Year: 2010
Type: article
Abstract: Until recently, different commercial arrays with 300,000 to one million SNPs have been used for studies in human genetics.Samples from genome-wide association studies have been genotyped on different genotyping platforms, while much more markers were used for reference panels.To increase power of association studies, different data sets are combined by imputing the missing markers.Imputation techniques have proven to be very efficient and allelic imputation error rates were estimated to range between 0.5 and 2.5 %, depending on the study.Although markers densities are lower, these techniques can be applied in cattle too.Thanks to strong familial information and larger samples, low imputation errors rates (below one percent) can be achieved.Imputation error rates are directly influenced by marker density and genetic relationships between target and reference individuals.Even with arrays of 3000 SNPs, imputation can yield reasonable marker predictions.As in human genetics, the development of publicly available databases of influential animals genotyped at high density or even re-sequenced would be beneficial to the field.
Author: Tom Druet
Language: English
Subfield: Genetics
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W1580025105