Analysis of Runs of Homozygosity in Cattle Breeds02 April 2013
Irish researchers looking into runs of homozygosity in beef and dairy breeds could have discovered that more inbreeding takes place in pedigree dairy cattle as analysis showed greater incidences of longer runs of homozygosity in dairy breeds.
Runs of homozygosity (ROH) are contiguous lengths of homozygous genotypes that are present in an animal due to parents transmitting identical sister haplotypes to their offspring. The length of an ROH can be an indication of consanguinity as the longer the ROH segment, the more likely recent inbreeding occurred in the pedigree.
However, long runs of homozygosity may also persist in outbred animals, believed to be due to the existence of multiple-megabase-scale ancestral haplotypes as a result of the mutation, linkage disequilibrium (LD) and recombination rates at their genomic location (Gibson, et al., 2006).
The presence of more ancient relatedness, which is often unaccounted for in an animal’s pedigree due to limitations of recording, can also account for shorter ROH lengths (Kirin, et al., 2010). As a result, past and present breeding practices will have played an important role in determining the length of the ROHs for a particular animal. The objective of this study was to quantify the levels of ROH that exist in different breeds of cattle.
Materials and Methods
Genotypes from 891 AI bulls from eight purebred dairy (n=360) and beef cattle breeds (n=420) and one Holstein-Friesian crossbreed (n=111), generated from the Illumina Bovine high-density (HD) panel were used. Only biallelic SNPs on the 29 autosomes were retained and both animals and SNPs with genotype rates <90 per cent were discarded.
Also, monomorphic SNPs and SNPs that deviated from the Hardy Weinberg equilibrium (P<0.0001) within breed were discarded. PLINK v1.07 was used to identify ROHs, by taking a sliding window of 5 Mb across the genome, and allowing no more than two missing genotypes and one possible heterozygous genotype per window. A minimum threshold length of 0.5 Mb was set for identification of ROH.
The HD genotypes were used for validation of the BovineSNP50 density panel to identify ROH by retaining the 48,734 SNPs (reduced HD panel) common to both the HD and the Illumina BovineSNP50 Beadchip. To establish that this reduced density also predicted the correct ROH length category for ROH, the quantity of ROHs that were correctly assigned to the HD ROH length category was plotted.
SNP involvement in a ROH was also calculated by counting the number of times each SNP appeared in a ROH. Measures of homozygosity (FROH) and the pedigree inbreeding coefficient were calculated and correlated for all animals that had all animals that had a complete generation equivalent value ?6 (n=230). FROH was calculated as FROH = ? LROH / Lauto where LROH was calculated as the sum of all ROHs >0.5Mb in length and Lauto is the total length of the genome covered by SNPs (i.e., 2.51 GB). The pedigree inbreeding coefficient was calculated according to the Meuwissen and Luo (1992) algorithm.
Results and Discussion
ROHs were common and frequent across all breeds as all animals had at least one ROH present in their genome that was between 1-5 Mb in length. The dairy breeds had a greater incidence of long ROHs (>30 Mb in length) indicative of recent inbreeding, whereas the Angus and Hereford had substantially more of their genome covered in shorter ROHs (1-5 Mb) suggesting a high level of distant ancestral relatedness possibly due to their geographical isolation during breed formation.
Analyses of SNP involvement in a ROH revealed that certain areas of the genome were more commonly involved in a ROH than others (Figure 1). In particular a region on chromosome 14 involving the TOX gene, an immune gene, was observed in 87.4 per cent of the HD genotyped population in a ROH. Strong correlations existed between the pedigree-based inbreeding coefficient and FROH (r=0.75;P<0.0001).
A total of 157,600 ROHs were identified using the HD panel, whereas only 19,078 ROHs were identified using the reduced HD panel. The majority of ROH that the reduced panel failed to recognise were between 0.5-1 Mb in length but almost all ROHs in length categories of >5Mb were recognised and assigned to the correct length category >70 per cent of the time, validating the BovineSNP50 array panel for use.
We show that the extent of a genome under ROH in cattle provides a sufficient predictor of the pedigree inbreeding coefficient, suggesting that in the absence of pedigree data, this method may be used to infer aspects of recent population history.
The authors gratefully acknowledge funding from Science Foundation Ireland.
Gibson J., Morton N.E., & Collins A., (2006). Hum. Mol. Genet.15:789-795
Kirin M., R. McQuillan, Franklin C.S., Harry C., McKeigue P.M., &Wilson J.F.,(2010). PLoSONE, 5:e13996
Meuwissen, T.H.E., & Luo, Z. (1992). Genet. Sel. Evol. 24, 305-313