Fine-scale estimation of inbreeding rates, runs of homozygosity and genome-wide heterozygosity levels in the Mangalarga Marchador horse breed

Abstract

With the availability of high-density SNP panels and the establishment of approaches for characterizing homozygosity and heterozygosity sites, it is possible to access fine-scale information regarding genomes, providing more than just comparisons of different inbreeding coefficients. This is the first study that seeks to access such information for the Mangalarga Marchador (MM) horse breed on a genomic scale. To this end, we aimed to assess inbreeding levels using different coefficients, as well as to characterize homozygous and heterozygous runs in the population. Using Axiom ® Equine Genotyping Array—670k SNP (Thermo Fisher), 192 horses were genotyped. Our results showed different estimates: inbreeding from genomic coefficients (FROH) = 0.16; pedigree-based (FPED) = 0.008; and a method based on excess homozygosity (FHOM) = 0.010. The correlations between the inbreeding coefficients were low to moderate, and some comparisons showed negative correlations, being practically null. In total, 85,295 runs of homozygosity (ROH) and 10,016 runs of heterozygosity (ROHet) were characterized for the 31 horse autosomal chromosomes. The class with the highest percentage of ROH was 0–2 Mbps, with 92.78% of the observations. In the ROHet results, only the 0–2 class presented observations, with chromosome 11 highlighted in a region with high genetic variability. Three regions from the ROHet analyses showed genes with known functions: tripartite motif-containing 37 (TRIM37), protein phosphatase, Mg2+/Mn2+ dependent 1E (PPM1E) and carbonic anhydrase 10 (CA10). Therefore, our findings suggest moderate inbreeding, possibly attributed to breed formation, annulling possible recent inbreeding. Furthermore, regions with high variability in the MM genome were identified (ROHet), associated with the recent selection and important events in the development and performance of MM horses over generations.