Submitted to: American Society of Sugar Cane Technologists
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/21/2007
Publication Date: 8/27/2007
Citation: Edme, S.J., Tai, P.Y., Miller, J.D. 2007. Relative efficiency of spatial analyses for non-replicated early-stage sugarcane field experiments. Journal of American Society of Sugar Cane Technologists. 27:89-104.
Interpretive Summary: To improve efficiency in any plant breeding program, proper experimental designs and statistical analyses need be adopted to discriminate genotypes under selection by their true genotypic values. Replications and multi-locations trials, the accepted norms in plant breeding trials to remove the interaction of genotypes with their environments, cannot be used when large numbers of clones are the units of selection in early-stage sugarcane field experiments. Stage 2 in the Canal Point sugarcane breeding program involves field testing of about 1200-1500 genotypes without replications along with five replicated check cultivars. Spatial analyses were compared with the traditional method for their efficiency at removing field trends in the 2000 and 2002 Stage 2 field trials at Canal Point, FL. The two spatial analyses proved to be superior to the standard analysis, with one of them achieving much greater gains in efficiency (95 to 492%) depending on the trait being evaluated. Their suitability came from being able to adjust for the complex patterns of yield and related traits in the field. Breeders always want to appreciate interrelationships among the traits under selection. With the removal of field trends, some of the negative correlations between yield and quality traits disappeared or improved. The spatial analysis was implemented in 2002 to improve efficiency in selecting Stage 2 clones and genetic gains in the Canal Point breeding program. Of the 135 selections advanced to Stage 3, 35 clones selected exclusively by the spatial method were included. One of these 35 clones was subsequently selected to make the 13 clones normally advanced from Stage 3 to Stage 4, which is the last selection stage before release to the industry. The spatial analysis has identified one highly elite clone that is now planted in replicated tests at ten locations in the Canal Point program. This clone would have been dropped based on traditional selection methods.
Technical Abstract: In the early stages of sugarcane (Saccharum spp.) selection programs, testing in replicated plots is impractical with large numbers of clones and limited number of planting materials. Field trends are likely to affect the performance of these experimental genotypes and mask their true genetic potential. The purpose of this study was to evaluate different spatial analyses for their efficiency at accounting for field trends in early-stage sugarcane trials. A Moving Means (MM) and an autoregressive spatial (SP) methods were used to adjust genotype values in the 2000 and 2002 unreplicated Stage-II field experiments within the Canal Point (FL, USA) sugarcane breeding program. Variogram plots showed more complex patterns for fields of larger size and suggested the existence of both local trends and gradients across the fields. Based on reduced error variance for the checks, the relative efficiency (RE) obtained with MM was inconsistent and ranged from 4 to 188%. Based on average standard error of differences, gains in RE ranged from 95 to 492% with SP. Adjusting for field trends resulted in the improvement of phenotypic correlations among traits, even annulling some of the negative associations between cane yield and quality traits. An overall 75% congruence was obtained between the standard and SP, which corresponded to 35 selections being in disagreement to make the 135 clones normally advanced to the next stage. SP was the preferred method for giving consistent and greater gains in RE, thereby indicating its suitability for analysis and selection of early-stage sugarcane trials.