Investigation of the Sweetpotato Trial Network for Efficient and Cost-effective Variety Development in Uganda

Late-stage evaluation of sweetpotato breeding trials is routinely conducted at multiple locations, years, or sea-sons to provide a better understanding of the genotype-environment interaction and test for genotype stability. In addition, increased testing serves to reduce the experimental error and therefore provide a better estimate of the actual genetic component. This increased testing comes with a huge financial burden for the breeding pro-gram. The aim of this study was to determine the variance components of multi-environment Advanced yield trials (AYT) and identify the most appropriate sweetpotato breeding trial network. The optimization of the trial network in terms of number of replications, locations, and seasons would ensure efficient resource allocation to further streamline the breeding program strategy. The storage root yield data obtained from 13 AYTs of the sweetpotato breeding program in Uganda was used for this study. In each trial, a total of 23 genotypes were es-tablished in three replications using Randomised complete block design (RCBD) and evaluated for three sea-sons. The data was analysed using mixed models to estimate the variance components used for the simulation of the trial network. The genotype effect (0.21) provided the highest proportion of variance, followed by genotype by season (0.14). Based on the simulation results, the best trial network for about the same number of genotypes is 4 locations, 2 seasons, and 3 replications. This network minimizes the standard error of the difference and, at the same time, maintains the one-year (2-season) evaluation for quicker variety development. This article has also provided a guideline for a trial network for earlier and later-stage gates with genotypes in the range of 06–120 evaluated in the same target population of environments. Keywords: multi-environment trials, resource optimisation, stage gate, sweet potato breeding, trial network, Uganda