One referred to a paper on multiple imputation but gave no further details, the other stated that "1000 imputation sequences" were used.
Only 89 sequenced animals from three breeds were used as imputation reference population and, therefore, imputed sequences were expected to have relatively low imputation accuracy, especially for SNPs with low minor allelic frequencies.
Imputation of sequence variants from related individuals or extended pedigrees can also boost imputation accuracy.
Simulations have shown that imputation of sequence information from a restricted number of highly informative individuals is feasible [ 21].
Therefore, imputation to sequence data using SNP genotypes is an attractive and cost-effective approach to obtain a large training set of sequenced individuals, and to investigate the benefit of using sequence data for relevant populations.
This may be due to imperfect imputation of sequence data due to the small reference size or may reflect that the density of the HD data was sufficient for QTL detection and further density increases only give small improvements.
The cost reduction in modern sequencing technology enables investigators to generate a reference panel for genotype imputation by sequencing a subset of the study sample.
This will likely include the DNA sequencing of individual animals [ 51], such as widely-used AI sires, and the imputation of sequences.
