Use of genomics to simultaneously improve feed efficiency and meat quality in grow-finish pigs

For the pig industry, feed is the largest cost of production, with the largest proportion (approx. 74%) consumed in the grow-finish phase. Efforts to improve grow-finish feed conversion will significantly reduce production costs and consequently increase total profitability. Feed efficiency is usually defined as lean growth efficiency evaluated through feed intake, daily gain, loin depth and backfat thickness, but without any focus on pork quality. Better pork quality is another high priority for the pork industry to satisfy consumer demand for an enhanced eating experience. Selection based on pedigree and phenotype have shown that high emphasis on lean growth efficiency improves feed efficiency (lower feed intake and higher lean growth), but also reduces pork quality in terms of less marbling and tenderness, and lower pH and lighter meat color (Suzuki et al. J. Anim. Sci. 2005, 83: 2058-2065; Gilbert et al. J. Anim. Sci. 2006, 85: 3182-3188; Cai et al. J. Anim. Sci. 2008, 86: 287-298; 2008; Lefaucheur et al. J. Anim. Sci. 2011, 89: 996-1010). Improving lean growth efficiency without deterioration of pork quality is now a priority. We have performed large scale genomic studies with industry on both feed efficiency and meat quality for grow-finish pigs. We are utilizing whole genome sequence and different densities of SNP genotypes (60K, 80K and 650K) for genome-wide association studies. Our aims are to investigate the genetic architectures and relationships between these economically important traits and to develop genomic tools to increase genetic gain for both feed efficiency and meat quality. Significant genomic regions and markers have been identified for feed intake, loin depth, backfat thickness, meat color, pH, drip loss and marbling. Preliminary genomic prediction results show high accuracy for most feed efficiency component traits, but somewhat lower for meat quality. We are continuing to optimize the genomic selection methods by making good use of such abundant information to improve prediction power and validate the accuracy of the genomic estimated breeding values.