| dc.description.abstract | Accurate estimation of genetic parameters is important in animal evaluation to effectively sustain reproductive performance, while accounting for sources of variation in animal performance. The pig improvement program is under implementation; however, accurate estimates of performance parameters are largely lacking. KALRO is at the forefront of pig genetic improvement, breeding research, and dissemination of improved pig technologies. They have implemented breed evaluation programs, particularly for the Large White and Landrace breeds. KALRO also develops feeding regimes, disease control programs, and farmer training modules.
This study aimed to conduct a genetic analysis of the reproductive performance of sows in Western Kenya by: (i) evaluating the non-genetic sources of variation in sow reproductive performance, which include herd, season of birth, year of birth, season of farrowing, and year of farrowing, and (ii) estimating variance components and genetic parameters of sow reproductive performance. Data were collected on reproductive performance records of Large White sows obtained from 2010 to 2022. Data on reproductive traits, namely litter size at birth (LSB), number of piglets born alive (NPBA), litter size at weaning (LSW), and inter-farrowing interval (IFI), were obtained from pig farms in Kisumu and Trans-Nzoia counties. Non-genetic sources of variation in reproductive traits were determined by fitting reproductive data to a fixed-effect model in a generalized linear model using R software. Estimates of variance components and genetic parameters were obtained by fitting a multivariate animal model to the data in WOMBAT. The fixed part of the model comprised herd, year of birth (Yob), season of birth (Sob), year of farrowing (Yof), season of farrowing (Sof), parity, and LSB, among which Yof, Yob, herd, and LSB were found to have a significant effect on reproductive performance. There was low genetic variance compared to phenotypic variance, and high phenotypic correlation estimates were observed between LSB and LSW (0.66±0.12) and between NPBA and LSW (0.82±0.01). Heritability estimates were 0.014±0.040 for LSB, 0.011±0.039 for NPBA, 0.001±0.035 for LSW, and 0.039±0.038 for IFI. These low heritability estimates suggest that genetic improvement through selection will be slow. Analysis of non-genetic sources of variation revealed that herd management and seasonality of years in terms of rainfall were significant sources of variation in sow reproductive traits, underscoring the importance of management and environmental factors in optimizing sow productivity. Additionally, the estimated variance components and genetic parameters showed that heritability estimates for sow reproductive traits were generally low, suggesting limited genetic variability and potential for response to selection. While strong phenotypic correlations were observed among the litter traits, some genetic correlations were antagonistic, indicating that selection for one trait may negatively affect others if not carefully balanced. Therefore, there should be cautious implementation of genetic selection to avoid compromising correlated traits. | en_US |
