GGE叠图法─分析品种×环境互作模式的理想方法

GGE Biplot-An Ideal Tool for Studying Genotype by Environment Interaction of Regional Yield Trial Data

本文介绍一种分析作物区域试验结果的方法－ＧＧＥ叠图法。首先，将原始产量数据减去各地 点的平均产量，由此形成的数据集只含品种主效应Ｇ和品种－环境互作效应ＧＥ，合称为ＧＧＥ。对ＧＧＥ 作单值分解，并以第一和第二主成分近似之。按照第一和第二主成分值将各品种和各地点放到一个平 面图上即形成ＧＧＥ叠图。借助于辅助线，可以直观回答以下问题：（１）什么是某一特定环境下最好的 品种；（２）什么是某一特定品种最适合的环境；（３）任意两品种在各环境下的表现如何；（４）试验中品 种×环境互作的总体模式是怎样的；（５）什么是高产、稳产品种；（６）什么是有利于筛选高产、稳产品 种的环境。

This paper introduces a GGE biplot as a graphic method for analyzing the yield data of crop regional trials. First, the average yield at each environment is subtracted from the original yield data so that the derived data contains only genotype main effect G and genotype by environment interaction GE, which are collectively called GGE. The GGE data is then subjected to singular value decomposition and is approximated by the first two principal components. Plotting the first principal component against the second for all genotypes and environments results in a GGE biplot. In light of supplement lines, the following questions can be visually addressed: 1) what is the best cultivar for a given environment; 2) what is the most favorable environment for a given cultivar; 3 ) how do a pair of cultivars compare at different environments ; 4) what is the general pattern of genotype by environment interaction ; 5) which are the cultivars that are both high-yielding and stable; and 6) what are the environments that facilitates identification of such cultivars.