Should the Markers on X Chromosome be Used for Genomic Prediction?

This study investigated the accuracy of imputation from LD (7K) to 54K panel and compared accuracy of genomic prediction with or without the X chromosome information, based on data of Nordic Holstein bulls. Beagle and Findhap were used for imputation. Averaged over two imputation datasets, the allele correct rates of imputation using Findhap were 98.2% for autosomal markers, 89.7% for markers on the pseudo autosomal region of the X chromosome, and 96.4% for X-specific markers. The allele correct rates were 98.9%, 91.2% and 96.8%, respectively, when using Beagle. Genomic predictions were carried out for 15 traits based on 54K marker data, imputed 54K for test animals, and imputed 54K for half of reference animals. GBLUP models with or without residual polygenic effect were used for genomic prediction. For all three data sets, genomic prediction using all markers gave slightly higher reliability than prediction excluding the X chromosome. Averaged over 15 traits, the gains in reliability from the X chromosome ranged from 0.3% to 0.5% points among the three data sets and models. Using a model with a G-matrix accounting for sex-linked relationship appropriately or a model which divided genomic breeding value into an autosomal component and an X chromosomal component did not led to better prediction based on the present data where all animals were bulls. A model including polygenic effect did not recover the loss of prediction accuracy due to exclusion of the X chromosome. It is recommended using markers on the X chromosome for routine genomic evaluation.