Biometrical approaches for investigating genetic improvement in wheat breeding in South Africa

[摘要] English: Wheat is the biggest winter cereal crop in South Africa and the second largest cereal to feed the population of South Africa. The population of South Africa grows with approximately one million people a year. Consistent wheat production is necessary for food security and is therefore of extreme agricultural and economic significance. Future production increases depend on the ability to improve, or at least maintain, the rate of increase to feed the population. The study was undertaken to investigate genetic improvement (genetic advance) in wheat by various statistical methods of analysis. This was done to determine the most suitable procedure to evaluate genetic improvement in the three wheat production areas of South Africa, namely the Western Cape Province, the Free State province and the irrigation areas. The second objective of this study was to demonstrate the trend of yield and the two quality traits [HLM (hectolitre mass) and protein content] over 16 years (1995-2010) by various statistical techniques. The third objective was to compare the AMMI (additive main effects and multiplicative interaction) and the GGE (genotype plus genotype-by-environment interaction) analyses in assessing genotype-by-environment interaction (GEI) for yield and the two quality traits. The fourth objective was to study the relationship among wheat grain yield and the two quality traits by various statistical techniques. Linear regression (TRET) and various variance component methods were investigated to determine genetic advance. The recommended method of determining genetic advance in this study is TRET. In the Western Cape elite trials TRET predicted a genetic advance of 1% per year and genetic advance estimated at 1% genetic improvement for protein content in the cultivar trials. No significant trend was observed in the elite trials of the Free State with TRET. Yield showed 0.5% and 0.6% per year improvement for the two planting dates of the eastern cultivar trials of Free State. A yield improvement of 0.3% per year improvement for the two planting dates of the central cultivar trials of Free State was determined. A genetic advance for yield of 0.7% per year was found in the warm region of the elite irrigation trials and 9% yield improvement per year for the first planting date of the eastern region of the cultivar irrigation trials. A negative trend was observed for the second planting date of eastern region of both elite and cultivar irrigation trials. The effects of GEI on yield and quality traits were studied by comparing the AMMI and GGE analyses. These methods portrayed similar results. An advantage of these techniques is their complementary nature. Although both models portray GEI in various biplots, the AMMI provides statistical evidence to the visual presentation of the GGE biplots. Pearson product moment correlation matrix provided a linear relationship among the variables studied. Principal component analysis (PCA), cluster analysis (CA) and discriminant analysis (DA) offered auxiliary information on the relationship among the factors (e.g. genotypes, years, localities and/or environments) and the variables. DA was not able to indicate direction of genetic improvement in either of the three production areas in this study.

[发布机构] University of the Free State