Genetic gain prediction by different family-selection strategies

The choice of the best selection method is dependent on several variables, such as the heritability of the trait under selection, the resources available and the breeders' abilities. The objective of this study was to estimate the prediction of genetic gains from adopting different strategies of family selection. The families considered were obtained from 92 biparental crosses from the Flowering and Crossing Station of Serra do Ouro in Murici- AL, Brazil, in 2011. The mixed models approach was used for the prediction of the genotypic values and ranking of the families for cane yield (t/ha). The selection strategies considered were: individual BLUP selection (BLUPi), individual simulated BLUP selection (BLUPis), and Australian Sequential Selection (BLUPseq), and as a parameter of comparison (negative control) of the other selection methods, a random selection was also used. Overall, the method with the highest expected genetic gain was the BLUPseq (9.8%), followed by BLUPis (9.3%) and BLUPi (6.4%). As expected, the random selection method was very inefficient due to the large proportion of clones selected with low genotypic values. In practical situations, both methods of family selection may be recommended. However, the BLUPi method presents practical implementation difficulties.