Analysis and interpretation of genotype by environment interaction using cluster analysis
Loading...
Date
2005-11-10
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Agricultural Research, plant breeders conduct multi-environmental testing of genotypes to select the best genotype for farmers to grow on their farms. The genotype with relatively high mean yield across the test environments is usually considered as the best genotype to be selected. The problem with this selection is that some genotypes that might have performed well in some sites may have performed badly in some other sites. This phenomenon often referred to as genotype by environment (GE) interaction complicates the selection process.
This thesis is therefore concerned with the analysis and interpretation of genotype by environment (GE) interaction which occurs when a wide range of genotypes are tested over a wide diversity of environmental conditions. Various statistical methods for the analysis of GE interaction have been proposed. One of the methods currently in use, Cluster analysis technique that is efficient at extracting information on genotype by environment (GE) interaction is of interest.
Cluster analysis techniques are appealing ways of explaining GE interaction, but can suffer from lack of clarity on the objectives of the analysis. Building on the work reported by Byth et. al., (1976), cluster analysis is extended with the particular objective of predicting the “best” genotype for each environment. The objective is achieved by attaching to each environment the predicted genotype. This approach is illustrated using Wheat yield data consisting of 40 genotypes in 45 environments from Centre for Maize and Wheat Yield Improvement Trial centre in Mexico (CIMMYT).
Description
A thesis submitted to the Department of Mathematics in partial fulfilment of the requirements for the degree of Master of Science, 2005