Genetic analysis of behavior traits in swine production

Authors: Rohrer, Gary; Brown-Brandl, Tami; Rempel, Lea; Schneider, James; HOLL, JUSTIN - Pig Improvement Company

Submitted to: Livestock Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/3/2013
Publication Date: 10/1/2013
Citation: Rohrer, G.A., Brown-Brandl, T., Rempel, L.A., Schneider, J.F., Holl, J. 2013. Genetic analysis of behavior traits in swine production. Livestock Science. 157(1):28-37.

Interpretive Summary: Stress in livestock reduces production and is a well-being or welfare concern. Stressors can be production environments, interaction with humans, or other animals in their pen among many other factors. Aggressive animals can cause injuries to other pigs and decrease performance due to the stress they inflict on pen mates. Selection may be one method that could be applied to improve animal well-being as well as reduce animal aggression. However, before selection programs can be designed to produce more docile pigs, we need to determine whether differences in pig behavior in response to stressful situations are inherited. Pig behavior was evaluated in a Landrace-Duroc-Yorkshire composite population. Piglets were evaluated for their response to handling at 1 d of age (n = 11,069), being placed on their back for 60 seconds at approximately 24 days of age (n = 975), and being confined in a scale while backfat measurements were being collected (n = 9,035). Feeding behavior was monitored in a growing-finishing facility (n = 1,162) where approximately 40 pigs were placed in pens containing one 5-position feeder. Feeding information collected included number of meals/day, average length of meals, and preferences for specific position at the feeder. Feeders were placed along a fence with one end adjacent to a gate (gate-end) and the other end open to the pen. An animal model was fit to estimate heritability of each measurement and whether the inheritance of one measurement influences the inheritance of another (genetic correlation). All estimates of heritability were greater than zero. Activity scores and backtest traits had the lowest estimates of heritability (0.15 to 0.19), the heritabilities of measures of feeding behavior were more variable (0.16 to 0.60), while production data had high heritabilities (> 0.5). All traits measured during the backtest had strong genetic correlations and similar estimated heritability. Among feeding behavior traits, number of meals/day and average meal length were genetically correlated with total daily meal time. In addition, animals that typically ate alone avoided the open-end position at the feeder. The only behavioral traits with genetic correlations significantly different from zero with production traits were associated with feeding behavior where animals that ate longer meals and spent more time at the feeder/day tended to be heavier and fatter at 154 days of age. In addition, animals that ate more meals/day were fatter and animals that preferred the gate-end position of the feeder were heavier. Pigs with more reactive responses to stress tended to eat fewer meals/day, each longer in duration, and preferred the gate-end feeder position. The measures of pig behavior studied were heritable and because of low genetic correlations between measurements, selection for more docile pigs should not have large detrimental effects on performance.

Technical Abstract: Estimates of genetic parameters related to pig behavior under stressful situations are required before selection programs can be designed to produce more docile pigs. Pig behavior was evaluated in a pedigreed Landrace-Duroc-Yorkshire composite population. Piglets were evaluated for their response to handling at 1 d of age (n = 11,069), being placed on their back for 60 s at ~24 d of age (n = 975), and being confined in a scale while backfat measurements were being collected (n = 9,035). Feeding behavior was monitored in a growing-finishing facility (n = 1,162) including preferences for feeding positions. Feeders were placed along a fence with one end adjacent to a gate (gate-end) and the other end open. An animal model was fitted to the data using WOMBAT where litter was included for d 1 activity scores and backtest traits. Fixed effects of sex, pen/year-season/date of collection in all analyses along with scorer (d 1 activity score) and a covariate of age (d 154 weight and backfat). Multiple trait models were fit to estimate genetic covariances among traits. All estimates of heritability were significantly different than zero. Activity scores and backtest traits had the lowest estimates of heritability (0.15 to 0.19), measures of feeding behavior were more variable (0.16 to 0.60) while production data had high heritabilities (> 0.5). Genomic heritability estimates were similar to standard heritability estimates for most traits, except traits measured at a young age. All traits measured during the backtest had strong genetic correlations and similar estimated heritability. Among feeding behavior traits, number of meals/d and average meal length were highly correlated with total daily meal time. In addition, animals that preferred to eat alone avoided the open-end position at the feeder. The only behavioral traits with genetic correlations significantly different from zero with production traits were associated with feeding behavior where animals that ate longer meals and spent more time at the feeder/d tended to be heavier and fatter at 154 d. In addition, animals that ate more meals/d were fatter and animals that preferred the gate-end position of the feeder were heavier. Pigs with more reactive personalities tended to eat fewer meals/d, each longer in duration, and they preferred the gate-end feeder position. The measures of pig behavior studied were heritable and selection for more docile pigs should not have large detrimental effects on performance.