Two types of biplots to integrate multi-trial and multi-trait information for genotype selection

Abstract

Genotype × environment interaction (GE) and unfavorable associations among breeding objectives are the two key challenges in genotype evaluation and selection. Dealing with GE includes utilizing repeatable GE and accommodating nonrepeatable GE, and analytical tools for both steps have been developed in recent years. The genotype by yield × trait (GYT) analysis was also developed to address the issue of genotype selection based on multiple traits. However, a method to integrate both multi-trial and multi-trait information has been lacking. The purpose of this study was to fill the gap. Two types of biplots were described and demonstrated, using an oat (Avena sativa L.) dataset as an example. The G + GE biplot of GYT index graphically displays the mean and stability of the genotypes, considering all breeding objectives. The GYT biplot across trials graphically displays the overall superiority and the strengths and weaknesses of the genotypes, after accommodating the nonrepeatable GE. The two types of biplots rank the genotypes in the same order of superiority; they complementarily provide a complete picture of the genotypes and allow confident genotype evaluation, selection, and recommendation.