Using Pedigree and Genomic Data toward Better Management of Inbreeding in Italian Dairy Sheep and Goat Breeds

The inbreeding coefficient is an important parameter for livestock management. Small ruminant breeders and associations mainly rely on pedigree information, but genomic tools are gaining relevance, overcoming possible pedigree inconsistencies. This study investigates the relationship between pedigree-based and genomic inbreeding in two goat and four sheep dairy breeds. Pedigree and genomic data (medium-density SNPchip) were obtained for 3107 goats and 2511 sheep. We estimated pedigree depth (number of fully traced generations, FullGen) and inbreeding (F_PED), as well as two genomic inbreeding indexes, using runs of the homozygosity (F_ROH) and genomic relationship matrix (F_GRM). The correlation between the inbreeding coefficients was assessed. A linear regression model (LRM) was fitted for estimating F_PED from F_ROH. After quality control on genomic data, we retained 5085 animals. Mean inbreeding values were low, with higher F_ROH than F_PED and F_GRM. Breed differences can partially depend on different managements. The correlation between F_PED and F_ROH was the highest and directly related to pedigree depth. The best LRM was chosen for FullGen ≥4 and ≥6 for goats and sheep, respectively; after excluding animals with extreme residuals, a new refined regression equation was calculated. Since massive genotyping is not affordable to small ruminant breeders, it is important to understand the distinction and relationship between differently calculated inbreeding coefficients, also in view of the introduction of genomic enhanced breeding values. Our study highlights the importance of accurate pedigree information and, especially if not obtainable, of calculating genomic-based inbreeding coefficients. A better estimation of animals' relatedness contributes to improve animal breeding and conservation.