Using the Griffing’s experimental design method I, model II. Apple breeding - a case study as a proposed methodology of the statistical and genetic analysis

Abstract

The diallel mating design is widely used in plant breeding to estimate combining ability of genitors and useful genetic parameters. In this work, a complete diallel mating design with four apple cultivars used as genitors was set, in order to elaborate a pattern for a quantitative trait, respectively the height of apple hybrids, measured at three months after emergence. Griffing?s method I, model II, with random effects, also known as complete diallel mating design, was included in a broader context of a methodology for both statistical and genetic analysis of the experimental data. An algorithm was developed based on the proposed methodology and tested on the experimental data as well as on other two simulated scenarios of complete diallel mating design with respectively three and seven parents fed with random values. The results have illustrated the pattern suitability for the quantitative inherited trait as was the height of the apple seedlings, respectively the speed of growth of the very young plants. In addition, the proposed algorithm can help young researchers to understand and use adequately for different similar traits the complexity of statistical and genetic analysis of the full diallel mating design. The information obtaining using diallel crosses offers the breeders the possibility to choose the appropriate breeding and selection strategies for quantitative traits.