A linkage disequilibrium method to reposition single nucleotide polymorphism and improve genotype imputation accuracy

Incorrect positioning of single nucleotide polymorphisms (SNP) can affect imputation accuracy and decrease the accuracy of genomic prediction. This study aimed to develop a method to identify the most likely genomic position of the misplaced SNPs which have low imputation accuracy by fitting a Spline curve using linkage disequilibrium (LD) information. The accuracy of the method was validated by correctly identifying the masked position of 2,560 out of 45,918 SNP with a 100% correlation between the original and estimated positions. Candidate SNPs with low imputation accuracy (< 0.5) were assumed to be incorrectly positioned on the genome assembly. The pair-wise LD between these SNPs and other SNPs on the genome was used to fit a Spline curve. The Spline peak was considered the most likely position for the candidate SNPs. This LD-based method assigned the new position for 92% of the SNPs with low imputation accuracy and improved the mean imputation accuracy of these repositioned SNPs from 0.21 to 0.97 in Australian Brahman cattle.