PubMed1988
A computer-aided analysis based on soft-laser densitometry of electrophoretic patterns is described. The system has the potential to automatically identify wheat cultivars from normalized sodium dodecyl sulfate-polyacrylamide gel electropherograms. The principle encompasses autocalibration of the relative mobilities of components of the unknown pattern using standards of Mr that bracket each pattern on the pictures, automatic identification of peaks and baseline correction, and simplified estimation of the peak intensities. The algorithm provides a normalized array that can be automatically compared with a library of standard cultivar patterns in order to determine those having the highest similarity percentage. The outputs of the program are 1) relative similarity plots which show the percentage of similarity between the unknown cultivar to that of each of the common standard cultivars put in a pattern library, 2) the names of cultivars in the library having the most similar patterns and 3) computer-generated electrophoretic graphics of these cultivars. This system is not intended to identify closely related cultivars but it is particularly recommended for cultivar identification without well-trained personnel, familiar with band location and nomenclature.