Reassembly of cattle immune gene clusters for quantitative analysis

Authors: Bickhart, Derek; HARRISON, DOROTHEA - The Pirbright Institute; SCHWARTZ, JOHN - The Pirbright Institute; Cole, John; HAMMOND, JOHN - The Pirbright Institute; Smith, Timothy - Tim

Submitted to: Research Workers in Animal Diseases Conference Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 12/2/2017
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Objective Animal health is a critical component of dairy cattle productivity; however, current genomic selection genotyping tools have a paucity of genetic markers within key immune gene clusters (IGC) involved in the cattle innate and adaptive immune systems. We sought to assemble alternative haplotypes of regions of the cattle genome that may harbor alleles that confer increased disease resistance or susceptibility. Methods We selected bacterial artificial chromosome (BAC) clones harboring key IGC genes from public and private libraries and sequenced the inserts with long-read sequencing technologies. We then used a hierarchical assembly approach to combine the assemblies into alternative haplotype regions. In order to identify single nucleotide polymorphism (SNP) markers that would be suitable for assays, we aligned sequence data from 125 Holstein bulls to the alternative haplotypes. These variants were then used on custom genotyping arrays to genotype a population of 1,800 Holstein cows with bovine tuberculosis resistance phenotypes. Results Using a hierarchical assembly approach with long-read sequencing technologies, we assembled nine IGC haplotypes from 48 pre-selected BACs. These haplotypes were compared to the recently released ARS-UCDv1.2 cattle reference assembly and showed good contiguity with that reference. Alignment of whole genome shotgun data from 125 Holstein bulls to these alternative haplotypes revealed 55,410 single nucleotide polymorphisms (SNP); however, many of these variant sites were unsuitable for use on custom genotyping arrays. Using model-based and machine-learning approaches, we selected 67 of these markers for custom genotyping. Approximately 60% of our markers had genotype call-rates greater than 80% in this population. Conclusions We demonstrate that a hierarchical assembly and variant-calling approach is able to identify suitable genetic markers to tag alternative IGC alleles in reference populations. This data will be used in genome-wide association analyses to identify suitable genetic markers to track disease resistance phenotypes in dairy cattle.