Evaluation of Bread Wheat (Tritium aestivum L.) Genotype in Multi-environment Trials Using Enhanced Statistical Models

In varietal selection field trials, spatial variation and genotype by environment (GxE) interaction are frequent and present a major challenge to plant breeders comparing the genetic potential of several cultivars. To consistently select superior cultivars that increase agricultural production, bread wheat breeding studies must be evaluated using efficient statistical techniques. By modeling the interactions of geographical field trends and genotypes by environment interaction, this work aimed to forecast the genetic potential of bread wheat varieties across settings and improve selection tactics. The dataset utilized in this investigation consisted of sixteen multi-environment trials (MET) that were carried out using a randomized complete block design (RCBD), with two replications arranged in plot arrays of rows and columns. The findings showed that the factor analytical and spatial models were effective ways to analyze the data for this study under the linear mixed model. By ranking average Best Linear Unbiased Predictions (BLUPs) within clusters, the 16 bread wheat environments were grouped into three mega environments (C1, C2, and C3) based on yield. This served as a selection indicator. Ranking average BLUPs helped in the selection of superior and stable genotypes. The first cluster (C1)'s mean BLUP values were used to score the genotypes' performance; C2 and C3 were excluded because of their limited genetic variety and low genetic connection with the other trials. The genotypes with the highest potential based on this cluster were EBW192346 and EBW192347, chosen for a subsequent verification study to release a variety. The estimates for variance component parameters ranged from 0.013 to 3.024 for genetic variance and from 0.072 to 0.37 for error variance. Hence, scaling up the use of this efficient analysis method will improve the selection of superior bread wheat varieties.