The effect of haplotype diversity on genotype imputation accuracy of Australian Hereford cattle

Accurate genotype imputation enhances genetic gain by improving the accuracy of genomic prediction. Haplotype diversity across chromosomes plays a crucial role in the accuracy of genotype imputation. This study aimed to investigate the effect of haplotype diversity on genotype imputation accuracy in Australian Hereford cattle. Genotypes were randomly masked from 80k to 15k, and then FImpute was used to impute missing genotypes to the original density with a reference population of 6,000 animals. Haplotypes were defined as non-overlapping 10-SNP length windows across the genome, and the number of unique haplotypes on each window was used to quantify haplotype diversity. The relationship between the sum, mean, and variance of the total number of unique haplotypes across all windows of chromosomes and the median imputation accuracy of chromosomes was investigated, and the Pearson's correlations of 0.79, -0.94, and -0.03 was observed between the median imputation accuracy of each chromosome and the sum, mean and variance of the total number of unique haplotypes across all windows of the chromosomes, respectively. Chromosomes with lower mean haplotype diversity and a higher number of common haplotypes exhibited higher imputation accuracy. These findings highlight that haplotype diversity within the reference population affects genotype imputation accuracy in Australian Hereford cattle. A knowledge of haplotype diversity, which quantifies the number and length of shared haplotypes, can help optimise the size of the reference population and the design of a region-specific or chromosome-specific marker density chip, thereby improving genotype imputation accuracy. It can also provide insights into recombination and genetic diversity patterns in the population to be considered in breeding programs, ultimately leading to improved genetic gain.