Genotype by environment interactions for reproductive performance of North American purebred sows between North America and Southeast Asia

Abstract

Importing improved Western pig genetics into Southeast Asia has been a common practice to enhance the reproductive performance of pork production in the region. This study aimed to investigate the presence and magnitude of genotype-by-environment (GxE) interactions for sow reproductive performance of purebred North American genetics between temperate (North America) and tropical climates (Southeast Asia). Reproductive data from North American purebred Landrace (LR) and Large White (LW) sows, raised in Canada and in 2 Southeast Asia nucleus herds were used to estimate genetic parameters and quantify GxE. Data were recorded from 2015 to 2023 for 6 reproductive traits: total number born (TNB), number born alive (NBA), number stillborn (NSB), number mummified, age at first farrowing (AFF), and farrowing interval (FI). On average, TNB and NBA were lower in Southeast Asia than in Canada for both LR (by 13.0%) and LW (by 11.1%). The Canadian data showed higher estimates of heritability and repeatability than the Southeast Asia data for TNB, NBA, and NSB. Estimates of genetic correlations between parities for TNB, NBA, and NSB were not significantly different from 1 in Southeast Asia for both LR and LW, but they were significantly different from 1 for both breeds in Canada. Estimates of genetic correlations between Canada and Southeast Asia for TNB, NBA, and NSB were significantly different from 1 for the LW breed, ranging from 0.54 to 0.66, but were higher for the LR breed, ranging from 0.81 to 0.92, and not significantly different from 1. Estimates of genetic correlations between the 2 Southeast Asian herds; however, also revealed the potential presence of GxE within Southeast Asia, although these estimates were associated with high standard errors and were not significantly different from 1. Estimates of the genetic correlation between regions for FI and AFF were found to differ between breeds, with LR showing negative genetic correlations (-0.10 ± 0.33 and -0.59 ± 0.29 for FI and AFF, respectively), while LW showed positive genetic correlations for these 2 traits (0.73 ± 0.41 and 0.50 ± 0.11, respectively). The higher estimates of genetic correlations for reproductive traits between Canada and Southeast Asia for the LR breed indicate that LR sows may be more robust when exposed to a tropical climate, although there was no difference between the 2 breeds in the drop in average reproductive performance between Canada and Southeast Asia, nor was a seasonal effect on performance within Canada and Southeast Asia more pronounced for LW than LR. Further research is needed to investigate the differences in robustness and adaptability between these 2 breeds.

Keywords: genotype-by-environment interactions; purebred sows; reproductive performance; swine.

Figure 1.

Monthly THI averaged over the years 2015 to 2023 and average estimates of corresponding contemporary group effects in Canada CA) (A, B, C) and 2 herds in Southeast Asia (AS1, AS2) (D, E, F) for the Landrace (LR) and Large White (LW) breeds. Contemporary group effects are for the total number born (A, D), number born alive (B, E), and number stillborn (C, F).

Figure 2.

Heatmaps of the additive relationship matrix between phenotyped animals from Canada (CA), Southeast Asian herd 1 (AS1), and Southeast Asian herd 2 (AS2). A) Landrace breed, B) Large White breed. Captions: *Matrix was ordered by population and increasing date of birth. Dashed lines delineate cohorts AS1, AS2, and CA. Within each population, animals are arrayed from left to right and bottom to top from the oldest to the youngest based on their date of birth.

Figure 3.

Principal component analysis (PCA) of pedigree-based relationships between phenotyped animals from Canada (CA), Southeast Asian herd 1 (AS1), and Southeast Asian herd 2 (AS2). A) Landrace breed, B) Large White breed.

Figure 4.

Estimates (and standard error bars) of genetic and phenotypic correlations among reproduction traits¹ within Canada (CA) and Southeast Asia (AS) and by breed (Landrace, LR; Large White, LW) based on bivariate models. *: P < 0.05, **: P < 0.01. ¹ Total number born (TNB), number born alive (NBA), number of mummified (NM), number of stillborn (NSB), farrowing interval (FI), age at first farrowing (AFF). DNC: did not converge.

Figure 5.

Estimates (and standard error bars) of genetic correlations for reproduction traits between Canada (CA) and each and the combined Southeast Asia herds¹ (AS1, AS2, AS) for A) Landrace (LR) and B) Large White (LW) pigs. ¹ Genetic correlation between a pair of regions: CA-AS1, Canadian herds and Southeast Asian herd 1; CA-AS2, Canadian herds and Southeast Asian herd 2; AS1-AS2, Southeast Asian herd 1 and Southeast Asian herd 2; CA-AS, Canadian herds and 2 Southeast Asian herds. *: significantly different from 1 at P < 0.05. DNC: did not converge.