DNA pooling and low-density SNP genotyping platforms offer promise as an effective and cost-efficient method for genetic evaluation of aquaculture species, particularly in situations where individual family rearing is not feasible or practical. This method involves combining tissue samples from multiple individuals and genotyping them as a single sample. The objective of this study was to estimate the accuracy of the genetic evaluation based on breeding values using DNA pooling on a simulated low-density SNP tool.
We tested this method using simulated data of 4,000 individuals from 40 different families, assuming a single trait with a heritability of 0.30. A 5 percent error rate in genotyping was assumed. We tested different SNP densities ranging from 100 to 500 and two different pool sizes, 10 and 20 individuals per sample. We pooled individuals by phenotype, except for the top-performing 200 individuals from the tail of the phenotypic distribution, which had individual (non-pooled) simulated genotypes. These mimics realistic scenarios in aquaculture breeding.
Our results show that the accuracy of assigning individuals genotyped in pooled samples to their family of origin ranged from 0.53 to 0.91, with the highest accuracy achieved with a pool size of 10 and 500 SNPs. We also estimated the accuracy of the estimated breeding values (EBVs) from the pooled data, which ranged from 0.69 to 0.95, with the highest accuracy achieved when the pool size was 10 with 500 SNPs. Based on these simulated data, DNA pooling coupled with low-density SNP genotyping appears an effective and economical approach for some aquaculture selective breeding programs.