Scanning Microscopy


In cell biology, electron probe X-ray microanalysis can reveal the distribution of chemical elements inside a single cell. The full description of a biological system (cell population, tissue) requires a great number of spot measurements. In quantitative analysis, the measurements are subject to experimental errors of several types; moreover, the relations between the resulting values are usually more interesting than the absolute concentrations. Nevertheless, the proper evaluation of quantitative values can discover information more on the object of study.

A system of simple statistical tests is suggested here which can solve several problems. Some concentration values can be far from the statistical average due to errors in measurement; therefore, a statistical test of plausibility of the measured values is carried out. In the compartments (e.g., nucleus, cytoplasm or other selected areas), the distribution of an element can be nonhomogeneous, and hence a statistical test of homogeneity of the element distribution in specified areas is provided. The tests continue with a test for correlation, in which the concentrations of a given element in a pair of specified areas are compared. These tests proceed step-by-step for all elements of interest. Subsequently, the relations of concentrations in all possible pairs of elements in the area in question are calculated. Moreover, cells within a population can be different from the point of view of elemental concentration; a statistical test of homogeneity of the cell population is provided. In the case of nonhomogeneity, the concentration values and/or cells within a population are clustered into homogeneous groups.

The evaluation is carried out automatically, with a simple program. The system of programs, in which the program for evaluation is incorporated, is included semi-on-line in the EDAX9900 system, where the measurement and evaluation are carried out in sequence. The results for a population of Srrepromyces aureofaciens are shown as an example.

Included in

Biology Commons