Optimizing selection indices in sugarcane seedlings

Selection of superior clones in sugarcane (Saccharum spp.) breeding programs follows a multistage process, starting in the first stage with selection of seedlings grown from seeds obtained from crosses. Effective selection of seedlings is important but notoriously difficult. Selection validity is affected by experimental error variation and interplant competition effects and is also complicated by the need to consider many different traits that affect economic value of cultivars (e.g., cane yield, sugar content, fiber content, disease resistance, etc.). In this study, we developed an optimized protocol for selection of sugarcane seedlings that balances the desire to maximize genetic gains but also be cost and labor efficient. A population of seedlings in commercial sugarcane breeding programs was first evaluated in a field trial (Stage 1). A subset of the population was then taken at random and grown in a subsequent trial (Stage 2) for evaluation in small plots. Minor modifications to application of the complete selection index that would greatly reduce labor requirements but have only a small impact on reducing genetic gains were then explored, to identify optimal selection protocols for routine application in commercial breeding programs. A procedure is recommended for routine selection of seedlings that involves (a) selection of families based on an index of four traits (stalk diameter, stalk height, trashiness, and pith rating, all derived from measurements of a small sample of seedlings in each family), followed by (b) sequential measurement and culling of individual seedlings in each selected family based on (in order) trashiness, stalk diameter, and pith rating.