Estimates of strain additive genetic, heterosis and reciprocal effects for male proportion in Nile tilapia, Oreochromis niloticus L.

Data from the GIFT complete diallel cross with eight strains of Nile tilapia reared at seven different grow out environments were analyzed to evaluate the feasibility of using breeders from the best performing strains or strain combinations to increase male proportion in the population. Of the three strain genetic by test environment interaction effects evaluated, i.e., strain additive, strain reciprocal and strain heterosis, only the latter was found to be significantly different from zero (P < 0.05). However, it was difficult to see any trend with respect to the ranking of the strain total heterosis estimates across different environments, a prerequisite for a commercial utilization of these heterosis effects in a breeding program. Of those effects found to have a significant effect (P < 0.05) on male proportion, the environment effect accounted for 0.12%, the batch effect for 0.23%, the strain additive genetic effect for 0.06%, the strain reciprocal effect for 0.25% and the strain total heterosis effect for 0.42% of the total variation in male proportion. The largest difference between two strains was 13.3 ± 4.6 percentage points (P < 0.001) for the additive genetic effect, 12.0 ± 2.2 percentage points (P < 0.001) for the reciprocal effect and 5.7 ± 2.2 percentage points (P < 0.01) for the general heterosis effect. Average heterosis for male proportion was not significantly different from zero (1.8 ± 1.2 percentage points; P > 0.05), however for some strain crosses the total strain heterosis effect was substantial with 19.7 ± 4.4 percentage points (P < 0.001) as the largest difference between two strain crosses. The strain additive effect explained only 3%, the strain reciprocal effect 25% and the strain total heterosis effect 59% of the variation in total performance (strain additive + strain reciprocal + strain total heterosis) in male proportion. However, most of the variation in total heterosis was due to the specific heterosis effect which explained a large proportion (53%) of the variation in total performance in male proportion among the crosses, and thus is more important than any of the other studied strain genetic effects; i.e. the strain additive, reciprocal and general heterosis effects. Therefore to maximize gain in male proportion the strain cross or crosses with highest total performance should be chosen. It can be concluded that genetic improvement of male proportion through the use of breeders from the strain and strain crosses with highest male proportion in this study will have an insufficient impact for the immediate commercial applications under the tested farming conditions in the Philippines.