Accounting for spatial trends to increase the selection efficiency in potato breeding

Abstract

A crucial point in agricultural experimentation is to compare treatments with high accuracy. However, agricultural experimentation is prone to field heterogeneity, and a common source of error is the spatial variation between the plots used in an experiment. With plant breeding experiments, the high number of tested genotypes requires breeders to use large areas, which invariably increases the likelihood of spatial variation. The use of models that do not address this variation can lead to errors in selecting the best genotypes. Our goal was to evaluate the effects of two spatial models—first‐order autoregressive (AR1) and spatial analysis of field trials with splines (SpATS)—to control the spatial variation in 30 experiments from potato (Solanum tuberosum L.) breeding programs. Specifically, we sought to control for three traits: total tuber yield (TTY), marketable tuber yield (MTY), and tuber specific gravity (SG). The results obtained with the use of spatial models were compared with the base model (independent errors) based on precision, heritability, and the impact on the selection of the best clones. Spatial models were effective in controlling local and global errors and achieved greater accuracy and efficiency over the base model. The spatial approach also showed greater heritability for all analyzed traits. The spatial models led to differences in the clone ranking and consequently in the selection of the best clones. Thus, spatial analysis has the power to make more precise analyses, which leads to more accurate selections and should be used to analyze phenotype data of potato breeding programs.

Accounting for spatial trends in multi-environment diallel analysis in maize breeding

“…When comparing the NSPA and SPA models, the results demonstrated the importance of spatial analysis. Relevant studies on potato [ 80 ] and soybean [ 44 ] also presented satisfactory results when spatial trends were considered. Higher selective accuracies were encountered in the SPA models, reinforcing the importance and potential of the SPA analyses in crop breeding.…”

Phenomics-Assisted Selection for Herbage Accumulation in Alfalfa (Medicago sativa L.)

“…where all terms are the same as the model (1) other than the term ξ, which is the independent error random vector of residual, ξ∼ N (0, R e ), R e is the covariance matrix of ξ, and it is defined as: R e σξ 2 Σ c (ρ c )⊗ Σ r (ρ r ). Where ρ c and ρ r are the autocorrelation parameters for the spatial coordinates of row and column; Σ c (ρ c )and Σ r (ρ r ) represent the autoregressive correlation matrices; and ⊗ represents the Kronecker product ( Andrade et al, 2020 ).…”

“…A better way to control the spatial variation is to implement spatial analysis to detect and correct the variation patterns in multiple dimensions. Experimental units close to each other are expected to be higher correlated than those far apart, and improvements in model fitness and higher selection accuracy have been reported in plant breeding programs ( Andrade et al, 2020 ).…”

Phenotypic Diversity and Productivity of Medicago sativa Subspecies from Drought-Prone Environments in Mediterranean Type Climates

“…The first-order autoregressive anisotropic covariance structure (AR1 x AR1) was used to model spatial variation in row and column directions. This model is reportedly adequate to account for spatial variation in yield trials of cultivars and breeding lines [ 51 , 52 ]. For the unreplicated Australian experiments, the standard check information (Sardi 7) was used for calculating variance components and BLUPs in accordance with Piepho and Williams [ 53 ].…”