Evidence of Genetic Variation for Recombination Events in Purebred Swine Populations

Recombination can affect the genetic gain of a trait in different ways. A high recombination rate can cause instability of genomic predictions as a result of the linkage disequilibrium breaking between markers and QTL. Conversely, recombination rate can maintain and increase the ability to recruit genetic variability by virtue of the same process. Within this research, we investigated the potential effects of sex and breed as well as the genetic variation of recombination events in swine. Data originated from four breed/sex commercial nucleus populations of Smithfield Premium Genetics: Large White sires (LWS, n=270), Large White dams (LWD, n=1755), Landrace sires (LRS, n=281) and Landrace dams (LRD, n=1356). Individuals in the analysis were genotyped at 10k, 60k or 80k Illumina SNP chips then all imputed to 80k using the Fimpute software. The software FindhapV4 was used to obtain the total number of recombination events for each individual’s progeny (n=20,712 total progeny records). The R package MCMCglmm was employed to fit a model with the total number of recombination events in the genome as the predicted variable. Animal and contemporary group (herd, year, and season of observed recombination event) were random predictors, while sex and breed were fixed effects. Heritability estimates of recombination were obtained within each breed/sex combination using THRGIBBS1F90.The model included the number of recombination events as a predictor variable and a random sire or dam effect for each population. The sire/dam effects was assumed N(0, G/Aσ_s/d² ) where A and G were a pedigree or genomic relationship matrix, respectively. Two fixed effects were included, a contemporary group and a covariate for age at recombination event. Least squared mean estimates (LSME) of total number of recombination events for sex were 16.25(±0.152) in dams and 12.09(±0.181) in sires. LSME for breed were 14.32(±0.229) in LW and 14.05(±0.231) in LR. Sex and breed were both significant (p< 0.05).Heritabilities of recombination across the whole genome were 0.039(±0.036) for LRS, 0.074(±0.030) for LRD, 0.090(±0.062) for LWS, and 0.107(±0.034) for LWD. Heritabilities, when genomic data was included, were 0.050(±0.036) for LRS, 0.232(±0.028) for LRD, 0.084(±0.045) for LWS, and 0.257(±0.029) in LWD. These results show that recombination is heritable and that both sex and breed are significant contributors, with females and LW having a significantly larger number of recombination events. Further research should focus on environmental factors and the interaction between genetics and environment in determining recombination events.