Genetic improvement in the presence of genotype by environment interaction

Although the underlying mechanisms for triggering genotype–environment (GE) interaction are poorly understood, the potential impacts of GE interaction on genetic improvement are well recognized. Genotype–environment interaction may be classified into three levels: breed, individual and gene–environment interactions. Three measures of GE interaction (genetic correlation, interaction correlation, and commonality of individuals selected between environments) are discussed. Three options are currently available to deal with GE interaction: environmental, breeding and marker‐assisted approaches. Three possible selection strategies for improving global net merit were outlined: (i) selection of a specific genotype for each environment; (ii) selection in a single environment alone for overall response across environments; and (iii) global optimum index selection for high stability and average performance across environments. Global optimum index should be the method of choice from the standpoint of global marketing. Because of the complexity of GE interaction, it is impossible to develop a general strategy to deal with different types of GE interaction. Each type of interaction requires its own solution, depending upon a combination of the following six factors: (i) the intensity of GE interaction; (ii) relative economic weights among environments; (iii) the size of environments; (iv) the nature of environments; (v) the nature of GE interaction; and (vi) selection intensity. Profitability is a major concern in animal production. Extra genetic gain does not necessarily mean extra profit. Does additional genetic gain justify the associated costs of dealing with GE interaction? This is a fundamental issue that needs to be considered before a specific breeding strategy for GE interaction is developed.