Genomic selection (GS) uses genome-wide molecular marker data to predict the genetic value of selection candidates in breeding programs. In plant breeding, the ability to produce large numbers of progeny per cross allows GS to be conducted within each family. However, this approach requires phenotypes of lines from each cross before conducting GS. This will prolong the selection cycle and may result in lower gains per year than approaches that estimate marker-effects with multiple families from previous selection cycles. In this study, phenotypic selection (PS), conventional marker-assisted selection (MAS), and GS prediction accuracy were compared for 13 agronomic traits in a population of 374 winter wheat (Triticum aestivum L.) advanced-cycle breeding lines. A cross-validation approach that trained and validated prediction accuracy across years was used to evaluate effects of model selection, training population size, and marker density in the presence of genotype × environment interactions (G×E). The average prediction accuracies using GS were 28% greater than with MAS and were 95% as accurate as PS. For net merit, the average accuracy across six selection indices for GS was 14% greater than for PS. These results provide empirical evidence that multifamily GS could increase genetic gain per unit time and cost in plant breeding.

Genomic selection (GS) is a method to estimate the breeding values of individuals by using markers throughout the genome. We evaluated the accuracies of GS using data from fi ve traits on 446 oat (Avena sativa L.) lines genotyped with 1005 Diversity Array Technology (DArT) markers and two GS methods (ridge regression-best linear unbiased prediction [RR-BLUP] and BayesCπ) under various training designs. Our objectives were to (i) determine accuracy under increasing marker density and training population size, (ii) assess accuracies when data is divided over time, and (iii) examine accuracy in the presence of population structure. Accuracy increased as the number of markers and training size become larger. Including older lines in the training population increased or maintained accuracy, indicating that older generations retained information useful for predicting validation populations. The presence of population structure affected accuracy: when training and validation subpopulations were closely related accuracy was greater than when they were distantly related, implying that linkage disequilibrium (LD) relationships changed across subpopulations. Across many scenarios involving large training populations, the accuracy of BayesCπ and RR-BLUP did not differ. This empirical study provided evidence regarding the application of GS to hasten the delivery of cultivars through the use of inexpensive and abundant molecular markers available to the public sector.T HE DECREASING COST of high-density molecular markers allows saturation of crop genomes with genetic markers and off ers an approach to predict genetic merit. Th ese markers can help capture the eff ects of many quantitative trait loci (QTL) controlling polygenic traits regardless of location of the QTL in the genome by using linkage disequilibrium (LD), the nonrandom association of alleles at diff erent loci (Falconer and Mackay, 1996). Meuwissen et al. (2001) proposed genomic selection (GS) based on prediction of the genetic value of individuals or the genomic estimated breeding values (GEBV) from high-density markers positioned throughout the genome. Because GS includes all markers, major and polygenic eff ects can be captured, potentially explaining more genetic variance (Solberg et al., 2008). Th erefore, the objective of GS is to predict the breeding value of each individual instead of identifying QTL for use in a traditional marker-assisted selection (MAS) program.Selection methods can be evaluated by measuring accuracy, a major component of the response to selection equation, R = irσ A , in which R is the response, i is the selection intensity, r is the accuracy, and σ A is the additive genetic standard deviation (Falconer and Mackay, 1996). As a general term in statistics, accuracy is the degree of similarity

The economic and environmental costs of weed management in soybean (Glycine max [L.] Merr.) have led to interest in developing weed suppressive soybean varieties to enhance traditional herbicide and tillage‐based approaches. We evaluated 104 inbred progeny from three crosses among elite soybean lines to determine optimal selection criteria for weed suppressive ability (WSA). We grew the lines in 1996 and 1997 at Becker, MN, an irrigated sandy site, and Rosemount, MN, a rainfed silt loam site, in a split‐plot, with and without white mustard (Brassica hirta Moench). We measured soybean height 7 wk after emergence (WAE), light interception 5 and 7 WAE, specific leaf area 7 WAE, and date of full bloom. We harvested aboveground mustard biomass 8 WAE and calculated each soybean line's WSA as the difference between mustard biomass when grown in competition with that line and the overall mean mustard biomass. We estimated genetic correlations between soybean morphological traits, WSA, and the agronomic traits lodging, maturity date, and yield. Soybean early height's heritability true(h2=0.64true) and genetic correlation with WSA (r = 0.81) made it an ideal selection criterion. Indirect selection on height increased predicted selection efficiency by 70% relative to direct selection on mustard dry weight. Restricted index selection combining information on early height and lodging or yield eliminated undesirable correlated responses of lodging and yield while maintaining genetic gain for early height and WSA. Nevertheless, continuing rapid gains in agronomic performance while incorporating WSA may be difficult.