Development and evaluation of a haplotype reference panel for low-coverage whole genome sequencing genotype imputation in turbot (Scophthalmus maximus)

Abstract

As a core molecular breeding tool, whole genome sequencing (WGS) can improve the speed of selection and breeding of superior variety and promote the progress of biotechnology-based breeding. However, the high cost of WGS limits its wide application for most aquaculture species. Low-coverage whole genome sequencing is a cost-effective alternative to genotype imputation to obtain a high density of genome-wide SNPs and ultimately Genomic selection (GS) and Genome-wide Association Studies (GWAS). In this study, we constructed the first haplotype reference panel for turbot (Scophthalmus maximus), which is an economically important marine farmed fish in China. We performed WGS of fish from major turbot aquaculture areas. We used 183 fish from which six populations as the reference group and 30 individual turbots from a seventh population as the validation group. In total, more than 239 million single nucleotide polymorphisms (SNPs) were identified in the reference group, and 4092,338 of them were screened for the construction of the haplotype reference panel. The reference panels were evaluated based on several factors that affect the accuracy of reference panel imputation. Twenty-two chromosomes had imputation accuracies ranging from 0.96 to 0.99, and there were no significant differences among them (p > 0.05). As the sequencing depth increased, the imputation accuracy increased and stabilized at 0.1 × , ultimately reaching a imputation accuracy of 0.96. As the minor allele frequency (MAF) or genotype probability (GP) increased, the imputation accuracy increased accordingly and eventually stabilized. The results indicated that the depth of coverage could range from 0.1 × to 1.0 × and that the imputation accuracies were all as high as 0.96, which would result in high-quality sequencing data. In conclusion, we constructed a reference panel of turbot haplotypes and determined its depth of coverage for low-coverage whole genome sequencing. This is a cost-effective alternative to whole genome sequencing to provide new ideas for fish breeding.