Abstract
BACKGROUND: The aim of this study was to assess genome-wide autozygosity in a Nellore cattle population and to characterize ROH patterns and autozygosity islands that may have occurred due to selection within its lineages. It attempts also to compare estimates of inbreeding calculated from ROH (F(ROH)), genomic relationship matrix (F(GRM)), and pedigree-based coefficient (F(PED)). RESULTS: The average number of ROH per animal was 55.15 ± 13.01 with an average size of 3.24 Mb. The Nellore genome is composed mostly by a high number of shorter segments accounting for 78% of all ROH, although the proportion of the genome covered by them was relatively small. The genome autozygosity proportion indicates moderate to high inbreeding levels for classical standards, with an average value of 7.15% (178.70 Mb). The average of F(PED) and F(ROH), and their correlations (- 0.05 to 0.26) were low. Estimates of correlation between F(GRM)-F(PED) was zero, while the correlation (- 0.01 to - 0.07) between F(GRM)-F(ROH) decreased as a function of ROH length, except for F(ROH > 8Mb) (- 0.03). Overall, inbreeding coefficients were not high for the genotyped animals. Autozygosity islands were evident across the genome (n = 62) and their genomic location did not largely differ within lineages. Enriched terms (p < 0.01) associated with defense response to bacteria (GO:0042742), immune complex reaction (GO:0045647), pregnancy-associated glycoproteins genes (GO:0030163), and organism growth (GO:0040014) were described within the autozygotic islands. CONCLUSIONS: Low F(PED-)F(ROH) correlation estimates indicate that F(PED) is not the most suitable method for capturing ancient inbreeding when the pedigree does not extend back many generations and F(ROH) should be used instead. Enriched terms (p < 0.01) suggest a strong selection for immune response. Non-overlapping islands within the lineages greatly explain the mechanism underlying selection for functionally important traits in Nellore cattle.