Use of Allele Specific Expression to Fine Map QTL and Enhance Genomic Selection

A new tool emerging to fine map QTL associated with treatments or conditions, such as disease, is allele specific expression (ASE).  We used ASE to find QTL associated with Marek's disease in poultry.  RNAseq was performed in replicated F1 birds produced by crossing 2 inbred lines and either exposed or not to MDV.  We then tested for intra-locus differential expression (DE) that is dependent on (E), i.e. a DE x E interaction.  This resulted in 1,750 ASE SNP loci putatively associated with MD.  In order to estimate QTL effects associated with the ASE SNPs, we genotyped and obtained phenotypes on 1,000 pullets exposed to MD from a resource population consisting of the F₆ generation of the above cross.  The data were analyzed using a mixed model analysis with SNPs and/or pedigree effects treated as random.  In the F₇ generation roosters from each family were genotyped and BLUP EBVs calculated based on either ASE SNP effects or pedigree information.  Roosters were then bi-directionally selected, based on the ASE SNP effects, and progeny tested for MD resistance.  The accuracy of selection was determined from the correlation of EBVs with progeny test performance.  Results showed that 1800 ASE SNPs accounted for a greater percent of the genetic variation than 3097 random SNPs uniformly distributed across the genome (83% vs. 67%), demonstrating that on a per SNP bases, ASE SNPs account for more than twice the genetic variance as random.  Response to bi-directional selection based on the ASE SNPs resulted in a 20% difference in MD in one generation.  The accuracy of selection was 61% higher for selection based on ASE SNPs compared to pedigree based.  The results suggest that ASE SNPs are functionally linked to causative polymorphisms in promoter regions inducing regulatory changes in genes that manifest in the phenotype.