GGE Biplot Analysis of Yield Stability in Multi-environment Trials of Lentil Genotypes under Rainfed Condition

Rahmatollah Karimizadeh,Ali Akbar Mahmoodi,Naser Sabaghni,Fariba Akbari,Barzo Roustami,Mohtasham Mohammadi,Faramarz Seyyedi

doi:10.15835/nsb529067

Rahmatollah Karimizadeh, Ali Akbar Mahmoodi + Show 5 more

Open Access

https://doi.org/10.15835/nsb529067

Copy DOI

Abstract

This investigation was done to study GE interaction over twelve environments for seed yield in 18 genetically diverse genotypes. Grain yield performances were evaluated for three years at four locations in Iran using a randomized complete block design. The first two principal components (IPC1 and IPC2) were used to create a two-dimensional GGE biplot that accounted percentages of 49% and 20% respectively of sums of squares of the GE interaction. The combined analysis of variance indicated that year and location were the most important sources affecting yield variation and these factors accounted for percentages of 50.0% and 33.3% respectively of total G+E+GE variation. The GGE biplot suggested the existence of three lentil mega-environments with wining genotypes G1, G11 and G14. According to the ideal-genotype biplot, genotype G1 was the better genotype demonstrating high mean yield and high stability of performance across test locations. The average tester coordinate view indicated that genotype G1 had the highest average yield, and genotypes G1 and G12 recorded the best stability. The study revealed that a GGE biplot graphically displays interrelationships between test locations as well as genotypes and facilitates visual comparisons.

Highlights

Plant breeders perform multi-environment trials (MET) to evaluate new improved genotypes across test environments, before a specific genotype is released for production to supply growers
Results of combined ANOVA for the yearly datasets are shown in Tab. 4, which provides a general picture of the relative magnitudes of effects of genotype, location and the interaction (G × L) due to G + L + genotype × location (GL) variations
The relatively large yield variation due to location, which is disjointed to genotype evaluation and mega-environment identification (Gauch and Zobel, 1996), justifies selection of site regression statistical model as the suitable tool for investigation of the multi-environment trials dataset

Summary

Introduction

Plant breeders perform multi-environment trials (MET) to evaluate new improved genotypes across test environments (several locations and over years), before a specific genotype is released for production to supply growers In such experiments, genotype × environment (GE) interaction is a commonly evaluated (Annicchiarico 2002; Kang, 1998; Karimizadeh et al, 2012a; Yan et al, 2007). A GE interaction refers to differential ranking of genotypes across environments and may complicate the selection process and recommendation of a genotype for a target environment (Ebdon and Gauch, 2002; Gauch, 2006) It may reduce the selection efficiency in different breeding programs because in a GE interaction, measured traits are less predictable and cannot be interpreted using main effects (genotype or environment) and need more analysis (Gauch et al, 2008). Plant breeders perform MET to select favorable genotypes based on both mean yield and performance stability; and to determine whether a test environment is homogeneous or should be divided into various mega-environments (Gauch, 2006; Yan and Kang, 2002)

Methods

Results

Conclusion