ROH inbreeding and within-breed structure of Red Steppe cows based on SNP genotyping: relevance for veterinary-biological monitoring of breeding herd

The aim of this study was to assess genomic diversity, ROH-based inbreeding and within-breed structure in Red Steppe cows as complementary indicators for veterinary-biological monitoring of a breeding herd. The study included 40 cows from the breeding female stock of the Agrofirma Sogratl herd in the Republic of Dagestan, Russia, genotyped using the Illumina BovineSNP50 BeadChip. After marker-level quality control, 42,105 autosomal SNPs were used for heterozygosity and population-structure analyses. A separate reproducible ROH branch included 42,242 autosomal SNPs. Individual observed heterozygosity averaged 0.377 ± 0.012 and ranged from 0.335 to 0.389. Runs of homozygosity were detected in PLINK v1.9 using a minimum length of 1 Mb, at least 50 SNPs per run, a maximum inter-marker gap of 1,000 kb, and the documented sliding-window settings. In total, 87 ROH segments were detected in 34 of 40 animals. Mean F_ROH was 0.00568, the median was 0.00518, and the range was 0-0.01814; mean total ROH length was 14.25 Mb per animal. The three animals with the greatest total ROH length had 45.51, 45.29 and 40.13 Mb, respectively, but even these values remained within a low overall inbreeding range for the studied herd. ADMIXTURE was evaluated with 10-fold cross-validation and repeated runs across K = 2-6. The lowest median cross-validation error was obtained at K = 2 (0.608), whereas K = 3 yielded 0.630 and progressively higher values were observed for larger K. Therefore, K = 2 was retained as a parsimonious working description rather than evidence of two biologically isolated subpopulations. In the 2025 PCA analysis used in this study, PC1 and PC2 accounted for 15.7% and 13.9% of the reported variation, respectively. Because PCA and ADMIXTURE were derived from the same genotype matrix, their qualitative agreement was interpreted as complementary description rather than independent validation. The results indicate low average ROH-based inbreeding with measurable between-animal heterogeneity. SNP-based heterozygosity, ROH metrics and population-structure analyses can therefore complement pedigree, health, reproductive and productivity records in herd monitoring, but should not be used as stand-alone criteria for culling or mating decisions.