LAUSR

Gastrointestinal nematode infections have caused expressive losses in sheep production worldwide. The improvement of host genetic resistance to worms has been used as a strategy to mitigate this problem. In this sense, the inclusion of genomic information has shown potential to increase the accuracy of prediction of breeding values and speed up selection. In this study, we aimed to compare estimates of genetic parameters and breeding values for traits that indicate the resistance to gastrointestinal nematode infection in Santa Inês sheep using the pedigree-based BLUP or including genomic information. There were 1478 animals in the pedigree, of which 271 were genotyped using the OvineSNP50 BeadChip (Illumina, Inc.). The host resistance was assessed using the following traits: fecal nematode egg counts (FEC); FAMACHA score (FAMACHA); and resistance to gastrointestinal nematode infection (RGNI) as a combination of FEC, FAMACHA, body condition score, and hematocrit. The genetic parameters and breeding values were estimated using single- and multi-trait analyses. For RGNI, the heritability estimates ranged from 0.25 using the single-trait genomic model (S-H) to 0.54 using the traditional multi-trait model (M-A). The heritability estimates for FEC ranged from 0.06 to 0.36, using the single-trait pedigree-based model (S-A) and the multi-trait genomic model (M-H), respectively. For FAMACHA, the heritability estimates ranged from 0.46 (M-H) to 0.54 (M-A). Estimates of genetic correlation ranged from 0.22 to 0.69. The inclusion of genomic information provided gain in accuracy for all traits. All estimates of predictive ability obtained using genomic data in a multi-trait setting were higher than those obtained using single-trait models. The estimates of predictive ability ranged from 0.03 (S-A) to 0.46 (M-H). The heritability estimates obtained using genomic information showed that all traits evaluated are suitable for genomic selection. Despite the low accuracies obtained, the use of the genomic model provided more accurate estimates of breeding values in comparison to the pedigree-based model.