Towards a sustainable tilapia breeding program in Tanzania

Sammanfattning: A structured breeding program is an important step towards sustainable tilapia aquaculture in Tanzania. The aim of the thesis was to generate baseline information for the establishment of tilapia breeding program in Tanzania. In the first study of this thesis we determined the genetic diversity between and within local and exotic Nile tilapia (Oreochromis niloticus) and Rufiji strains (Oreochromis urolepis urolepis) in Tanzania. Strains of FETA, Victoria, Igunga and TAFIRI had low genetic variation ranging from 0.057 to 0.1 while Kunduchi, Karanga and Ruhila showed highest genetic diversity from 0.214 to 0.212. Strong genetic differentiation was revealed between Karanga and the closely related strains of FETA, Lake Victoria, and Igunga with values from 0.533 to 0.548. STRUCTURE analysis revealed highly admixture among Karanga, Kunduchi, and Ruhila strains while FETA, Victoria, Igunga and TAFIRI showed no admixture. Higher genetic variation was also revealed among Rufiji strains compared to exotic and local Nile tilapia strains. High FST values (0.6- 0.8) were observed between Rufiji strains and the local or exotic Nile tilapia strains. The second part of this thesis based on a common garden experiment where the different strains of tilapia were compared in two environments with differing salinity. Differences in growth performance for all body traits were significant among Nile tilapia strains (P< 0.001) and strains ranked differently across the two environments. In freshwater environment, Karanga strain ranked first while TAFIRI strain ranked last. Regarding brackish water environment, Igunga strain was ranked first and the Victoria strain ranked last. Heritability estimates for harvest weight were low in both freshwater (0.10) and brackish water (0.09) environments compared to weight at tagging. Genetic correlations were low (0.35) for harvest weight and families ranked differently across the two environments indicating the existence of substantial GxE. Our results suggest that some strains are better suited than others to form the basis of a selective breeding program of Nile tilapia in Tanzania. Base population for the breeding program needs to strike the right balance between picking the best performing strains and having a broad genetic basis. Caution though is needed due to the high GxE across the tested environments.