Low-cost SNP array for genotype imputation optimized for breeding programs in the fish pacu Piaractus mesopotamicus

Pacu (Piaractus mesopotamicus) is one of the main native fish species for aquaculture in South America. Cost-effective genotyping strategies, involving low-density SNP arrays followed by genotype imputation, can facilitate the broader adoption of genomic selection for this species. In this study, we developed a 1K SNP array to be applied with genotype imputation strategies in breeding program of pacu. First, we assessed imputation accuracy across various SNP densities (9K, 7K, 5K, 2K, 1K, and 0.5K) from the commercial Axiom 30 K SerraSNP array to select the best subset for further analysis. Then, 15,166 (91.1%) imputed SNPs with high accuracy were obtained when we selected 1,068 SNPs to comprise the 1K SNP array (average accuracies of 0.90 ± 0.12 at the SNP level and 0.92 ± 0.06 for 290 genotyped individuals). The 1K SNP array was developed using the Agriseq tGBS platform. In the evaluation of array performance using 95 individuals, no discordant genotyping was observed between the Agriseq tGBS platform and the Axiom 30 K SerraSNP array. Ninety-one individuals had a call rate of 90% and above, and 92% of the SNPs were identified in at least 90% of the samples. Additionally, an average minor allele frequency (MAF) of 0.356 (± 0.108) and individual heterozygosity values ranging from 0.321 to 0.470 were found. The results showed that the constructed low-density SNP array exhibited high imputation accuracies and can explore genetic variations within pacu populations, making it a cost-effective tool for enhancing pacu production through genomic selection.