Daily feed intake patterns of purebred nucleus boars as genetic indicators for disease resilience of crossbred barrows under a natural polymicrobial disease challenge

Resilience is an important selection target in pig production to reduce the impact of stressors on performance and welfare, in particular disease stressors. Disease resilience is, however, difficult to select for because purebred selection candidates must be raised under high biosecurity. Previous research showed that patterns of feed intake and feeding behavior of pigs under a disease challenge are genetically correlated with disease resilience. Given the wealth of individual feed intake data that is collected on purebred selection candidates, the objective of this study was to determine whether patterns of feed intake derived from such data can be used as genetic indicators to select for disease resilience of crossbred pigs. Daily feed intake on 27,880 boars from 5 Landrace and Large White breeding populations were used to derive three potential disease resilience indicators: the square root of the standard deviation (RSD), the lag-one autocorrelation (AC), and the skewness (SK) or residuals of linear regression of feed intake on age. Heritability estimates were 0.13 for RSD, 0.08 for AC, and 0.06 for SK. Estimates of genetic correlations with growth rate and feed intake of these same purebreds were high positive for RSD, close to zero for AC, and moderate negative for SK. Estimates of genetic correlations of the purebred traits with traits of their crossbred barrows (n = 1,818) that were exposed to a natural polymicrobial disease challenge indicated that resilience measures derived from purebred nucleus data are different genetic traits than similar measures (i.e., RSD) on their crossbreds under disease, as are corresponding performance traits such as growth and feed intake. Estimates of genetic correlations of the three indicator traits of purebreds with resilience traits of crossbreds under the disease challenge, including growth rate, mortality, and veterinary treatment rates, were highly variable and on average close to zero. We conclude that the purebred feed intake pattern traits evaluated here are not ready to be used to select for disease resilience because of inconsistent results and large standard errors of genetic correlation estimates. However, results do suggest that resilience measures derived from feed intake and behavior traits (e.g., based on duration) of purebreds in high-health nucleus herds may contain information that is genetically correlated to disease resilience in the field. Additional research is needed to identify such measures.