Data Mining in the Italian Peach Germplasm Database: a Preliminary Analysis

In genetic resources conservation and plant breeding research, multivariate data on continuous and categorical traits are collected with the main objective to select genotypes and accessions which best represent the entire population or gene collection, without losing genetic variability, and also to select superior accessions for utilization in crop improvement programmes all over the world. In this context, the application of means which enable “managing” reference collections is highly desirable. For this purpose, a set of fifty-seven peach varieties, maintained in the National Collection of Fruit Tree Germplasm at CRA – FRU Research Centre in Rome (Italy), were chosen, with the aim to develop a method to explore the potential of useful traits in cultivars conserved in the collection. In order to detect relationships among genotypes concerning interesting traits related to fruit size and fruit quality, Principal Component Analysis (PCA) was used. Results showed a considerable phenotypical diversity in the analysed peach germplasm. The observations carried out confirmed that the varieties carry a wide range of genetic variability concerning a range of interesting agronomic and/or organoleptic traits