This dissertation develops mixed family selection for Pacific oysters using marker-based pedigree reconstruction. It focuses on improving the efficiency of parentage assignment, determining the optimum life stage to mix oyster families for rearing and selection, comparing mixed-family and separate-family selective breeding in the field, and applying the mixed method to estimate the heritability of shell shape. We developed novel computer software, P-LOCI (available at http://marineresearch.oregonstate.edu/genetics/ploci.htm), which identifies the most efficient set of codominant markers for assigning parentage, accounting for marker linkage, mating design, null alleles and genotyping error, and found that the most efficient group of loci for assignment is not necessarily comprised of the top individually ranked loci, or best for all populations. We determined the optimum time to mix oyster families for rearing and selection in the field; overall, planting size is the most prudent time to mix families for MFS, due to high variability in family representation produced during larval and nursery stages. Mixing families at stages earlier than this for selection on field traits would require pre-planting genotyping of large samples for estimation of initial family representation, which would add substantial cost, or other special considerations. Rearing mixed family groups of oysters in the field yielded very similar results to rearing the same families separately, (r = 0.817 for two-site average individual weight at harvest), demonstrating it is unlikely associative effects are of great importance in the Pacific oyster. Our results show that the mixed method was well-suited for individual traits and walk-back selection, but would incur higher costs than the separate method to estimate survival with lower precision. Finally, we utilized the mixed method to estimate the heritability of shell shape using midparent-offspring regression; we estimated shell depth heritability as 0.404 ± 0.14 and shell width as 0.287 ± 0.11, nearly equal to the only other study for the Pacific oyster, demonstrating potential for selective breeding on these traits in this U.S. population, and similar results between methods. Overall, we found that mixed-family rearing is viable for Pacific oyster breeding, given some important restrictions.
Free download: printable .pdf [National Sea Grant Library]