Análise da estabilidade e adaptabilidade de genótipos de milho na resistência a doenças por meio dos modelos Gammi

Abstract

The AMMI methodology is widely used to study the stability and adaptability of genotypes to different environments. The vast literature on the subject facilitates the implementation and interpretation of results. However, there are few examples of generalization of the AMMI model by the generalized additive main effects and multiplicative interaction models (GAMMI). This study aimed to identify the Genotype by Environment (GxE) interactions using the methodology GAMMI, taking as dependent variable the incidence of two maize foliar diseases, the Gray leaf spot and Diplodia blight, whose pattern does not fit a normal distribution of probability. The process involves concepts formalized from the AMMI model, expanded by the generalized linear models (GLM) and amplified by the quasilikelihood models. Two samples of 36 genotypes observed in 9 locations were analyzed. Genotypes 1F5634 and 1D2195 were identified as stable and widelyrecommended for their reaction to gray leaf spot (Cercospora). For the Diplodia blight, cultivars 1F5864 and P30F35 were considered the most stable, according to the quasi-likelihood model with logit link function and varianc e V (μ) = [μ(1-μ)]2 . However, both had a high incidence of the disease, and therefore not widely recommended. The computational process was developed with the software R, an open source language, available at http://www.rproject.org/, featuring the functionality, accessibility and relevance of the process.