Heritability and genotype by environment interaction estimates for harvest weight, growth rate, and shape of Nile tilapia (Oreochromis niloticus) grown in river cage and VAC in Vietnam

Harvest weight has been the main selected trait for the GIFT strain of Nile tilapia (Oreochromis niloticus). However, growth rate and body shape are traits of increasing interest. In the Mekong Delta of Vietnam, intensive river-cage culture and low-input ponds are the most important production environments, whereas the breeding program is conducted in an intensive nucleus pond. The first objective was to estimate heritability and phenotypic and genetic correlations for harvest weight (HW), growth rate expressed as daily growth coefficient (DGC), condition factor (K), and shape expressed as ellipticity: mid-sagittal plane (EL-H), transverse plane (EL–T), and frontal plane (EH–T). The second objective was to estimate genotype by environment interactions (G × E), expressed as the genetic correlation (rg) between the nucleus and two production environments, cage and VAC (Vietnamese acronym for garden, pond and livestock pen), for these traits. Data were obtained from the 13th generation of selection of GIFT. Within the breeding nucleus, heritability was high for HW, DGC, but low for K, EL–H, EL–T, and EH–T. DGC was positively correlated with condition factor K (rg = 0.59), while the rg of HW with K was non-significant. This suggests that selection for harvest weight alone will not result in fish with higher condition factor. Genetic correlations between HW and body dimensions (L, H, T) were 0.89–0.98, but genetic correlations of DGC with ellipticity showed that fish selected for high growth rate will become more rotund rather than simply larger. GxE was minor for harvest weight and for growth, but substantial for shape traits. For DGC, genetic correlation was 0.77 between cage and VAC, but higher between the breeding nucleus and cage or VAC. For EL-H, substantial GxE (rg 0.54) was found between cage and nucleus pond. GxE was also found for EL–T between cage–VAC (rg 0.51), and for EH–T across all three environments, although with high standard errors of estimates. We conclude that selection in nucleus ponds will produce desired correlated responses in Nile tilapia grown in river-cages as they are expected to develop a more rotund and thicker body shape at the same length compared to fish grown in ponds.

► Heritability estimates for harvest weight, growth rate, condition factor and shape.
► Genetic correlation estimates for Nile Tilapia grown in cages and ponds.
► Minor genotype by environment interaction for harvest weight and growth rate.
► Substantial genotype by environment interaction for condition factor and shape.
► Ellipticity is a useful parameter to describe shape.