A cautionary tale! The hazards of genotype imputation.

Although imputation of missing SNP results has been widely used in genetic studies, claims about the quality and usefulness of imputation have outnumbered the few studies that have  questioned its limitations. We studied the  imputation of  variants that have become frequent  due to selection,  and compared imputed genotypes with directly genotyped data, then examined the haplotype pairs of all  individuals with discrepant and missing genotypes.   The errors are non random. Most incorrect and missing imputations result from long haplotypes that are evolutionarily closely related to those carrying the derived alleles, while some relate to rare and recombinant haplotypes. This is a cautionary tale.

"The hazards of genotype imputation in chromosomal regions under selection: a case study using the Lactase gene region" Ann Hum Genet 2021.  Aminah T Ali, Anke Liebert, Winston Lau, Nikolas Maniatis and Dallas M Swallow. (https://pubmed.ncbi.nlm.nih.gov/34523124/)