Comparison of phenotypic and marker-assisted selection for quantitative traits in sweet corn

This investigation was designed to empirically compare the efficiency of marker-assisted selection (MAS) and phenotypic selection (PS) in enhancing economically important quantitative traits in sweet corn (Zea mays L.). Marker-assisted selection and PS were applied to three F-2:3 base populations (CO) with either the sugary1 (su1), sugary enhancer1 (se1), or shrunken2 (sh2) endosperm mutations. One cycle of selection was applied for both single and multiple traits including seedling emergence, kernel sucrose concentration, kernel tenderness, and hedonic rating (taste panel preference). Twenty percent of the families in each of the base populations were selected and intermated to constitute MAS- and PS-based C-1 composite populations. Selection efficiencies were evaluated on the basis of gains over one cycle and estimated evaluation costs. A total of 52 paired comparisons were made between MAS and PS composite populations. In 38% of the paired comparisons, MAS resulted in significantly higher gain than PS across the three C-1 composite populations, while PS was significantly greater in only 4% of the cases. The average MAS and PS gain across all composite populations and selected traits, calculated as percent increase or decrease from the randomly selected controls, was 10.9% and 6.1%, respectively. Use of MAS is most appropriate when traits are difficult and costly to measure. However, for some traits, the higher gain from MAS can compensate for the higher costs of MAS. It was concluded that incorporating DNA markers to traditional breeding programs could expedite selection progress and be cost-effective.