Applying the AMMI models to understand genotype-by-environment (G E) interactions in rice reaction to blast disease in Africa

Abstract

Additive, Main-effects and Multiplicative Interaction (AMMI) models are being widely used for analysing main-effects and genotypeby-environment (G E) interactions in multilocation variety trials. This paperpresents an AMMI analysis ofthe reaction to rice blast( Pyricularia oryzae ) by 25 rice genotypes grown in unreplicated trials in six locations. The object is to gain insight into G E in rice blast, and identify genotypes with high and stable resistance to the disease. Our results showed that contributions to treatment main-effects were: genotype (33%), environment (29%), and G E (38%). Analysis of the interaction revealed decreasing magnitude of contributions to the G E sum of squares in the order: interaction principal component axis 1 (IPCA1) = 39%, IPCA2 = 22%, IPCA3 = 20%, and IPCA4 = 12%. Correlation analysis related IPCA3 to total rainfall in the first month of experimentation. Considerable diversity was observed among the environments and in the reaction of the differential entries to blast disease. An overlap between two differentials suggest one could replace the other. Four improved upland varieties (IRAT13, LAC23, ITA257, and ITA235) and three improved lowland plant types (ITA116, ITA118, and ITA120) showed high and stable resistance to blast. Compared with their parents, this suggests progress in varietal improvement against blast in Africa. The paper discusses some applications of AMMI in the analysis and interpretation of multilocation blast screening data.