particle characterization

The CytoSense flow cytometer produces data on forward scatter, sideward scatter and one or more wavelengths of fluorescence for many individual particles. Particles of a certain kind tend to have properties that are about the same for all particles of that kind. For instance, phytoplankton cells have chlorophyll that emits red light when exited by the laser light, and some kinds of phytoplankton have auxiliary pigments also emit shorter wavelengths. So, fluorescence can be used to separate phytoplankton cells from zooplankton and sediment. And if fluorescence at shorter wavelengths than red is also measured then phytoplankton cells having auxiliary pigments can be separated from those having not. The measured scatter and fluorescence of each particle are a set of correlated numbers belonging to that particle. With thousands of particles measured, a plain list of all those numbers provides no insight. These numbers are visualized therefore in scatter plots (we call them dotplots) and distributions. The graph shows three bivariate scatter plots of an aquatic sample containing several species.

Graph: Analysis of plankton and coccoliths by flow cytometry. Data are displayed as dualparameter plots of combinations of light scatter and fluorescence depending on the particles being analysed (Graph by Glen Tarran, Plymouth Marine Laboratory).

The CytoClus data analysis program offers options to draw boxes or polygons around the clusters in such dotplots to make a specific selection set for a target cluster. Such a selection set is a collection of boundaries on the measured parameters, which means that it is a morphological description. The discrimination power is increased if a cluster is selected in more dotplots. Statistics for this and every other distinguishable group can be generated (number of counts, averages and standard deviations etc.). Instead of single measured values, the Cytosense flow cytometer produces complete signal courses (scans) of measured scatter and fluorescence of each particle. From these digitized signal courses the software programme CytoClus extracts properties that can be represented by a single number also called 'parameter'. This parameterizing is a software version of the more basic hardware signal quantification in standard flow cytometers (viz. the section on 'scanning').