Within the scope of current genetic diversity analyses, population structure and homozygosity measures are independently analysed and interpreted. To enhance analytical power, we combined the visualization of recently described high-resolution population networks with runs of homozygosity (ROH). In this study, we demonstrate that this approach enabled us to reveal important aspects of the breeding history of the Haflinger horse. We collected high-density genotype information of 531 horses originating from seven populations which were involved in the formation of the Haflinger, namely 32 Italian Haflingers, 78 Austrian Haflingers, 190 Noriker, 23 Bosnian Mountain Horses, 20 Gidran, 33 Shagya Arabians, and 155 Purebred Arabians. Model-based cluster analysis identified substructures within Purebred Arabian, Haflinger and Noriker that reflected distinct genealogy (Purebred Arabian), geographic origin (Haflinger) and coat colour patterns (Noriker). Analysis of ROH revealed that the two Arabian populations (Purebred and Shagya Arabians), Gidran and the Bosnian Mountain Horse had the highest genome proportion covered by ROH segments (306Mb - 397Mb). The Noriker and the Austrian Haflinger showed the lowest ROH coverage (228Mb, 282Mb). Our combined visualization approach made it feasible to clearly identify outbred (admixture) and inbred (ROH segments) horses. Genomic inbreeding coefficients (FROH) ranged from 10.1% (Noriker) to 17.7% (Purebred Arabian). Finally it could be demonstrated, that the Austrian Haflinger sample has a lack of longer ROH segments and a deviating ROH spectrum, which is associated with past bottleneck events and the recent mating strategy favouring out-crosses within the breed.
https://doi.org/10.1093/jhered/esx114Population networks associated with runs of homozygosity reveal new insights into the breeding history of the Haflinger horse, Journal of Heredity,