensated signal. A disadvantage of compensation is that the positive population, due to the logarithmic amplification, is pulled apart in the direction of compensation, i.e. it is scattered more widely.
Gating is the process of defining a group of cells and gating them into another plot. However, since more than one cell population or property is studied in the research, gating can be very time-consuming. It is also a common criticism of flow cytometry data in general, as it represents a subjective evaluation.
With the proliferation of new automated analysis techniques, this problem is also being addressed while assuring that the data extracted for downstream statistical analysis comes from a robust, peer-reviewed process.