Validation of a method for molecular genotyping of fruit varieties

Microsatellite based genotyping for the identification of cultivar varieties in crop plants is nowadays a widely used and proven method. Many variations of this method are used that differ sometimes strongly with respect to work load and efficiency. We have developed a robust and efficient method that we use with great success since several years. The aim of this project was to test how well this method performs in another laboratory, and to compare the results elaborated in this and the other laboratory. We could show that the method can easily be implemented in another laboratory and that it delivers identical results. We found a highly significant overall correlation between both analysis systems (R2 = 0.9993). As expected, despite this excellent correlation, we found differences in running distances of individual gene fragments (alleles). However, these differences are only relevant if no standard cultivar sample is known/available for a comparative analysis. We developed a simple method to solve this problem. We compare the sum of the absolute differences of all alleles between the test cultivar and those of known cultivars in a database. Provided the correct standard cultivar is present in that database, its deviation from the test cultivar will be least. The candidate standard cultivar thus identified will then be used in a direct comparison. This method thus allows an unambiguous identification of a test cultivar even with moderate systematic deviations between different analysis systems.