Parameter optimization for runs of homozygosity (ROH) detection in the pig genome

The present study analyses the influence of various parameters on the accuracy and reliability of ROH identification procedures, as well as their relationship with inbreeding and genetic diversity in Duroc pig populations. The genotyping data of 70 Duroc pigs obtained using PorcineSNP80v1 chips were used, and the effects of LD pruning and ROH calling parameters were evaluated. It was found that the use of strict basic filters during the execution of the --homozyg command reduces the likelihood of false positives and contributes to the more accurate identification of ROH. The results show that LD pruning reduces the detection of random segments, allowing for a more accurate identification of the true ROH, and increases their total number through the elimination of errors. It can thus be concluded that the applied ROH search model leads to more reliable results.