GGEbiplot™

Once a biplot is constructed, it can be visualized in many ways to achieve diferent research objectives. The most important views include:

• Relationships among variables (environments, traits, etc.)
• Which won where or which had largest values for what traits
  
• Mean vs. stability (of genotypes)
• Discrimitiveness vs. representativeness (of test environments)
  

Biplot

The biplot (Gabriel 1971) is a scatter "plot" to approximately display a two-way table with labels for both the rows and the columns. Here "bi" means "both". If a two-way table can be sufficiently approximated by a rank-2 matrix, then it can be effectively investigated using a 2-D biplot. Likewise, a 3-D biplot can be used to effectively investigate a rank-3 matrix. A biplot always displays the most important patterns of a two-way table, no matter how large and complex it is. 

GGE biplot

A GGE biplot (Yan et al. 2000) is a biplot that displays the genotypic main effect (G) and genotype by environment interaction (GE) of a genotype-by-environment dataset. 

GGE biplot analysis

GGE biplot analysis is a system that consists of a set of biplot graphs that are designed to address various research objectives when genotype by environment two-way data are analyzed; the most important ones are:

1. Mega-environment analysis based on genetic correlation between location and the which-won-where pattern;
2. Test location evaluation based on their discriminating ability and representativeness; and
3. Genotype evaluation based on their mean performance and stability across a mega-environment

See REFERENCES for details.