SNP genotyping allows for SNPs associated with a phenotype to be identified and used, but SNPs with a smaller effects can be overlooked. Genomic Selection would fit all SNPs by including each one as a regression dependent on all other SNPs. The allele substitution effect would be random, and this would avoid overestimating the impact of a SNP. The Genomic Estimated Breeding Value can then be calculated for an individual from the SNP estimates. The SNP estimates can be obtained by BLUP, Bayes-A, or Bayes-B methods. The Genomic Selection method would allow the breeding value of individuals to be calculated, rather than just those with a certain phenotype. However, SNP re-estimation would have to be updated continuously because the SNP effects are from the training population, and other genetic interactions can contribute. Commercial pigs are often crossbred, so the pure parental lines are where genetic selection is conducted. Also, the parental lines are often kept in a facility with a higher biosecurity. If crossbred phenotypes could be recorded the parental breeding value could be more accurately estimated, but cost and the use of mixed semen makes this difficult. Genomic Selection allows the SNP effect to be estimated for crossbreeds of a phenotype, and eliminates the need for tracking pedigrees. High-density SNP genotyping is still not cost-effective for individual analysis and using smaller panels still require a large amount of SNPs. Instead, genotype selection of around 400 SNPs can be done on  individuals, and then portions of high-density SNPs from the parents used. Genomic Selection has the potential to provide more accurate breeding values, and selection effects seen more clearly in production settings.