Eight turbot (Scophthalmus maximus) and eight rainbow trout (Oncorhynchus mykiss) microsatellites were selected for parentage assignment in fish-farmed populations. The number of alleles, gene diversity, polymorphic information content, and the probabilities of exclusion of these loci were 8, 0.76, 0.73, and 0.55 in turbot and 4, 0.65, 0.59 and 0.39 in rainbow trout, respectively. The power of the markers for parentage assignment was assessed by computing the frequency of good and unique decisions (fgu) in a population of genitors defined by its allele frequencies and assuming three different types of mating schemes. The eight turbot microsatellites gave larger maximal mating schemes (the largest mating structure with a fgu >= 0.95) than the eight rainbow trout loci: 1 female (F) mated to 520 males (M) (paternity retrieval scheme), more than 140F x 140M (factorial scheme), and more than 15 000 independent pairs (natural population scheme) for turbot, and 1F x 88M, 34F x 34M, and 7000 independent pairs for rainbow trout. The variation of the fgu values with the number of loci confirmed that the turbot set of microsatellites was more efficient for parentage assignment than the rainbow trout markers.
Accuracy of BLUP breeding values in a factorial mating design with mixed families and marker-based parentage assignment in rainbow trout Oncorhynchus mykiss
