Towards deciphering the structure of long homozygous stretches in cattle genome

Abstract

Up-to-day there is no as universally accepted software tool and threshold parameters to identify runs of homozygosity (ROH). The relative position of POH segments in the cattle genome has not been studied extensively. Specific objective of this study was to evaluate the effect of allowed missing and heterozygous SNPs in ROH on their number, on the estimate of inbreeding level, and on structure of ROH segments in the cattle genome. In this study 371 Holsteinized cows from six herds were genotyped with BovineSNP50 array. To identify ROH, the consecutive and sliding runs were carried out with detectRUNS and Plink tools. Neither effect was shown for missing SNPs genotype calls. Allowing even one heterozygous SNP resulted in significant bias of ROH data. Furthermore, the sliding runs identified less ROH than consecutive runs. The mean coefficient of inbreeding across herds was 0.111 {+/-} 0.003 and 0.104 {+/-} 0.003 based on consecutive and sliding runs respectively. It was shown how, using the heterozygous SNPs in ROH, may be possible to derive a distribution of ROH segments in the cow genome. We suggested it was similar to normal distribution. Furthermore, frequency of ROH in the chromosomes did not depend on their length. Of 29 chromosomes, the most abundant with ROH were BTA 14, BTA 7, and BTA 18. The result of this study confirmed more accurately identification of ROH with consecutive runs, uneven their distribution in the cattle genome, significant bias of the data due to allowing heterozygous SNPs in ROH.