Abstract:

The objective of this study was to evaluate the efficiency of spatial statistical analysis in the superior genotype selection of plant breeding using the data from a yield trial of wheat with 56 test lines in a block design. The spatial analysis used linear mixed model with spatial covariance for residuals which was estimated using semivariogram defined in geostatistics. Results showed a typical spatial correlation of residuals, which allowed a reduction in standard errors of line effect estimate and a greater efficiency of F- and t-test in discrimination of lines when spatial analysis was applied. Furthermore, in estimation of line effect the spatial analysis was less influenced by assay condition variation, and led to a different ranking and selection of genetic materials, in comparison with traditional analysis of variance.

Key words: Spatial statistics, Plant breeding, Superior genotype selection